PDA

View Full Version : Advanced Stats Thread



Pages : [1] 2

PlayerToBeNamedLater
07-22-2013, 11:43 AM
I thought this topic deserves it's own thread.

And this a great article from grantland for the first post (minus the typical and predictable Leafs shot at the end):




The Faker's Guide to Advanced Stats in the NHL
By Sean McIndoe on July 19, 2013 2:30 PM ET

So apparently it’s time for the hockey world to fight about advanced stats again.

The most recent outbreak of hostilities was focused (as these things have a disturbing tendency to be) on the Toronto Maple Leafs. First it was forward Joffrey Lupul tweeting about Corsi. Next up was assistant coach Greg Cronin sharing his views on possession stats. If you missed out on those particular skirmishes, don’t worry. Another one should be along any minute now.

If you’re an old-school type who doesn’t like advanced stats, you’re … well, you’re not reading this, because you instinctively slammed your face into your browser’s “close” button as soon as you saw the headline. And if you’re already an advanced stats proponent, you don’t have time to read this because you’re already busy fighting a civil war over what the various stats should even be called.

So that leaves the fans who haven’t chosen a side yet. And in many cases, those who still find themselves on neutral ground might be there because they don’t actually understand what the whole debate is about in the first place.

If you’re interested in learning more about advanced statistics but aren’t sure where to start, there are plenty of good resources available. Eyes on the Prize is running an ongoing summer school feature that will walk through some of the basic concepts. Broad Street Hockey has a basic glossary of terms, and Behind the Net has a more detailed one. Arctic Ice did a whole “Understanding Advanced Stats” section a few years back that has a ton of good stuff. And those are just a few examples. If you’re willing to roll up your sleeves a bit, you can educate yourself fairly quickly.

But what if you’re not interesting in learning more? What if you just want to seem like you did? What if you’d rather know just enough to fake your way through a conversation about advanced stats, while doing as little actual thinking as possible?

If so, I may be able to help you out. Let’s see if we can make our way through this together, with my handy 12-step program for faking your way through the world of advanced hockey statistics.

Step 1: Don’t be afraid
First things first: Hockey’s “advanced” stats aren’t actually all that advanced, at least as far as the math goes. Are you comfortable with tricky concepts like addition and subtraction? Awesome, you’re halfway there. If you also understand what a percentage is and/or can divide by 60, you’re going to be golden.

And beyond the math, the concepts themselves aren’t all that complicated, either. If you can get past the new terminology, most of them will make intuitive sense to anyone who’s ever played shinny.

Practical tip: You can do this. Stay with me.

Step 2: Have realistic expectations
You may be aware of the sabermetrics movement in baseball, and how it’s slowly but surely been revolutionizing our understanding of the game. Hockey hasn’t really come close to that yet, for two key reasons. One is that baseball’s advanced stats work had a several-decade head start. The other is that baseball lends itself to much cleaner data, and has so much of it to draw from. By contrast, much of the data that advanced hockey stats rely on has been tracked for only a few years. We can go back and calculate Babe Ruth’s career WAR, but we don’t know Wayne Gretzky’s Corsi and never will.

The advanced stats movement in hockey is still in its infancy. It’s making progress, but it hasn’t really revolutionized all that much yet. For the most part, its proponents are well aware of that, and you should be too.

Practical tip: If you’re looking to learn more about a sport that you love, you’re going to find this worthwhile. If you’re setting the success bar at “life-changing breakthroughs,” prepare to be disappointed.

Step 3: Don’t say “moneyball”
Just don’t. Yes, you read the book. (OK, fine, you saw the movie and heard about the book.) Yes, it was about advanced stats. But don’t bring it up. You’re just going to have to trust me on this one.

Practical tip: Saying “moneyball” in a discussion about hockey stats is like saying “totally radical” while working as an undercover narc at a high school.

Step 4: If you remember only one thing, make it this word: "possession"
Hockey is a game of scoring goals and preventing the other team from scoring them. But goals are relatively rare events in hockey, so relying purely on goal-based stats to predict future performance is dicey.

If we can’t rely on goals, though, we need something that’s more common and acts as a reasonably good stand-in. For many of hockey’s advanced stats, that’s puck possession. After all, you can’t score if you don’t have the puck, and you can’t be scored on if you do. (Unless you are Chris Phillips.) And since one team or the other has the puck for the majority of every hockey game, we’ve got far more to work with than if we relied just on goals.

Again, this way of thinking is not especially advanced. You’re basically asking “Which team controls the puck?,” which is something fans have understood was important for decades. Possession was one of the intermission stats in NHLPA Hockey 93, and that game kept like only four stats.

So using possession seems relatively intuitive. That’s the good news. The bad news is that there isn’t really a stat that directly measures puck possession. So we look for a decent proxy. That’s where shots come in, which we’ll get to in Step 5.

Practical tip: During any discussion of a specific player, casually say, “Yeah, but those possession numbers …” and then trail off knowingly.

Step 5: Corsi and Fenwick
Corsi and Fenwick are two of the terms you’ll hear most often as you’re faking your way through the world of advanced stats. They measure essentially the same thing: how many shots are directed at each net at even strength.

Note that that’s “shots directed at the net,” not just plain old shots. That’s because we’re including traditional shots (i.e., goals and saves) as well as shots that miss the net. Corsi also includes blocked shots, while Fenwick doesn’t. (Yes, there’s a reason for that. No, you will never need to actually know what it is.)

You can express Corsi and Fenwick as a plus/minus or a percentage, and apply them to either a team or an individual player. If a team is taking more shots than it's allowing, that means it's (very probably) possessing the puck more than its opponents, which means that (all else being equal) you’d expect it to score more goals than the other team.

This is where you’ve probably started mumbling about shot quality, since Corsi and Fenwick and any variations seem to assume that all shots are created equal. That’s obviously not true, since a breakaway or a goalmouth tap-in are more likely to result in a goal than an unscreened shot from the point. But as it turns out, there’s little evidence that teams or players can get consistently outshot while still getting better chances. And while there have been efforts to track scoring chances as opposed to just shot attempts, the results usually just wind up mapping pretty well to each other. So over the long term, we can largely ignore shot quality — not because it doesn’t exist, but because it tends to even out over time.

Corsi and Fenwick turn out to be great predictors of future success. Learn to love them.

Practical tip: As a side note, Corsi and Fenwick are both named after people. They’re not acronyms. This doesn’t sound important, but it is. If you ever write “CORSI,” someone will immediately point at you and yell “Fraud!” At this point, they’ll quite possibly stab you.

Step 6: A word about takeaways and giveaways
These are relatively new NHL stats that have only become prominent in the last decade or so, but if possession is so crucial, you’d probably assume that takeaways and giveaways are especially meaningful. But as it turns out, having a lot of takeaways is often bad, and having a lot of giveaways is often good.

That sounds crazy, but stay with me. If your team has the puck, it can’t record a takeaway. So players/teams that have great puck possession numbers often do rather poorly in terms of takeaways, and vice versa. The same logic applies in reverse to giveaways — you need to possess the puck before you can give it away.

(This also applies to hits, another stat that fans would assume is a positive but can actually be a symptom of a team/player with lousy underlying possession numbers.)

That doesn’t mean that a hit or a takeaway is a bad thing. Obviously, if the other team already has the puck, then taking it away is a good thing. It’s just that if you spend the whole season racking up big numbers in these sorts of categories, there’s a good chance it’s at least partly because the other team always has the puck. And that’s part of the reason why these particular stats don’t correlate well to wins.

One more thing: Takeaways, giveaways, and hits all fall into the category of “real-time” stats, which are incredibly subjective and unreliable, and prone to huge home/away swings that can be attributed to rink bias. Most people disregard them.

Practical tip: Anytime someone mentions a real-time stat, scoff and ask if they’re referring to road splits only.

Step 7: Death to plus/minus
OK, so you kind of understand basic possession stats. At this point, you may have found yourself saying “Wait, Corsi and Fenwick are basically the same idea as that old-fashioned plus/minus stat that I already know and love, only with shot attempts instead of goals.” And you would be right! You may then have continued, “That must mean advanced stats guys love plus/minus!” And you would be more wrong than you have ever been about anything in your entire life.

Remember when we said that goals were relatively rare events and purely goal-based stats could be unreliable? Plus/minus is a purely goal-based stat. And it’s one that penalizes players with defensive roles and/or bad linemates (more on that in a second).

Plus/minus is basically hockey’s answer to baseball’s RBI: a once-cherished statistic that, upon reflection, is actually more or less useless. Ignore it.

Practical tip: If you mention plus/minus, advanced stats folks will actually hiss at you.

Step 8: Context is king
Possession is the building block. Once you can fake an understanding of that, you’re halfway home. But you can also expect to encounter a whole lot of additional stats that sound much more complicated. They are, but in a way that makes enough sense that they shouldn’t trip you up.

Opponents of advanced stats sometimes try to dismiss the numbers by talking about context — you have to watch the game, man! But the advanced stats crowd is one step ahead of them, and has already developed plenty of stats that at least try to factor context into the equation.

Don’t teams take more shots when they’re trailing and fewer when they’re protecting a lead? Yes, which is why we have Fenwick Close. What if a good player is just stuck on a crappy team? Let’s check his Relative Corsi. What if he’s stuck with bad linemates? Let’s look at Quality of Teammates. Does he always line up against the best players? Quality of Competition. But his coach uses him as more of a defensive type of … Zone Starts! You get the idea.

Practical tip: These stats come up less often, so if you’re a faker you can probably get away without really understanding any of them. Just don’t try to shout “context!” at somebody who’s reached a conclusion you don’t like, because they probably have a stat in their back pocket that’s going to make you look bad.

Step 9: Goaltenders: Save percentage good, wins bad
Wins are a bad way to measure a goalie’s worth in hockey for the same reason that they’re a bad way to measure any individual player’s worth in any sport (i.e., they’re just too team-dependent). Goals-against average is better, but still flawed because a goalie can’t control how many shots he has to face. We’d rather look at save percentage, because it doesn’t punish a great goalie who’s stuck on a team that’s giving up a ton of shots every night.

So save percentage is good. Even-strength save percentage is even better, because it eliminates the quality of special teams (which the goalie can’t really control) from the equation. But if that number isn’t handy, you can usually just go with save percentage anytime the subject of goaltending comes up.

Practical tip: Don’t say “wins.” God help you if say “wins.”

Step 10: Understand the role of random chance
Random chance is a nice way of saying “luck,” and that’s good because luck is a dirty word in sports. If we say a team was lucky in a win, it sounds like we’re saying they didn’t deserve it. And as fans, it’s more satisfying if we can tell ourselves that every result was the right one, and that wins and losses are earned by skill and effort alone.

But the truth is that, like any sport, luck plays a huge role in hockey. Some stats can offer insight into its impact. PDO is a simple but crucial stat that takes a player/team’s on-ice save percentage and shooting percentage and adds them together, with anything significantly above (or below) 1.000 being a red flag of especially good (or bad) luck. Luck isn’t sustainable, so keep an eye on teams and players who’ve had an especially lucky or unlucky stretch. They’re great candidates to bounce back or crash back to earth.

There are other ways to try to measure luck, but they all lead us back to the same place: A lot of what happens in this sport is just plain random.

Practical tip: Not everything needs a narrative. Embrace randomness.

Step 11: When things go bad, your safe word is “sample size”
If you want to create a catchy narrative, you can do it based on a very small amount of data (or, if you’re really good, no data at all). But if you want to be confident that you’re drawing conclusions that are actually useful, you’re going to need to base them on enough data to filter out the flukes and the noise. That’s where sample size comes in.

One game isn’t a big enough sample size. Neither is one playoff series. In many cases, even a full season isn’t enough. This ends up coming up a lot in advanced stat discussions, and can make it tricky to draw firm conclusions.

Practical tip: If you’re trying to fake your way through an advanced stats conversation and feel like it’s getting away from you, just furrow your brow and say, “Well, that’s a good point, but of course you have to be careful of sample-size issues.” Everyone will nod solemnly and look at the ground, at which point you can make a break for it.

Step 12: Be prepared for counterintuitive conclusions
If you’ve made it this far, you probably want to know what exactly these stats guys have been learning about the NHL. And you’re in luck, because they’ll tell you. (Good lord, will they tell you.) But there’s one final caveat — you’re going to hear some things that don’t seem to make any sense.

That is to say, you’re going to be told things that are deeply counterintuitive to what you’ve come to believe during your life as a hockey fan. That’s a good thing — if after all this work advanced stats only confirmed what everyone already knew, there’d be no point — but it can still be pretty jarring.

Here’s an example: Does a good defenseman help his goaltender record a better save percentage? Of course! A great defensive defenseman pounces on rebounds and clears the front of the net and doesn’t let his opponents get great scoring chances. He makes his goaltender better. Anyone who watches hockey knows this.

Except … well, this. The numbers don’t play nice with common sense.

Once you go down the advanced stats rabbit hole, this stuff happens all the time. Sometimes it’s big things and sometimes it’s little things, but it’s not rare and it takes some getting used to.

Practical tip: If, after all of this, you hear somebody say something that seems to go against your hockey fan common sense and (this is the key) they back it up with numbers, which are you going to trust? Can you at least accept that it’s possible, just possible, that the numbers are right?

If you can’t, then you’re not going to be able to fake this after all. This whole thing has been a waste of your time. Sorry about that.

But at least you can probably get a job with the Toronto Maple Leafs.



http://www.grantland.com/blog/the-triangle/post/_/id/69005/the-fakers-guide-to-advanced-stats-in-the-nhl

PlayerToBeNamedLater
07-22-2013, 11:56 AM
I'm of the opinion that Fenwick / Corsi is a part of the overall equation, a big part actually, but not the entire equation. And that's where a lot guys like Mirtle falter in their evaluations. While a lot of the best teams post good-great Corsi's, there are exceptions to the rule every year.

Most good teams post a good Corsi, but not all good teams post good a Corsi. Goaltending, special teams, what you actually do with the possession you get are important factors that needed to be accounted for as well.

UWHabs
07-22-2013, 12:19 PM
I want to believe some of the numbers, I truly would like much better measures of players than we have, but there's still plenty of cases that simply don't line up with reality.

For now, my biggest beef with defensive stats are simply that we've been thrown a lot of them, and have no way to measure them against one another. I still go with Phaneuf as my example, as he is a polarizing figure.

His QoC scores are very good - meaning, it appears that he faces some of the toughest competition in the league. But his Corsi scores are some of the worst on your team. So while he's out there, your team is getting massively outshot. And has one of the worst on-ice sv% on your team. There's other stats that point one way or another too, but I can't seem to balance them all out to figure out where to rank him.

zeke
07-22-2013, 12:43 PM
There's only one sentence in that article that I quibble with:


Corsi and Fenwick turn out to be great predictors of future success. Learn to love them.

This actually isn't true.

Corsi and Fenwick turn out to be SLIGHTLY BETTER PREDICTORS OF FUTURE SUCCESS than most other stats, but they are not "great" predictors of future success, or even "good" ones. Their correlation with future performance is still a rather weak one, even if it is slightly better than most goals-based metrics.

JaysCyYoung
07-22-2013, 12:45 PM
I still go with Phaneuf as my example, as he is a polarizing figure.

His QoC scores are very good - meaning, it appears that he faces some of the toughest competition in the league. But his Corsi scores are some of the worst on your team. So while he's out there, your team is getting massively outshot. And has one of the worst on-ice sv% on your team. There's other stats that point one way or another too, but I can't seem to balance them all out to figure out where to rank him.

Because you fail to take into account he played literally half the season with Korbinian Holzer. He had one of the best metrics in the league when paired with a competent NHL partner in Gunnarsson (and his offensive totals are miles better in that situation too).

Context is incredibly important when evaluating QualComp scores.

zeke
07-22-2013, 12:50 PM
Do we have Splits for advanced stats yet?

PlayerToBeNamedLater
07-22-2013, 01:07 PM
I want to believe some of the numbers, I truly would like much better measures of players than we have, but there's still plenty of cases that simply don't line up with reality.

For now, my biggest beef with defensive stats are simply that we've been thrown a lot of them, and have no way to measure them against one another. I still go with Phaneuf as my example, as he is a polarizing figure.

His QoC scores are very good - meaning, it appears that he faces some of the toughest competition in the league. But his Corsi scores are some of the worst on your team. So while he's out there, your team is getting massively outshot. And has one of the worst on-ice sv% on your team. There's other stats that point one way or another too, but I can't seem to balance them all out to figure out where to rank him.

Not only did he face tough competition, he had a very low quality of teammate (worst on the team) and he only started about 40% of his shifts in the offensive zone. He had everything stacked against him. Tough comp, playing in mosty defensive situations, with a poor defensive partner.

Plus, his rel corsi isn't even that bad. His -7.3 is comparable with guys like Robidas, Bouwmeester, Weber, Suter etc who were used in similar roles as Phaneuf, who also posted negative corsi's.

With corsi, you need to compare players who play similar roles, otherwise, it's useless.

PlayerToBeNamedLater
07-22-2013, 01:09 PM
Do we have Splits for advanced stats yet?

I haven't seen them.

PlayerToBeNamedLater
07-22-2013, 01:39 PM
Also, Phaneuf played in nearly an identical role the year before this one, and his rel. corsi was +3.3. Only major difference between this year and last year was his defensive partner.

Phaneuf's defensive partner for a large portion of the year, Hozler, posted ungodly bad numbers across the board. His corsi was -24. This is where I'd love to have splits. I almost guarantee that Phaneuf was a healthy + corsi player without Holzer.

theREALkoreaboy
07-22-2013, 01:47 PM
I can just imagine all the goofy arguments the zekes of the world are going to spin out of these stats. Can't wait.

PlayerToBeNamedLater
07-22-2013, 01:48 PM
These "goofy arguments" have been happening for years.

theREALkoreaboy
07-22-2013, 01:49 PM
Exactly. And now they're all going to be in one place.

PlayerToBeNamedLater
07-22-2013, 01:51 PM
This board was one of the first to use them. Rest of the hockey world is slowly catching up.

PlayerToBeNamedLater
07-22-2013, 02:26 PM
12/13, Fenwick Close Ranking
(with Standings Rank)

1. LA: 8th
2. CHI: 1st
3. NJ: 22nd
4: BOS: 5th
5. DET: 14th
6. STL: 6th
7. NYR: 12th
8. MTL: 4th
9. SJ: 10th
10. OTT: 12th
11. NYI: 15th
12. VAN: 7th
13. CAR: 26th
14. PHX: 19th
15. PIT: 2nd
16. WPG: 19th
17. FLA: 30th
18. MIN: 15th
19. PHI: 20th
20. CGY: 25th
21. ANA: 3rd
22. WAS: 10th
23. DAL: 21st
24. COL: 29th
25. NAS: 27th
26. CBJ: 15th
27. TB: 28th
28. EDM: 24th
29. TOR: 10th
30. BUF: 21st



Fenwick got 12 of the 16 playoff teams correct. The teams missed were -- TOR, WAS, ANA, MIN. What all these teams have in common is that they have quite a bit of elite talent / gamebreakers (that may not need much possession to score) and good goaltending. Fenwick doesn't pick this up obviously.

Also, 3 of the 4 conference finalists were top 4 in fenwick during the regular season, only exception was the Pens.

zeke
07-22-2013, 02:51 PM
Now check out how well Goal Differential predicted the playoff teams.

PlayerToBeNamedLater
07-22-2013, 02:55 PM
Now check out how well Goal Differential predicted the playoff teams.

It does very well. As does a stat like SV%.

I don't see Fenwick replacing all other stats, I just think it's one of a few key stats (like GD, SV etc) that are good predictors of success.

PlayerToBeNamedLater
07-23-2013, 04:35 PM
Two great sites:

http://stats.hockeyanalysis.com/

http://www.behindthenet.ca/

PlayerToBeNamedLater
08-01-2013, 08:19 AM
Sharks scout: "We're using Corsi for Junior players"


The application of analytics to hockey is a relatively new phenomenon, and one that has been skeptically received by many fans, media personalities and even some NHL organizations. Increasingly, however, it has become clear that some of the most successful franchises in hockey are incorporating quantitative analysis into their decision making processes when it comes to hockey operations.

One of those teams, as we've long suspected, is the San Jose Sharks. Despite a relative lack of postseason success, the Sharks have managed to consistently restock their cupboard on the fly and have sustained a high level of play over the past decade.

On Tuesday night Sharks amateur scout Brian Gross appeared on an Edmonton radio station and discussed how technology and analytics have altered the way he does his job. He particularly singled out "Corsi" a metric which essentially measures shot attempt differential for players at even-strength as a proxy for puck possession, and the impact it has had on the club's player evaluations.

From Fear the Fin:

"We have now the Corsi system we use and for people in the NHL, probably Corsi is used almost exclusively for NHL players. We're using Corsi for junior players..."

In the interview Gross goes on to describe the organizatoin's procedure of hand-counting "Corsi events" from game tape. Gross also says that these numbers played a major role in the club's decision to trade up at the 2013 NHL entry draft in order to select WHL defenceman Mirco Mueller (whose nickname should probably be Mirco Stat Mueller).

As Fear the Fin writer The Neutral points out, the Sharks have a habit of "signing and trading for players like Manny Malhotra, Kyle Wellwood, Ian White, Daniel Winnik, Scott Gomez and Tyler Kennedy to trading away the likes of Douglas Murray, Michal Handzus, Jamie McGinn and Devin Setoguchi," or in other words building their team around "players who can drive shot differential."



http://www.thescore.com/nhl/articles/1243405-sharks-scout-we-re-using-corsi-for-junior-players



Sharks use Corsi to evaluate junior players according to scout Brian Gross


The field of hockey analytics might still be in its infancy compared to its counterparts in other sports but there's no question about the major strides it has taken into the mainstream over the last few years. We're at the point where many teams have full-time stats gurus on their payrolls, GMs and other team personnel are common attendees of analytics conferences, coaches cite shot attempt differential during intermission interviews, and major television personalities bring up analytics in panel discussions.

Most of the work done regarding "advanced" statistics in hockey revolves around a stat that really shouldn't be considered advanced in the slightest. Despite its esoteric name, Corsi is simply the difference between even-strength shot attempts for and against, a measure that when adjusted for how often teams play with the lead tends to be one of the strongest predictors of future success available. Despite that, some teams remain skeptical of the numbers and others simply unaware.

And then there are the San Jose Sharks. Sharks scout Brian Gross let slip in an interview with an Edmonton radio station last night that the organization is so far ahead of the curve that they not only use shot metrics to guide their decisions at the NHL level but have people who manually track Corsi for the major junior leagues to supplement scouting, which is unheard of outside of the Soo Greyhounds' excellent work. You can listen to all three parts of the interview here, but I've transcribed the most relevant bits which include a fascinating and unprecedented look behind the curtain at the Sharks' scouting process as well as their interest in Mirco Mueller and the draft-day trade that allowed them to select him.

***

Guy Flaming: Anything stand out, big significant changes in how you do your job?

Brian Gross: Well there's no doubt about it that technology is playing a big role. We have a fairly consistent approach technically. We use RinkNet but RinkNet is a system that you can design to package for yourself, for your own team. So we have made a real big change in the last two years in how we identify players, how we grade them and we're moving even farther ahead of that. We have now the Corsi system we use and for people in the NHL, probably Corsi is used almost exclusively for NHL players. We're using Corsi for junior players.

GF: You as in the San Jose Sharks?

BG: The San Jose Sharks. Simply because we have access to almost every tape of every game. So once we highlight a player, our guys down in San Jose start doing that procedure right away. And we have four meetings a year and we start breaking things down really early. For example, the boy we drafted, Mirco Mueller, out of Everett. We knew about him a year before the draft. What we did is when we got him into the Western League, we really started to pick him apart and then we built a whole profile on him. And, at the time we drafted, he was one of maybe four guys we were very interested in. And the reason we moved up was because that small little package of guys were starting to go and so we had to move up to get after this guy.

GF: Even though it was only two spots. I mean, you identified the guy you want so you had to go up.

BG: Because we felt the team in front of us, the team we made the switch with, was probably interested in him.

GF: That was Detroit.

BG: That was Detroit because that's Detroit's kind of d-man. Mobile, skilled guy that plays good in all three zones. And they took a good guy behind us so they got what they wanted but that's how thin it can become and you utilize everything in your power to determine those players that you're isolating. So it might sound a little silly but when we do the draft we build it into blocks. So our blocks are, say, maybe first pick to ten. Then it's eleven to fifteen. Then it's fifteen to twenty and then it's twenty-one to twenty-five, twenty-five to thirty. So we build those blocks and we put players in those blocks and we highlight the guys within those blocks that we're really interested in. It's a lot of work and it's a lot of arguing. It's a lot of, you know, finger-pointing and things going on.

GF: You're an old school guy and you've been around a long time--a really long time--but the whole advanced stats stuff is fairly new. Is that something you're learning on the fly or do the San Jose Sharks have kind of a separate scouting staff, if you will, that kind of breaks down all those numbers and then bring those to your scouting meetings?

BG: Well for the people that are all excited about it, I'm excited when I get to see it and I don't have to develop it on a player. Our job is hard enough, just putting in reports of what we see as opposed to developing Corsi for each individual player that we're interested in. We have some very intelligent young guys in our organization in San Jose that can snap a tape off any system in a matter of minutes and they can isolate a player and they can define all the Corsi information they need on any player, whether it be a pro or a junior.

GF: Now do they do all of the--I don't even know all of the big words and big terms but the zone starts and the quality of competition stuff, all those things are available?

BG: We would evaluate our own players, number one, and number two, we would definitely evaluate players we are interested in on a trade or [who] may become unrestricted free agents.

GF: So if people want to send in some questions they can text us or tweet them in at me. SkinnyFish says, "That's all great about Corsi and stuff but how many Cups has that won the San Jose Sharks? I rest my case." So not a big believer in all the advanced stats stuff but what do you say to that?

BG: You know what, we have 23 cups in our dressing room and each player has one.

GF: And you said it's fairly new for San Jose. So they haven't won a Cup yet.

BG: Yes, this is fairly new. The whole process is fairly new. And exciting.

***

There's a lot to digest there, including Brian's artful response to the tongue-in-cheek comment about the Sharks' championship record but it's intriguing stuff on many levels. One because it's the first instance I can recall of a team official discussing these things so openly but mainly because the Sharks seem to believe they've identified a market inefficiency here in the way teams scout junior hockey players to the extent that they've gone to what sounds like the painstaking measure of hand-tracking the Corsi numbers of prospects they're considering. Reading between the lines, it sounds like Mueller was a guy who rated highly by these measures which makes sense given descriptions of his game.

Less surprising is the confirmation of what we've generally known for a while: the Sharks use analytics, specifically Corsi, to evaluate NHL players they're looking to acquire. Doug Wilson came close to saying as much in an interview with Jason Plank on this very website a few years ago and a good deal of the moves the Sharks have made, both in terms of acquiring and dumping players, have generally carried the Corsi seal of approval. From signing and trading for players like Manny Malhotra, Kyle Wellwood, Ian White, Daniel Winnik, Scott Gomez and Tyler Kennedy to trading away the likes of Douglas Murray, Michal Handzus, Jamie McGinn and Devin Setoguchi, there's always been a sense that the organization values players who can drive shot differential; the types of players who are generally undervalued by the market.

Granted, they've also made their share of moves at the margins that aren't necessarily compatible with analytics (signing Handzus in the first place, picking up Adam Burish, trading for Kent Huskins and Niclas Wallin) but Corsi alone isn't the be-all, end-all and the Sharks are understandably not treating it as such. But it's a valuable tool and it should be encouraging to Sharks fans that it's not only part of the team's toolkit but that they're apparently using it on a larger scale than any other club we're aware of.



http://www.fearthefin.com/2013/7/31/4575152/sharks-use-corsi-to-evaluate-junior-players-brian-gross-mirco-mueller-analytics

zeke
08-01-2013, 12:05 PM
were they using advanced stats back in 2003? because that's the last time they had a good draft.

TheMightyIgor
08-01-2013, 12:20 PM
http://www.youtube.com/watch?v=oHol7WW2A8g

JaysCyYoung
08-01-2013, 01:07 PM
The Sharks might be the worst-drafting team in the NHL...

worm
08-01-2013, 01:46 PM
Sharks scout: "We're using Corsi for Junior players"



http://www.thescore.com/nhl/articles/1243405-sharks-scout-we-re-using-corsi-for-junior-players



Sharks use Corsi to evaluate junior players according to scout Brian Gross



http://www.fearthefin.com/2013/7/31/4575152/sharks-use-corsi-to-evaluate-junior-players-brian-gross-mirco-mueller-analytics

was listening to that interview in the car... was a pretty good discussion

TheMightyIgor
08-01-2013, 02:01 PM
The Sharks might be the worst-drafting team in the NHL...

The Oilers are actually much worse. Having lots of high draft picks and the easiest roster in the NHL to crack obscures that a bit. I'd have Calgary as worse than them also. Or Vancouver. Let's face it. Western Canadians don't know shit about running a hockey team.

worm
08-01-2013, 02:23 PM
Oilers have done alright (just before all the top picks) ... Eberle and Cogliano .. both late first rounders.

Sharks seem horrible lately.... but 2005 they got Setoguchi and Vlasic and Couture in 2007

Flames have been just brutal.... though I love Baertschi.

PlayerToBeNamedLater
08-08-2013, 12:09 PM
Look out for a stat called WOWY. Could eventually replace Corsi as the go to metric. It's being used in some NBA circles as well.


'With or Without You' is a style of analysis used to determine which players benefit the most from their linemates and which players are driving play.



This guy argues for it and against corsi:






Accounting for Teammates Part 1: Corsi Rel and WOWY


The quality of teammates influences almost every stat in all major sports. This is particularly true of the base stats we tend to use, such as on-ice Corsi or Fenwick rate, because they don't just take something a player has done (score a goal) but also include what his teammates did while he was on the ice. The reason we opt for on-ice stats instead of individual is simple - on-ice stats allow us to measure, albeit noisily, all the contributions a player makes to the thing you are measuring.

How often do you hear an announcer say "that kind of play doesn't show up on the stat sheet, but was very important"? If you're measuring on-ice stats instead of individual stats, and you have a large enough sample, those small plays will show up. The trick is accounting for teammate quality, or at the very least taking it into consideration.


Challenges

The first thing one should recognize is that it's often difficult or even impossible to separate out guys that spend the majority of their time together. The Sedins are an obvious example, but it is very common for a top defense pairing to stick together most of a season and somewhat common for pairs or even entire lines of forwards to play the vast majority of their time together.

It all comes down to sample size. With apologies to Nashville fans in mourning, consider Shea Weber and Ryan Suter last season. Going by hockey analysis, at 5-on-5 they spent 1,293 minutes together, Suter played just under 213 minutes without Weber and Weber played almost 195 minutes without Suter. So they each played roughly 3 games' worth of time apart. With a sample size that small, we can't say much of anything.

Any team can look good or terrible over a three-game stretch. Last year Weber did substantially better when they were apart, but to figure out if one was driving play more than the other you would need several seasons' worth of data and even then it's just going to be a rough idea because those two spent so much of their time together.


A First Attempt - Relative Corsi

The most well known and, unfortunately, most widely used method of dealing with teammates is relative Corsi (Corsi Rel). To calculate relative Corsi, you simply take the possession rate when the player is on the ice and subtract off the possession rate when he is off the ice. The idea, broadly speaking, is that if a player is helping his team then, they will be better off with him on the ice than off. That's a fine idea but I have to say I'm not a fan of this metric at all. I'd love for someone to defend it in the comments because it puzzles me that it is ubiquitous given its major flaws. My intuition is that zone-start adjusted Corsi and maybe even just raw Corsi are better metrics than Corsi Rel.

Fellow Driving Play and NHLnumbers blogger Brent Morris provided a nice article of criticisms. In fairness, one could argue that a lot of the problems are down to misuse. Corsi Rel doesn't really adjust for teammates so much as say how your line/pairing does compare to the others on your team. The adjustment, subtracting off the team's Corsi when you are off the ice, puts most of the weight on the guys the player in question never plays with. If you are a first-pairing defensemen, it's mostly going to be determined by how well the second and third pairings go.

What we'd like to do is see how much credit we should give a player versus the teammates he plays with, and Corsi Rel does that very poorly.


A Step Up - WOWY

WOWY, an acronym of "with or without you", is a pretty substantial improvement. In the basic version, one compares how each teammate does with a player on the ice to his results without that player. The idea is the same - if you are good, you will make your teammates better. If a guy is improving his line/pairing then the other player(s) will be better with him than without him.

WOWY adjusts for teammates (or teammate at least) by holding them constant. If you look at Crosby's performance with and without Pascal Dupuis, Crosby will appear on each side of that. This is a big improvement on Corsi Rel because if Dupuis spends a lot of time with Crosby, Sid would mostly be in the Corsi On part which would elevate Dupuis's numbers.

One big issue with WOWY is that you can still get a dragged-along-by-teammate effect if you aren't careful. If you look at a Vancouver defenseman's Corsi with and without Alex Burrows he will be drastically better with Burrows on the ice. Burrows may be a good possession player but as pointed out above it's tough to say since he spent so much time with the Sedins; so it seems likely that a healthy share of the credit there should go to Henrik and Daniel. For this reason it is generally best to run WOWY numbers against teammates on the same color line - forwards with forwards and defensemen with defensemen. If you run it on a defenseman for a forward or vice-versa the results will be very similar to Corsi Rel and carry all those problems.

It's beyond the scope of this article, but you can do more fancy versions of WOWY by looking at combinations of players. For example, you could look at Pittsburgh's Corsi rate with and without Dupuis when both Crosby and Letang were on the ice. This allows you to take into account the quality of both forward and defensemen. The downside is that you are going to shrink the sample sizes pretty dramatically so you might have to go to multiple seasons, which can cause its own problems.


Final Thoughts

Something to keep in mind with both of these is that you are always doing a comparison - in the case of Corsi Rel you are comparing a player, and his line, to the performance of his teammates when he is off the ice. If you have two equally skilled fourth liners, the one playing for Detroit or St. Louis will have a much worse Corsi Rel than the one playing in Minnesota or Nashville because the players in the comparison group are much better.

WOWY is the same way - Crosby's results without Dupuis depend on who tends to slot in when Dupuis is injured or shifted to another line. Keep in mind that you are always comparing a player's performance to some specific group of teammates. This is important if his team is very strong or weak at the same position or if you are using one of these to compare players on different teams.

In the next installment, I will look at a couple methods that get around this by assessing everyone at once - regression and Vic Ferrari's King Value.



http://nhlnumbers.com/2012/7/18/accounting-for-teammates-part-1-corsi-rel-and-wowy

zeke
08-08-2013, 12:14 PM
One thing I've noticed in looking at Corsi is that it seems to ring much truer for dmen than it does forwards.

and it would make sense that adding a third linemate and an extra line makes it many times harder to isolate an individual's performance.

PlayerToBeNamedLater
08-08-2013, 12:22 PM
Forwards are tougher because one great forward (or bad forward) can drive his linemates numbers either up or down.

With Corsi -- it's probably best applied comparing line vs line, d pair vs d pair as opposed to player vs player.

If you compared Phaneuf's numbers with Hozler and without Holzer, the numbers would be pretty extreme, I'd say.

PlayerToBeNamedLater
08-08-2013, 12:28 PM
hockeyanalysis.com did a pretty interesting and detailed look at J. McClement's true impact defensively using WOWY...results weren't positive.

If anyone is interested:

http://hockeyanalysis.com/2013/03/15/visual-wowy-representation-of-jay-mcclements-poor-possession-play/

LeafGm
08-08-2013, 12:34 PM
If an advanced stat comes back and says negative things about Jay McClement's defensive impact, I think that says negative things about the validity of the statistic rather than negative things about McClement's defensive impact.

JaysCyYoung
08-08-2013, 12:35 PM
Pretty much how I feel.

He was hands down one of the most valuable defensive players in hockey last season.

zeke
08-08-2013, 12:39 PM
I think McClement was highly overrated, like so many 4th liners in this market always are.

McClement was a decent 4th line C and good PKer, nothing more. (and remember, these advanced stats are all about 5 on 5 play, not special teams).

In fact, his surprising 3rd-line quality offensive contributions impressed me more than his defense, IMO.

leafman101
08-08-2013, 12:52 PM
They just keep making the same mistake and confusing possession with defense. Possession is possession. It might be a factor in defensive effectiveness, but it does not encompass all that is defense. If it were a venn diagram possession and defense might overlap, but they'd still be separate circles.

Of course McClement's possession numbers stink. He has no talent.

But given the fact that he doesn't have the puck very much, his efficiency and defensive production stats are great. The guy just flat out plays hockey the right way. its pretty hard not to appreciate it.

PKForce81
08-08-2013, 12:57 PM
I think McClement was highly overrated, like so many 4th liners in this market always are.

McClement was a decent 4th line C and good PKer, nothing more. (and remember, these advanced stats are all about 5 on 5 play, not special teams).

In fact, his surprising 3rd-line quality offensive contributions impressed me more than his defense, IMO.


Stop drinking so much during Leaf games, zeke. ;)

I disagree. Defensively Leafs would give up many more goals if we didn't have McClement in the lineup. His positioning and work rate isn't matched even by a guy like Kuly(although it's close). As for his possession numbers...Not sure what they are but he's actually quite good at that part as well.He just wasn't used much(in that sense) and if you look at who he had to play with most times(only so much 1 guy can do) he was good enough.

leafman101
08-08-2013, 01:09 PM
McClement started 27% of his shifts in the Leafs zone. He only finished 41% of his shifts in the offensive zone. His possession numbers are near the bottom of the team. What all that says is he spent a shit load of time in his own end.

But his GAon/60, and onice save%, were the best of all Leafs regular forwards and he had the 3rd highest Corsi Rel QOC on the team. So he played tough minutes, spent a shit load of time in his own end, but flat out less goals were scored when he was on the ice than any other regularly playing forward on the team.

Nothing about that says bad defense.

PlayerToBeNamedLater
08-08-2013, 01:10 PM
The whole point of hockey is to outscore your opponent. And you can't do that without the puck.

So, if you have a player, for whatever reasons, is always playing without the puck, he has a negative impact on your ability to win.

I don't like to isolate offense and defense -- because they are directly linked.

McClement either has a positive, negative or neutral impact 5 on 5.

PlayerToBeNamedLater
08-08-2013, 01:15 PM
McClement started 27% of his shifts in the Leafs zone. He only finished 41% of his shifts in the offensive zone. His possession numbers are near the bottom of the team. What all that says is he spent a shit load of time in his own end.

But his GAon/60, and onice save%, were the best of all Leafs regular forwards and he had the 3rd highest Corsi Rel QOC on the team. So he played tough minutes, spent a shit load of time in his own end, but flat out less goals were scored when he was on the ice than any other regularly playing forward on the team.

Nothing about that says bad defense.

And the Leafs allowed the 2nd most shots on goal when he was on the ice.

Good thing they had a good goalie behind him to stop all those shots.

PlayerToBeNamedLater
08-08-2013, 01:24 PM
The Leafs got killed 5 on 5 when they used their bottom 6 forwards lines. Their top 6 weren't too bad from a possession standpoint. But the bottom 6ers played way too much without the puck, and that has a lot to do with the fact guys like Orr, McClaren, Hamilton, McClement, etc just aren't talented enough even in comparison to other teams bottom line players.

Additions of Clarkson, Bolland should help, I think.

leafman101
08-08-2013, 03:30 PM
They didn't get killed though when McClement was on the ice. They gave up less goals than when anyone else was one the ice.

These need to generalize everything in hockey down to one stat is beyond nuts, and obviously doesn't work. An honest belief that everything in hockey can be narrowed down to corsi just flat out doesn't comport with the game of hockey IMO.

leafman101
08-08-2013, 03:31 PM
And the Leafs allowed the 2nd most shots on goal when he was on the ice.

Good thing they had a good goalie behind him to stop all those shots.

They had a good goalie to play behind everyone and stop all the shots. Still gave up less goals when McClement was on the ice than anyone else.

But for some moronic reason that had absolutely nothing to do with McClement right?

PlayerToBeNamedLater
08-08-2013, 03:43 PM
They had a good goalie to play behind everyone and stop all the shots. Still gave up less goals when McClement was on the ice than anyone else.

But for some moronic reason that had absolutely nothing to do with McClement right?


No, I don't think McClement gets credit for the goalie making more saves when he's on the ice.

leafman101
08-08-2013, 03:43 PM
The reality is, though I guess some will never admit, that nothing in sports can be reduced down to one measurement. And its a pretty silly "advanced stats" movement by supposed "stats" people if that is what it breaks down to. Corsi is one of several things that can be measured in a hockey game. it doesn't measure defense. It doesn't measure offense. Technically it doesn't even really measure possession.

This obsession to focus on one set of data is really the opposite of advanced statistics. Its more of a regression.

The more data the better. And focusing on one unit of measurement gets you less of a complete picture, not more of one.

leafman101
08-08-2013, 03:46 PM
No, I don't think McClement gets credit for the goalie making more saves when he's on the ice.

But why does the goalie make more saves when McClement is on the ice than anyone else? Especially if McClement sucks defensively so much. Luck? Magic?

Or maybe McClement is really just good defensively and makes the goalies job easier than Tyler Bozak for example.

Nah its probably magic.

PlayerToBeNamedLater
08-08-2013, 03:52 PM
Randy used him well though. McClement isn't a defensive stopper and he wasn't used like a defensive stopper.

He was 11th in ES minutes -- used like a 4th liner. On the PK, where he's useful, he led the team in icetime.

He's a 4th liner who plays a key role on the PK. Not a Selke type player. Not an elite defensive player. A depth player.

PlayerToBeNamedLater
08-08-2013, 03:57 PM
But why does the goalie make more saves when McClement is on the ice than anyone else? Especially if McClement sucks defensively so much. Luck? Magic?

Or maybe McClement is really just good defensively and makes the goalies job easier than Tyler Bozak for example.

Nah its probably magic.

Somehow, shots against are really high when the great defensive McClement is on the ice.

Now that's magic.

JaysCyYoung
08-08-2013, 03:57 PM
McClement is definitely a Selke calibre player though. He finished just outside the top five while playing in probably the most critical market in the NHL in terms of identifying and respecting defensive performance from its team's players.

PlayerToBeNamedLater
08-08-2013, 04:01 PM
McClement is definitely a Selke calibre player though. He finished just outside the top five while playing in probably the most critical market in the NHL in terms of identifying and respecting defensive performance from its team's players.



He got 4th line icetime at ES. So, Randy doesn't think so.

Neither does Nonis who brought in Bolland to play on the 3rd line.

Giving McClement the Selke would be like giving a bottom pair dman the Norris.

PlayerToBeNamedLater
08-08-2013, 04:03 PM
I have nothing against McClement. He's not paid a lot and he's a useful 4th liner, who plays well on the PK.

But this idea that he's a Selke level player is insane imo. He just doesn't impact the game on that level. He doesn't play enough to impact the game on that level.

JaysCyYoung
08-08-2013, 04:09 PM
He got 4th line icetime at ES. So, Randy doesn't think so.

Neither does Nonis who brought in Bolland to play on the 3rd line.

Giving McClement the Selke would be like giving a bottom pair dman the Norris.

Which is why Toews won. He logged quality defensive minutes while playing a tremendous amount of time on the ice.

But McClement is still an elite defensive forward.

zeke
08-08-2013, 04:10 PM
yep, selke talk is insane.

McClement is solid in his role - good 4th liner, borderline 3rd liner, good PKer - and that's it.

If we were forced to use him as the fulltime 3rd line C, then he'd be a weakness, not a strength. Carlyle/Nonis traded a good chunk to get Bolland to take that role instead of McClem, though, so at least they recognize it.

PlayerToBeNamedLater
08-08-2013, 04:14 PM
Which is why Toews won. He logged quality defensive minutes while playing a tremendous amount of time on the ice.

But McClement is still an elite defensive forward.

An elite defensive forward plays a hell of a lot more than 4th line icetime. Unless Randy is a dummy. Who would use an elite defensive forward sparingly at ES?

If McClement was a Selke level talent, he'd be the Leafs 2nd or 3rd line C right now.

PlayerToBeNamedLater
08-08-2013, 04:21 PM
Kulemin is by far, the best defensive player on the team.

leafman101
08-08-2013, 04:52 PM
Somehow, shots against are really high when the great defensive McClement is on the ice.

Now that's magic.

And goals against are really low.

But we all know goals are the least important part of hockey. Corsi!

More importantly, I'll take you changing the topic and avoiding answering the question as an admittance that its not magic or luck, but rather McClement.

leafman101
08-08-2013, 04:56 PM
And for the record, Money Ball was about finding overlooked value in stats and exploiting it. Not reducing evaluation of data down to one subset as the end all and be all of the sport. Somehow that lost its way into the "stats revolution" in hockey.

In baseball, where the statisticians are far more advanced, they realized that such a data set does not exist.

PlayerToBeNamedLater
08-08-2013, 05:02 PM
And goals against are really low.

But we all know goals are the least important part of hockey. Corsi!

More importantly, I'll take you changing the topic and avoiding answering the question as an admittance that its not magic or luck, but rather McClement.

Listen, if you want to give credit to McClement for the goalie stopping the puck, that's fine.

I don't agree.

leafman101
08-08-2013, 05:03 PM
Listen, if you want to give credit to McClement for the goalie stopping the puck, that's fine.

I don't agree.

Then why did the goalie stop the puck more when McClement is on the ice than other players?

And more importantly how can it have absolutely nothing to do with McClement, or the quality of chances faced?

JaysCyYoung
08-08-2013, 05:04 PM
McClement likely forces the opposition to make worse shots because of his checking ability and ability to cut off angles.

leafman101
08-08-2013, 05:06 PM
I agree that McClement is just a 4th liner though. He has absolutely no talent. He has no shot. He can't pass. All he can do with the puck is take it behind the net and tie it up along the boards. Which is pretty bad for a center.

He's a play killer. As noted by his awful corsi.

PlayerToBeNamedLater
08-08-2013, 05:08 PM
12/13, GA per 60, Leafs Leaders:
1. C. Orr: 1.07
2. L. Komarov: 1.92
3. F. McClaren: 1.96
4. M. Frattin: 2.30
5. C. MacArthur: 2.31
6. J. McClement: 2.44

PlayerToBeNamedLater
08-08-2013, 05:14 PM
Then why did the goalie stop the puck more when McClement is on the ice than other players?

And more importantly how can it have absolutely nothing to do with McClement, or the quality of chances faced?

I just find it hard to believe that a player is so good defensively, yet somehow still allows a lot of shots.

It's like team a team that allows a lot of shots, but yet, is somehow good enough to limit quality chances. Good defensive limit the amount of shots, bad defensive teams don't.

leafman101
08-08-2013, 05:15 PM
I don't know. People widely regard Phoenix as a good defensive team, but they don't limit shots at all.

Its a bit of an over simplification. There are far more factors to good defense than shots.

In fact taking more shots is a specific strategy against good defenses.

PlayerToBeNamedLater
08-08-2013, 05:27 PM
And for the record, Money Ball was about finding overlooked value in stats and exploiting it. Not reducing evaluation of data down to one subset as the end all and be all of the sport. Somehow that lost its way into the "stats revolution" in hockey.

In baseball, where the statisticians are far more advanced, they realized that such a data set does not exist.

Nobody has claimed there is one magical stat in which to judge players.

But the more good ones we have, the easier it is to piece together the real value of a player.

leafman101
08-08-2013, 05:35 PM
Yeah exactly. Thats also why putting so much focus on only one, like Corsi, is wrong.

PlayerToBeNamedLater
08-08-2013, 05:35 PM
McClement's big problem is usage. He just isn't used a lot at even strength. Until he plays a top 9 role consistently, you can't rank him with the very best imo.

PlayerToBeNamedLater
08-08-2013, 05:36 PM
Yeah exactly. Thats also why focusing on only one, like Corsi, is wrong.

McClement doesn't just have a corsi problem, he has a regular stats problem and a usage problem.

leafman101
08-08-2013, 05:39 PM
I agree. McClement is a limited player. But he's a pretty useable, even valuable, defensive player. He plays the PK. He allows you to give your 4th line a regular shift. He can move up on the wing if needed.

He's just a really smart, solid, dependable player that teams need. Just not a very talented one.

PlayerToBeNamedLater
08-08-2013, 05:41 PM
I think we're saying the same thing then. I agree, he's a useful 4th liner, with good PK ability. And he's paid in line with what he provides.

I was debating the validity of him being a legit Selke candidate.

leafman101
08-08-2013, 05:45 PM
Yeah, of course not. Bergeron and Toews play a completely different sport than guys like McClement.

PlayerToBeNamedLater
08-08-2013, 05:53 PM
On a different note, I was reading now that at least a couple teams have recognized the undervaluing of Russians, and they are focusing more attention there.

That's one way to exploit the system.

PlayerToBeNamedLater
08-09-2013, 01:42 PM
Interview with David Conte:


Gregor: How much, if any, has advanced stats entered into your scouting reports?



Conte: We talk a lot about it, the Moneyball theory and everything else, and I don’t think that the hockey playing public is a big a sample as all of that. It is totally relevant, clearly it is relevant, and clearly there are some trends. If you’re going to invest your picks and your money into high-level young talent, there should be some substantiation via productivity and height and weights and various factors, but the real factor is watching them play. It comes down to “Do you want them on the team or you don’t want them on the team?”



Later, on that nebulous issue of “character,” Conte said:





“Players that play under pressure, players that elevate their games in the most important situations, players that work the clock and the score and they play for the team, I think those are things that are trademarks of players that have survived and prospered with the Devils.”




http://yoursports-1.blogspot.ca/2013/06/advanced-stats-vs-old-school-scouting.html

JaysCyYoung
08-09-2013, 01:43 PM
New Jersey has been a pretty poor drafting team for a long time, so I'm not sure how much stock I would put in Conte's position.

PlayerToBeNamedLater
08-09-2013, 01:47 PM
I took from that that they still place a lot of emphasis on character and intangibles.

JaysCyYoung
08-09-2013, 01:49 PM
I took from that that they still place a lot of emphasis on character and intangibles.

Oh you're absolutely right.

They wouldn't be wasting first rounders on low-upside guys like Matteau if things were otherwise.

PlayerToBeNamedLater
08-09-2013, 01:51 PM
Washington has done a terrific job drafting. Grabbing high end forwards, dmen, goalies, from all parts of the 1st round.

They place a lot of emphasis on skill, at all the positions.

PlayerToBeNamedLater
08-09-2013, 02:02 PM
Some of the better players they've drafted from 02-09:

A. Ovechkin (1)
A. Backstrom (4)
K. Alzner (5)
S. Eminger (12)
A. Semin (13)
B. Gordon (17)
E. Fehr (18)
S. Varlamov (23)
M. Johansson (24)
J. Carlson (27)
J. Shultz (27)
M. Green (28)
M. Neuvirth (34)
B. Holtby (93)

TheMightyIgor
08-09-2013, 02:12 PM
Washington has done a terrific job drafting. Grabbing high end forwards, dmen, goalies, from all parts of the 1st round.

They place a lot of emphasis on skill, at all the positions.


And how well has that skill translated in the playoffs compared to the Devils who put an emphasis on character?

JaysCyYoung
08-09-2013, 02:18 PM
The Devils haven't won a Stanley Cup in a decade, and it was during the height of the trapping/obstruction era that they implemented, so I fail to see the efficacy of that comparison.

PlayerToBeNamedLater
08-09-2013, 02:30 PM
And how well has that skill translated in the playoffs compared to the Devils who put an emphasis on character?

Washington hasn't a won a cup, but they've been a consistent contender for years now.

That's really all you can ask for -- a consistent chance.

TheMightyIgor
08-09-2013, 02:30 PM
The Devils haven't won a Stanley Cup in a decade, and it was during the height of the trapping/obstruction era that they implemented, so I fail to see the efficacy of that comparison.

They did make a surprising run to the Cup finals a year ago though while the Capitals have been a consistent playoff choker.

corksens
08-09-2013, 02:46 PM
Some of the better players they've drafted from 02-09:

A. Ovechkin (1)
A. Backstrom (4)
K. Alzner (5)
S. Eminger (12)
A. Semin (13)
B. Gordon (17)
E. Fehr (18)
S. Varlamov (23)
M. Johansson (24)
J. Carlson (27)
J. Shultz (27)
M. Green (28)
M. Neuvirth (34)
B. Holtby (93)Very Senators-esque.

No cups to show for it, either.

PlayerToBeNamedLater
08-09-2013, 02:52 PM
Sens aren't close to the Caps, in terms of the draft, at least over that time period.

Sens haven't done a very good job of developing 1st line / 1st pairing talent beyond Karlsson.

PlayerToBeNamedLater
08-14-2013, 03:48 PM
Interesting article...A little old.


NHL teams ready to embrace advanced statistics
By Elliotte Friedman Posted: Wednesday, March 6, 2013 | 09:33 PM


few years ago, I was talking to a young player who was really excited after a meeting with his coach. He'd been told he'd scored high on a stat the team kept: how often they retained possession whenever he was responsible for dumping in and chasing the puck.

So, I asked the coach about it. He paused, then smiled and said, "I have no idea what you're talking about." He just didn't want to say anything about this "private" statistic. The next day, the player said, "Please don't report that... I didn't realize I wasn't supposed to say anything."

So it was a secret -- until now.

I've always believed NHL teams keep much more detailed player evaluation tools than they let on. Last weekend, the Massachusetts Institute of Technology hosted its seventh annual Sloan Sports Analytics Conference in Boston. Nicknamed "Dorkapalooza" by ESPN's Bill Simmons, it has evolved into one the most important non-game events on the sporting calendar.

One of this year's panels was entitled "Revenge of the Nerds," about "the meteoric rise and dominance of probabilistic thinking using objective data in decision-making across all disciplines."

It featured Moneyball author Michael Lewis, Dallas Mavericks owner Mark Cuban, and Nate Silver, the statistical genius behind FiveThirtyEight.com. Silver's latest triumph was an astonishingly accurate prediction of the electoral vote count in the recent U.S. presidential election.

Pretty impressive.


Looking for the 'Ah-ha!' moment


Four years ago, the Vancouver Canucks were the only NHL team to go to the conference.

"As we like to say, 'We don't know what we don't know,'" said Laurence Gilman, Vancouver's vice-president of hockey operations and assistant general manager. "What we're trying to figure out is what we don't know. This conference helps get us thinking in ways in which we weren't thinking."

Meanwhile, the Edmonton Oilers' Kevin Lowe and Craig MacTavish attended for the first time this year.

"It was thought-provoking," MacTavish said.

This year, Boston, Dallas, Edmonton, Tampa Bay, Vancouver and Washington were there. Technically, Anaheim attended too, although Brian Burke spent more time ripping statistical analysis than praising it. I couldn't stop laughing at his quotes ("Statistics are like a lamp post to a drunk: Useful for support but not for illumination"), picturing how much attendees wanted to strangle him.

"It's not only about analytics, but the critical decision-making process," said Don Fishman, the Capitals' assistant general manager and director of legal affairs, attending for the third time. "That's why I love it...You learn what other leagues and teams are going through to make the decisions they need to make."

"There was a panel talking about football decisions, in-game probabilities," he added. "They were going through plays -- if you have the lead and the ball on your own 40-yard-line with with under five minutes to go, the probabilities may be better to retain possession. [Former New York Jets and Kansas City coach] Herm Edwards was saying, 'Twenty years ago, you'd never think about it.'"

"Everyone wants a black-and-white answer: Will a trade work or not? Nate Silver said analytics won't provide that... You can't do that in sports. But they might let you know if there's a 60 per cent probability whether something will work. So, you can determine if there is a better chance the trade will succeed for you."

"There's a ton of data out there," MacTavish said. "How are you going to use or make it work for you? A little edge in today's game goes a long way. We're all looking for the 'Ah-ha!' moment, but no one's found it yet."

"If you're playing blackjack and you've got 15 while the dealer has a nine, you hit on that," said Stars assistant general manager Frank Provenzano, attending for the second time. "You might get a 10, but you still do it to shift the odds in your favour. I'm looking for the hockey equivalent of that."

"Can we uncover something of value? Something someone else doesn't have?"


Not the most statistic-friendly sport

The biggest problem for the NHL is the sport just doesn't have the statistical bent of others.

"We are third, behind baseball and basketball," Fishman said.

So teams are creating their own. Because there is no consensus, they are notoriously secretive. One thing I believe some teams do is remove "second assists" from players and see how many points are left over. But good luck trying to confirm that.

"We're looking to determine what is random, and what is true performance. Other than that, I'm not going to tell you what we do," Fishman laughed.

"This is a competitive business and sports analytics is untamed territory," Gilman says. "When people were out there discovering a new world, they didn't want anyone else to know what they were doing."

"But believe me when I tell you there are percentage results that allow you to coach and manage your team to hedge bets in certain events."

"Do we have a formula to evaluate players?" Provenzano pauses. "We do... And the question is do we have confidence in it to apply it yet? Not really ... We're working to get models that are more predictive at a higher confidence."

"Each variable has an element of error to it. That starts to add up, so it's full of uncertainty. The Holy Grail is to... find a model that brings you more goals and more wins."

More than just stats

One of the biggest issues with analytics is that its most hardcore followers tend to discount things like "heart" or "clutch performance" because they are not quantifiable. That's a bad idea in hockey, which is so highly dependent on one-on-one battles. Meanwhile, Oakland A's general manager Billy Beane, the hero of Moneyball, blamed his team's playoff failures on "luck." That's a cop-out.

"If you rely solely [on analytics], you do it at your own peril," says MacTavish, who played 1,093 NHL games. "Nobody's figured out a fail-safe method, not in baseball, football, hockey or soccer. But you're foolish not to be [looking at them], either."

Fishman and Gilman are lawyers. Provenzano has a masters' degree in business administration. They are trained to think logically and follow a process. But they aren't automatons. They agree the best way to succeed in the NHL is a marriage of statistical analysis and "feel," for lack of a better term. Provenzano's boss is a Hall of Famer who won three Stanley Cups and scored 564 goals. If anyone should understand what makes a hockey player, it's Joe Nieuwendyk.

Does he listen to analytical advice?

"He's open to it," the assistant says. "Like anyone sitting in [the GM's] chair, he will use it if what you're proposing makes sense."

"Can someone skate? Is he smart? These are things you can't capture numerically. But if he can do those things, then you plug his performance into a model and more good things happen than bad, that helps you."

Teams are trying to apply this to everything from what happens during a game to evaluating potential draftees to valuing free agents. Washington, for example, has a decision to make on Mike Ribeiro.

"You look at statistics, you look at age, you look at probabilities," Fishman says.

He does the contract work for the Capitals and it would be a stunner if every team in the NHL wasn't doing some variation of the same thing. The Oilers, for example, meet monthly with a volunteer "analytics advice group." It's led by Dan Haight, COO of Darkhorse Analytics and managing director for the University of Alberta's School of Business.

"It is amazing, the passion these people have for hockey and the Oilers," MacTavish said. "It's definitely led to philosophical conversations about what we should be doing. A bunch of people talking about different ways of doing things leads to innovation. It's good to surround ourselves with people who are rarely thinking like we are."

"I thought our guys would be at the top echelon of presenters [at the Sloan Conference]."

Edmonton runs a contest on the team website -- "Last Fan Standing." Basically, you choose a game on the schedule and answer five questions about it. The questions are pretty detailed:

"Will there be a penalty call for hooking?" "Will both teams combine to score six or more goals?" "Will the final goal be scored in the third period?"

The more correct answers, the more opportunity to enter prize draws. But is there something extra to this contest?

"You're always looking to flush out some brainpower," MacTavish says.



http://www.cbc.ca/sports-content/hockey/opinion/2013/03/nhl-teams-ready-to-embrace-advanced-statistics.html

PlayerToBeNamedLater
09-16-2013, 02:07 PM
Why corsi/possession should not be the goal


Let’s imagine a sport where two factors are equally correlated with winning so that FactorA is 50% correlated with winning and FactorB is 50% correlated with winning. Now for years general managers in this sport only ever knew that FactorA existed and when choosing how to build their team they only ever considered FactorA. Now let’s assume that in this idealist, yet uninformed about FactorB, world every general manager of every team allocated their financial resources perfectly based on their knowledge of Factor A. On top of that, every team is working under the same financial constraints meaning they spend the exact same amount of money.

The result is, in this fictional world, FactorA becomes perfectly evenly distributed across every team. Strangely though, even after accounting for luck, teams have statistically significant differences in winning percentages.

Now, along comes a smart individual who discovers the existence of FactorB and finds out that FactorB correlates 100% with winning percentage (after factoring out luck) and concludes that General Managers were wrong all along and that FactorB is all that matters to winning and FactorA is irrelevant (has to be since it has zero correlation with winning). Upon discovering this he gets hired to become a General Manager of a team and while every other GM was only signing FactorA players he chose to go out and sign solely FactorB players. He made signing FactorB players his goal. Strangely, despite FactorB seemingly showing a 100% correlation with winning, his team didn’t win any more than anyone else.

The reason for this is that FactorA is in fact important. It just doesn’t seem important because everyone knows about FactorA and FactorA is getting evenly spread out across teams. Ignoring FactorA for FactorB is equally wrong as ignoring FactorB for FactorA. Upon learning of the existence of FactorB and its high correlation with winning, the goal of a General Manager is not to optimize his team for FactorB but to recognize that there is undiscovered value in players that have FactorB as a skill while not ignoring other skills that we previously knew existed.

Bringing this back to hockey, lets call FactorA shooting percentage and FactorB shot generation. Teams have typically doled out contracts based on shooting percentage but not based on corsi as shown by Eric T. His conclusion was:

"most teams don’t give out contracts because of Corsi. But a team that does will get more wins out of their budget than a team that follows the conventional path and overvalues finishing talent"

My response is, not if it comes at the expense of ignoring finishing talent. Based on Tom Awad’s work, finishing talent is probably at least 50% of out scoring your opposition (note that shooting percentage is a combination of out finishing and shot quality in Awad’s terminology).

So, if teams have been doling out contracts based on, effectively, shooting percentage then it is perfectly reasonable to assume that shooting percentage talent is more evenly distributed across teams than corsi-talent is. Under these circumstances corsi would be highly correlated with winning percentage because that is where the differences lie between teams. This doesn’t mean that corsi is the main factor in out scoring the opponent though and valuing corsi at the expense of shooting percentage will be a detriment to any General Manager.

Furthermore, if General Managers as a whole started paying primarily for corsi we will start to find that corsi talent becomes more evenly distributed across teams and thus shooting percentage would become much more highly correlated with winning (even after adjusting for luck). Furthermore, paying players based on corsi would potentially lead to players altering their style of play to optimize their corsi statistics to the detriment of the ultimate goal, out scoring the opponent.

It is certainly possible in the current hockey universe in which players are paid more by shooting percentage than corsi that they play a style of game to optimize shooting percentage at the expense of winning so it is not unreasonable to see the flip side occur of corsi because a metric by which general managers dole out contracts.

Ultimately, the goal of any General Manager is to optimize his line up for out scoring the opposition, not out shooting percentage-ing them and not out corsi-ing them. Corsi or possession should never be considered the goal just as shooting percentage or any other identifiable skill shouldn’t be. The goal has been, is, and always will be out score the opposition and it’s the General Managers job to find the right balance of all the identifiable skills, not just those that seemingly correlate with winning.



http://hockeyanalysis.com/

PlayerToBeNamedLater
09-19-2013, 10:39 AM
This one Leafs fan watched every minute of every Leafs game from the first half of last season with a stop watch to track actual possession. And he found that his numbers were nearly identical to the CORSI/FENWICK numbers tracked by sites like behindthenet. His conclusion was that corsi/fenwick are in fact a good proxy for possession.

It was a small sample size though. He tracked just one team in a short season.

The other issue (and where a lot of these guys go wrong) is that he assumes that bad possession teams can't experience success long term. While possession is an important indicator of success, it isn't the only indicator. It's just one of many. It's like saying you can't win without great goaltending. It's easier to win with great goaltending, but it's not a exact predictor of success.

The link:

http://www.pensionplanpuppets.com/2013/9/16/4727746/leafs-attack-time-at-the-halfway-mark

madvillain
09-19-2013, 11:05 AM
This one Leafs fan watched every minute of every Leafs game from the first half of last season with a stop watch to track actual possession. And he found that his numbers were nearly identical to the CORSI/FENWICK numbers tracked by sites like behindthenet. His conclusion was that corsi/fenwick are in fact a good proxy for possession.

It was a small sample size though. He tracked just one team in a short season.

The other issue (and where a lot of these guys go wrong) is that he assumes that bad possession teams can't experience success long term. While possession is an important indicator of success, it isn't the only indicator. It's just one of many. It's like saying you can't win without great goaltending. It's easier to win with great goaltending, but it's not a exact predictor of success.

The link:

http://www.pensionplanpuppets.com/2013/9/16/4727746/leafs-attack-time-at-the-halfway-mark

I think he's just being honest in that if you combine the Leafs' terrible Fenwick/Corsi numbers and the fact it was a shortened season, the most likely explanation was that the Leafs were "lucky" last year to get into the playoffs. I can't stand the guys who are so adamant it's the case, but there's nothing wrong in making that assumption for the time being.

If the Leafs make the playoffs this year with terrible advanced numbers, then we can start making a case for finding the flaws (which definitely exist, just a matter of finding them).

PlayerToBeNamedLater
09-19-2013, 11:10 AM
I think he's just being honest in that if you combine the Leafs' terrible Fenwick/Corsi numbers and the fact it was a shortened season, the most likely explanation was that the Leafs were "lucky" last year to get into the playoffs. I can't stand the guys who are so adamant it's the case, but there's nothing wrong in making that assumption for the time being.

If the Leafs make the playoffs this year with terrible advanced numbers, then we can start making a case for finding the flaws (which definitely exist, just a matter of finding them).

It's not just about flaws though. The Leafs need to improve the possession side of their game, but that doesn't mean they can't have success if they don't.

I'm a big proponent of Fenwick -- but it isn't an exact predictor of success. Teams like Pittsburgh, Anaheim, Washington, Toronto etc all had relatively successful years with average to below average possession numbers. It wasn't like the Leafs were the only below average possession team to find success.

If you're good enough in other areas, you can overcome below average possession stats.

madvillain
09-19-2013, 11:29 AM
It's not just about flaws though. The Leafs need to improve the possession side of their game, but that doesn't mean they can't have success if they don't.

I'm a big proponent of Fenwick -- but it isn't an exact predictor of success. Teams like Pittsburgh, Anaheim, Washington, Toronto etc all had relatively successful years with average to below average possession numbers. It wasn't like the Leafs were the only below average possession team to find success.

If you're good enough in other areas, you can overcome below average possession stats.

Yeah good in other areas, or lucky (see PDO). I'm totally on board with the idea that they're not the be-all-end-all, especially when it comes to the Leafs one factor that gets ignored is despite the ugly advanced stats, they're a young team that will more than likely grow and get better.

And another thing I noticed from the super-advanced stats supporters was that despite David Clarkson's long contract, his advanced stats are pretty good, but they seem to have ignored them (and boast about Grabo while they're at it).

This season will tell us a lot going forward.

PlayerToBeNamedLater
09-19-2013, 11:39 AM
The Leafs must be the luckiest team ever, because they finished 6th in shooting % the year before last. The Leafs finish at the top or near the top in shooting %, because they have a lot of offensive talent.

That is a good point that's been ignored this offseason. The Leafs have added a couple good possession guys in Clarkson and Bolland, and they should have Gardiner/Lupul for most of the year and those 2 are good possession guys as well. It's an area where they should be vastly improved.

Wayward DP
09-19-2013, 12:10 PM
Not to mention addition by subtraction. Not having Holzer, Kostka, and hopefully Fraser in the lineup on a regular basis should help too.

zeke
09-19-2013, 01:17 PM
I don't like to disregard the possession stats, but sometimes you have to step back and see what you're actually predicting.

What the Corsi addicts are actually predicting about the Leafs this year is:

1) The Leafs' offense significantly overachieved last year. The thing is, best on previous years' production, they really didn't. They had a couple of surprises (Kadri, Franson) and some dissappointments (Grabo, Lupul, Gardiner, Liles) but overall scored at about the expected level. Predicting them to drop significantly offensively is probably a bad prediction regardless of shooting percentage, especially considering the fact that key offensive players like Lupul and Gardiner will most likely play significantly more this year.

2) The Leafs' goaltending significantly overachieved last year. The thing is, Reimer didn't do anything other than repeat his performance from his last healthy year in his rookie season. Predicting him to get significantly worse is probably not a great idea, especially with Bernier there as an extra layer of insurance if he does.

The corsi addicts are avoiding actually making the above 2 arguments, though, because they're much harder to make, and are instead hiding behind the possession stats from a shortened season and refusing to make any further analysis - and that's pretty telling.

Moreover, there's a third argument which only recently have they started to address in their predictions, as a bit of a copout preemptive excuse:

3) That the leafs' offseason changes did not improve the team, and actually made them worse. I've seen more predictions recently with the copout "UNLESS they significantly improve their Corsi", so at least they're starting to realize this one glaring weakness of their arguments so far. But we know this is a copout because with all the crying over losing Grabo's Corsi, and criticisms of Bolland's Corsi, we know that they have to be predicting the Leafs should be getting even worse Corsi-wise....but again, they won't actually go so far as to say that, because they know it's much harder to defense. More telling is that you never hear about the positive Corsi of additions like Clarkson and Gardiner....in fact, the Corsi crowd lambasted the Clarkson signing DESPITE his great Corsi, which is the most telling of all. But we can't let them copout on this one, because the necessary conclusion from all their arguments offseason is that these changes made the Leafs worse, not better, even if they refuse to actually come out and say it:

Scrivens ----> Bernier

Kostka ------> Gardiner
Holzer ------> Ranger
(Liles -------> Rielly?)

Frattin -----> Clarkson
Grabo ------> Bolland
MacArthur --> Raymond
Komarov ----> Colborne/Ashton

PlayerToBeNamedLater
09-19-2013, 03:08 PM
Taking a closer look at the fenwick numbers, the Leafs were better from an offensive standpoint than the overall numbers show.

The Leafs ranked 19th in missed shots directed towards the opponents goal and 26th in shots that hit the opponents target. Not great but not the absolute worst. But it's the defensive side of things where they ranked right at the bottom. The Leafs ranked 30th in missed shots directed towards their goal and 29th in shots allowed that hit the target.

It wasn't really about the ability to create offense, they were just allowing too many chances. Holzer was a complete disaster and he posted one of the worst Corsi's in league. Fraser, relative to the comp he faced, was not effective as well. I think the biggest difference will be Gardiner, who posted one of the best Corsi's in the league last year. A more mobile, puck savy blueline will go a long way in correcting possession related issues next offseason.

zeke
09-19-2013, 03:22 PM
if you have time see if you can find it - pensionplanpuppets (I think) wrote up a post criticizing Carlyle's "defensive system", and inadverdently showed clearly that the increase in shots against under Carlyle came entirely from an increase in long distance shots. they didn't even notice what they had done - they didn't even realize that the "negligible" decrease in shot from in close was actually fairly significant, percentage-wise.

PlayerToBeNamedLater
09-19-2013, 03:40 PM
Something is missing here.

12/13: 1st in Shooting % / 6th in Goals For
11/12: 6th in Shooting % / 10th in Goals For

I don't buy this 'lucky' argument because just the year before the Leafs were top 10 in shooting % and goals for as well. I find it very unlikely that the Leafs got 'lucky' over 130 games from an offensive standpoint. It's just too big of a sample size.

I think the increase in point production from 11/12 to 12/13 was based mostly on the massive increase in SV%. That alone accounts for a large portion of the point increase.

How are the Leafs able to score as many goals as they have over the last two seasons, but rank so poorly in Fenwick/Corsi. It's tough to figure out.

hockeylover
09-19-2013, 03:51 PM
Something is missing here.

12/13: 1st in Shooting % / 6th in Goals For
11/12: 6th in Shooting % / 10th in Goals For

I don't buy this 'lucky' argument because just the year before the Leafs were top 10 in shooting % and goals for as well. I find it very unlikely that the Leafs got 'lucky' over 130 games from an offensive standpoint. It's just too big of a sample size.

I think the increase in point production from 11/12 to 12/13 was based mostly on the massive increase in SV%. That alone accounts for a large portion of the point increase.

How are the Leafs able to score as many goals as they have over the last two seasons, but rank so poorly in Fenwick/Corsi. It's tough to figure out.

When pressed on the issue, people who argue high shooting percentage as a reason the Leafs got "lucky" have a hard time pointing to which players specifically they can expect a regression on. Other than Kadri, the other players have performed point wise at the levels they did last year before and for long stretches.

zeke
09-19-2013, 04:14 PM
Leafs just have guys with good career shooting percentages:

Kessel 10.9
Lupul 11.5
Kadri 14.0
Bozak 15.5
VanR 10.7
Clarkson 9.4
Bolland 14.4
Kuly 12.0
Raymond 9.5
McClement 8.8
McLaren 11.1
Orr 7.5
Colborne 8.3


Phaneuf 6.2
Gardiner 7.7
Franson 6.0
Liles 6.9
Gunnarsson 5.2
Ranger 5.2
Fraser 4.1

PlayerToBeNamedLater
09-20-2013, 11:32 AM
The Leafs have 9 forwards who are capable of 20+ goals, none of which would be surprise if they get to that level.

The Leafs will be one of the top scoring teams in the league again, because they have elite talent and good depth.

TheMightyIgor
09-20-2013, 01:27 PM
The Leafs have 9 forwards who are capable of 20+ goals, none of which would be surprise if they get to that level.

The Leafs will be one of the top scoring teams in the league again, because they have elite talent and good depth.

When was the last time a team had 9 forwards with 20+ goals? It's tough because there's only so much PP time to go around. And the Leafs don't typically use a forward on the point of the PP.

worm
09-20-2013, 01:38 PM
When was the last time a team had 9 forwards with 20+ goals? It's tough because there's only so much PP time to go around. And the Leafs don't typically use a forward on the point of the PP.

Sometime in the 80s?

Maybe in that first post lockout year?

But he didnt say ALL would get that...just that none would shock if they did.

zeke
09-20-2013, 01:56 PM
Last 2yr Goal Paces:

1 Kessel 36
2 Lupul 36
3 Clarkson 29
4 Kadri 27
5 JVR 26
6 Bozak 21
7 Bolland 19
8 Raymond 16
9 Kulemin 10 (McClement 12)

MindzEye
09-20-2013, 02:08 PM
When was the last time a team had 9 forwards with 20+ goals? It's tough because there's only so much PP time to go around. And the Leafs don't typically use a forward on the point of the PP.

I'm going to guess that it was the Cup winning Rangers team from the 90's. I remember them having a pile of 20+ goal scorers.

hockeylover
09-20-2013, 02:12 PM
So the only two guys that haven't actually already had 20+ goals in a season before are Bolland (19) and Kadri (18 in a shortened season)?

Not bad.

JaysCyYoung
09-20-2013, 02:21 PM
I'm going to guess that it was the Cup winning Rangers team from the 90's. I remember them having a pile of 20+ goal scorers.

Close but no cigar.

The Rangers from 1993-94 had eight 20 goal-scorers, not nine. ;)

http://www.hockey-reference.com/teams/NYR/1994.html

JaysCyYoung
09-20-2013, 02:22 PM
The 1992-93 Penguins also had eight 20 goal-scorers:

http://www.hockey-reference.com/teams/PIT/1993.html

worm
09-20-2013, 03:30 PM
The 85 Oilers had two 70 goal scorers.

Crazy.

MindzEye
09-20-2013, 04:34 PM
So the only two guys that haven't actually already had 20+ goals in a season before are Bolland (19) and Kadri (18 in a shortened season)?

Not bad.


Yeah, some guys will have up seasons, some guys will have down seasons, but we're do deep that it's really likely that enough guys will have up seasons to make us a very dangerous offence. Top 5 depending on how good our PP is.

PlayerToBeNamedLater
09-21-2013, 08:54 AM
When was the last time a team had 9 forwards with 20+ goals? It's tough because there's only so much PP time to go around. And the Leafs don't typically use a forward on the point of the PP.

It's next to impossible to have 9 goal 20 goal scorers because scoring isn't high enough in the league.

But it's good to gave guys capable of playing top 6 roles on the 3rd line in case there are injuries and they need to fill-in.

The Leafs 3rd line of Raymond-Bolland-Kulemin have all been top 6 players on good teams in the past.

mbow30
09-21-2013, 10:29 AM
77-78 b's had 11 guys hit 20

CH1
09-21-2013, 11:19 AM
77-78 b's had 11 guys hit 20

That year, they lost the final to the Habs who only had 5.

hockeylover
10-03-2013, 01:05 AM
James Mirtle ‏@mirtle 4m
BTW all those Leafs fans saying it's "negative" to write about PDO, regression etc.: It applies to more than one team. Ducks in similar boat

Maroon-Getzlaf-Perry
Silfverberg-MP22-Selanne
Winnik-Koivu-Cogliano
Beleskey-Bonino-Palmieri

Beauch-Fowler
Lovejoy-Fistric
Allen-Vatanen

That lineup aside from Perry/Getzlaf/Selanne kinda blows, IMO. I think it's a huge mistake to compare those lineups at least on paper.

zeke
10-03-2013, 12:21 PM
it's too bad Mirtle blocked me last year.

of course, i would have been blocked again by now anyways.

let him keep digging his grave. it'll be all the more delicious next year when his credibility is shot.

PlayerToBeNamedLater
10-03-2013, 03:41 PM
Mirtle's argument that the Leafs miss the playoffs is based on these 3 things:

*Leafs SV% is unsustainable
*Leafs ES Shooting % is unstainable
*Leafs Fenwick is too poor to compete

There's two big problems with his conclusions. He's basing all this on a small sample size -- a 48 game season. He also didn't into account for the changes the Leafs made -- he's using the numbers last years team posted.

He does defend Phaneuf though. I will give him that.

zeke
10-03-2013, 03:49 PM
actually, key difference in the first two there - their shooting percentage is unsustainable, agreed there. 11.5% doesn't happen over the long run. Arguing that that will come down is fine and sensible.

but his argument for save percentage is not that its unsustainable, because a .917sv% is clearly sustainable. Nothing unsustainable about that in the least. In this case he's simply arguing that the leafs goaltending WON'T sustain that save percentage, not that that kind of save percentage is unsustainable, because it's not. And I think you'll notice that he doesn't actually offer any arguments as to why the save percentage will drop, just that it will. And, of course, he makes no mention of bernier replacing scrivens, either.

He also refuses to discuss that actual personnel on the team - he won't discuss players like Lupul and Gardiner missing most of last year, and he won't discuss the fact that 1/3 of the roster has changed from last year.

PlayerToBeNamedLater
10-03-2013, 03:56 PM
The 11.5% is overall shooting % though. He's not even talking about that because the Leafs overall shooting % was elite (albeit lower) the year before as well.

He's talking specifically about ES shooting % which was really high and it will fall this year. But, if the Leafs increase their shot total or their shooting % increases on the PP, they can make up the difference in the drop.

PlayerToBeNamedLater
10-03-2013, 04:02 PM
It was interesting, last year 9 teams posted overall shooting %'s of 9.90+.

The yeat before only 2 teams posted 9.90+

Lots of lucky teams.

zeke
10-03-2013, 04:06 PM
well, you'll probably always see that kind of discrepancy in the smaller sample. probably more really low percentages, too.

BeLeafer
10-03-2013, 04:08 PM
Shooting percentage could easily be affected by rule changes - e.g., offensive zone faceoffs, smaller goalie pads, etc.. Y-O-Y measurement isn't going to be totally reliable.

Montana
10-03-2013, 04:10 PM
...and we just added someone in Clarkson who shoots a f*ckton.

PlayerToBeNamedLater
10-03-2013, 04:11 PM
Overall Shooting %:
(Listed -- Goals / Shots / %)




12/13:

1 Toronto Maple Leafs 145 1,264 11.47%
2 Pittsburgh Penguins 162 1,438 11.27%
3 Tampa Bay Lightning 147 1,323 11.11%
4 Washington Capitals 146 1,350 10.81%
5 Dallas Stars 128 1,261 10.15%
6 Chicago Blackhawks 149 1,494 9.97%
7 Anaheim Ducks 134 1,347 9.95%
8 Montreal Canadiens 146 1,468 9.95%
9 Calgary Flames 128 1,293 9.90%
10 Edmonton Oilers 123 1,285 9.57%

11/12:

1 Tampa Bay Lightning 232 2,227 10.42%
2 Nashville Predators 232 2,261 10.26%
3 Philadelphia Flyers 260 2,641 9.84%
4 Pittsburgh Penguins 273 2,779 9.82%
5 Boston Bruins 260 2,653 9.80%
6 Toronto Maple Leafs 227 2,322 9.78%
7 New Jersey Devils 216 2,253 9.59%
8 New York Rangers 222 2,338 9.50%
9 Washington Capitals 218 2,295 9.50%
10 Edmonton Oilers 207 2,186 9.47%

10/11:

1 Anaheim Ducks 235 2,334 10.07%
2 Vancouver Canucks 258 2,624 9.83%
3 Philadelphia Flyers 256 2,605 9.83%
4 Dallas Stars 222 2,269 9.78%
5 Calgary Flames 241 2,501 9.64%
6 Chicago Blackhawks 252 2,637 9.56%
7 St. Louis Blues 236 2,475 9.54%
8 Minnesota Wild 203 2,148 9.45%
9 New York Islanders 225 2,408 9.34%
10 Detroit Red Wings 257 2,757 9.32%

BeLeafer
10-03-2013, 04:14 PM
Is there a historical overall tracker (i.e., all teams/each year)? I'd bet there's some variability that's tied to rule changes. For instance, did shots go down as much as scoring after '95?

PlayerToBeNamedLater
10-03-2013, 04:16 PM
I got the numbers from here:

http://www.sportingcharts.com/nhl/stats/team-shooting-percentage/1990/



Goes back to 1990.

BeLeafer
10-03-2013, 04:19 PM
Good source. Yeah, looks like broader historical trends at play. Mirtle would do well to look back over the years. Not sure what he's using to benchmark things. Seems like a case of disbelief that the Leafs could be leading in an offensive category ... the number is not off the charts or anything.

hockeylover
10-04-2013, 08:34 PM
Zeke, you holding out on us?

Saw this on HFboards. Great post.


Originally Posted by zeke
they have basically taken a stat like fenwick which actually has a fairly weak correlation with future wins, and are treating it as if it is a much stronger predictor than it actually is. It's correlation with future wins isn't strong enough to suggest its a a clear "cause" separated from all significant "effects", yet they still treat it as if the numbers have shown that we can actual model predictions with fenwick as the "reality", and factors such as sh% and sv% as just luck or small sample variation factors.

fenwick and corsi are nice stats that tell us something, but the conclusions being drawn from them are far too strong to be justified.

and this year is ESPECIALLY glaring because they are insanely drawing these conclusions from a the tiny half-season sample we had last year. even a full season is a small sample for the advanced stats with a weak correlation to future wins - a half-season sample is even weaker by orders of magnitude.

to give a layman's example, let's look at, say, Bozak.

Bozak Career:

Career: .23gpg, .56ppg
12/13: .26ppg, .61ppg
11/12: .25ppg, .64ppg
10/11: .18gpg, .39ppg
09/10: .22ppg, .73ppg

What we see there is a fairly consistent producer over his career. A guy with a hot rookie year, a bit of a sophomore slump, who then settled in between there the next two years, just a tad above his career average.

We have a solid track record to work off of here, which would give us very good reason to predict him to come in again around .25ppg/.60ppg level this year.

But an "Analytics Guy" is not allowed to draw this fairly simple conclusion, because last year Bozak had a 19.7sh%, an 11.1on-ice sh%, and a 1027 pdo. The "analytics guy" will be forced to claim that Bozak is very likely to suffer significant regression this year from last year's production - despite the fact that that would mean Bozak would "regress" to a level well below his relatively consistent career norms - career norms that were established before last year, even with less "lucky" advanced statistics.

So for an Analytics Guy, Bozak overachieved last year and is due for regression this year - even though Bozak simply continued on scoring at his typical career level last year.

And in fact, when you actually look at all the leafs' career production, weighted for recentness and adjusted for minutes played, it's very hard to argue that many leafs overachieved offensively last year, despite what the advanced stats say. About the only two leafs who played a significant amount of games that could legitimately be said to overachieve were kadri and franson - but even then, those were young guys gettting their first legit offensive opportunities, so their lack of track record is hard to hold against them. And moreover, a number of players, either due to poor performance or injury, clearly underachieved last year to balance those guys out - Lupul, Grabo, gardiner in particular.

What we've ended up with is having so many people with a fairly poor understanding of statistics, who have developed unjustified tunnel vision when it comes to these possession stats - they have literally convinced themselves that last year's half-season sample of possession stats has to be the "base" of their analysis, and are actually completely disregarding contrary facts like the Bozak example I showed above. For them, it doesn't matter that Bozak scored at his usual rate last year - he has to be due regression, because the advanced stats from a half-season sample size say that he was lucky last year.

They're making a big mistake, IMO.

But hey, maybe I'm wrong.

PlayerToBeNamedLater
10-06-2013, 09:43 PM
Using Corsi/Fenwick to predict the standings is like using SV% or GF in isolation to the predict the standings. It's just one piece (stat) of a very complex puzzle.

PlayerToBeNamedLater
10-06-2013, 10:27 PM
Leafs Shooting %:
12/13: 11.4
13/14: 10.5

hockeylover
10-06-2013, 10:33 PM
Leafs Shooting %:
12/13: 11.4%
13/14: 11.4%

Mirtle only ever cites SV% at ES though, which was 10.7 % last year and 8.8% this year.

http://www.extraskater.com/teams/2012/all?sort=shooting_pct

And this site has it at 10.5% this year and 11.5% last year...

PlayerToBeNamedLater
10-06-2013, 10:38 PM
Mirtle only ever cites SV% at ES though, which was 10.7 % last year and 8.8% this year.

And the difference in the drop has been made up by an increased amount of shots, at least so far.

I don't think many people expected the Leafs to maintain their shooting % from last year, I slight drop is to be expected.

PlayerToBeNamedLater
10-06-2013, 10:46 PM
And this site has it at 10.5% this year and 11.5% last year...

Yup, that's right, 10.5% this year. It was 11.47 last year.

hockeylover
10-06-2013, 10:46 PM
And the difference in the drop has been made up by an increased amount of shots, at least so far.

I don't think many people expected the Leafs to maintain their shooting % from last year, I slight drop is to be expected.

Sure, I'd agree, I was just trying to figure out where your numbers were from...

PlayerToBeNamedLater
10-06-2013, 10:48 PM
Sure, I'd agree, I was just trying to figure out where your numbers were from...

My calculation error. It is 10.5 this season.

I adjusted post.

PlayerToBeNamedLater
10-06-2013, 11:05 PM
One thing to look at also is, the Leafs are good at drawing penalties. They generated the 10th most PP chances last year. They were also shorthanded the 13th fewest times last year. Good at drawing PP's and good at not giving up PP's. They spent a lot more time up a man, then they did down man. They've continued this success this season, so far.

The Leafs can offset some of their deficiencies at ES (or expected decline in shooting % at ES) by spending less time at ES and by outproducing their opponent on special teams.

hockeylover
10-06-2013, 11:11 PM
It's still early but a quick look at that site I posted seems to suggest some good returns from some of the new guys in terms of their corsi/fenwick. Bolland specifically. One thing about the advanced stats guys is that they don't seem to look very much at any additions made, they just take last years numbers and seem dismissive about any changes we've made to that club.

Montana
10-06-2013, 11:12 PM
What are Bolland and Raymond's Corsi/Fenwick like historically?

PlayerToBeNamedLater
10-06-2013, 11:15 PM
Bolland's haven't been good, but it's tough to really evaluate him because his strength of comp has been so strong typically.

Raymond's been about average.

zeke
10-10-2013, 12:06 PM
I hate the name "advanced stats" because totalling up shot attempts isn't exactly "advanced" stuff.

here's another new stat, that's not really "advanced", but is a simple and excellent idea.

Much like OPS+ and wRC+ in baseball, which adjusts these stats to league average of that season (i.e. 100 = league average)....may I introduce to you....


.....SV%+


http://www.hockey-reference.com/leagues/NHL_2013_goalies.html#stats::15



an excellent much needed stat given the changes in sv% over the years.

we now know, for example, that Reimer's .924sv% (115sv%+) last year was not quite as impressive as Belfour's .922 in '03 (117sv%+).

Wayward DP
10-10-2013, 12:08 PM
I hate the name "advanced stats" because totalling up shot attempts isn't exactly "advanced" stuff.

here's another new stat, that's not really "advanced", but is a simple and excellent idea.

Much like OPS+ and wRC+ in baseball, which adjusts these stats to league average of that season (i.e. 100 = league average)....may I introduce to you....


.....SV%+


http://www.hockey-reference.com/leagues/NHL_2013_goalies.html#stats::15



an excellent much needed stat given the changes in sv% over the years.

we now know, for example, that Reimer's .924sv% (115sv%+) last year was not quite as impressive as Belfour's .922 in '03 (117sv%+).

At first blush, I really like this stat. Really, really like.

hockeylover
10-10-2013, 12:18 PM
Oh, that's wicked!

JaysCyYoung
10-10-2013, 12:19 PM
I hate the name "advanced stats" because totalling up shot attempts isn't exactly "advanced" stuff.

here's another new stat, that's not really "advanced", but is a simple and excellent idea.

Much like OPS+ and wRC+ in baseball, which adjusts these stats to league average of that season (i.e. 100 = league average)....may I introduce to you....


.....SV%+


http://www.hockey-reference.com/leagues/NHL_2013_goalies.html#stats::15



an excellent much needed stat given the changes in sv% over the years.

we now know, for example, that Reimer's .924sv% (115sv%+) last year was not quite as impressive as Belfour's .922 in '03 (117sv%+).

Fantastic.

ERA+ and OPS+ are probably the two best metrics for evaluating individual performance in baseball. To have a statistic at our disposal that averages out factors to create a league aggregate of performance could prove to be incredibly useful. Not sure if there's ever been a comparable state derived for hockey to be honest.

zeke
10-10-2013, 12:39 PM
I wonder why they don't do something simple like PPG+

MindzEye
10-10-2013, 01:44 PM
JEOMKB

PlayerToBeNamedLater
10-10-2013, 02:21 PM
I wonder why they don't do something simple like PPG+

I did something like this a little while back, just with goals

http://www.forumice.com/showthread.php?54781-Let-me-know-if-you-think-this-stat-is-useful

zeke
10-10-2013, 02:26 PM
http://www.grantland.com/story/_/id/9799967/toronto-maple-leafs-advanced-stats


Collision Corsi
This year's test case for advanced stats in the NHL? It's the old-school Toronto Maple Leafs, so … great. We're sure everyone will be super chill about this.


It's a battle sports fans come to know well over the years. On one side, you have the relatively recent wave of advanced analytics proponents, using new types of stats and theories to challenge the conventional wisdom. On the other side are the old-school thinkers, many of whom have been around the game since before these stats guys were born and question how much insight can be pulled from a spreadsheet.

Hockey was late to the party, but the debate has arrived here, too. There have been some ugly skirmishes, often flaring up when a particular team's on-ice results didn't match what the numbers predicted. Every year seems to bring at least one example. In 2009-10, it was the Colorado Avalanche. In 2010-11, the Dallas Stars.1

Perhaps most memorably, the 2011-12 Minnesota Wild managed to lead the Western Conference for much of the season's first half, despite horrible underlying numbers. The stats guys predicted doom. Wild fans scoffed. Insults were hurled. And in the end, the stats guys were right. The Wild plummeted all the way to 12th, missed the playoffs by 14 points, and firmly established themselves as the cautionary example for anyone who'd dare roll their eyes at hockey's analytics community.

This year, we've moved into new territory, and the stakes suddenly seem higher. In any battle, real or metaphorical, things don't get really ugly until you threaten sacred ground. And there may be no more sacred ground in the hockey world than the Toronto Maple Leafs........

hockeylover
10-10-2013, 02:33 PM
Here's something on which both sides can agree: If you're even vaguely interested in the ongoing debate over the future of hockey metrics, the 2013-14 Toronto Maple Leafs are the test case you've been waiting for. They are the canary in the advanced-stats mine shaft. And that makes them this season's most fascinating team.

Pretty much. If the Leafs can continue to have team success despite the advanced stats, it'd be awfully hard to keep saying it's just luck after 130 games.

MindzEye
10-10-2013, 03:12 PM
I already don't like how this is being framed. If the Leafs are successful this season despite unflattering Corsi numbers, it won't show that 'advanced' stats are useless, simply that the current accepted thoughts in the hockey stats community on it's usefulness is wrong.

I'm strongly, strongly leaning towards the "it's part possession, part efficiency" (sv% & shot%) camp. I agree that there's a lot of people who are misusing possession stats to make claims that they simply can't support on their own.

CH1
10-10-2013, 03:25 PM
I just hope the Corsi number takes off so Jim Corsi can get a promotion from his current gig as goaltending coach in Buffalo.

http://cdn.nhl.com/images/upload/2008/06/Jim_Corsi_110x156.jpg

Metalleaf
10-10-2013, 03:42 PM
http://www.keepcalm-o-matic.co.uk/i/keep-calm-and-corsis-wrong-2.png

hockeylover
10-10-2013, 04:04 PM
I already don't like how this is being framed. If the Leafs are successful this season despite unflattering Corsi numbers, it won't show that 'advanced' stats are useless, simply that the current accepted thoughts in the hockey stats community on it's usefulness is wrong.

I'm strongly, strongly leaning towards the "it's part possession, part efficiency" (sv% & shot%) camp. I agree that there's a lot of people who are misusing possession stats to make claims that they simply can't support on their own.

Agreed... but people can probably stop using the word "luck" every 5 seconds to describe our team.

MindzEye
10-10-2013, 04:08 PM
Agreed... but people can probably stop using the word "luck" every 5 seconds to describe our team.

Yeah, it will deflate that little myth, definitely.

Montana
10-10-2013, 04:16 PM
I already don't like how this is being framed. If the Leafs are successful this season despite unflattering Corsi numbers, it won't show that 'advanced' stats are useless, simply that the current accepted thoughts in the hockey stats community on it's usefulness is wrong.

I'm strongly, strongly leaning towards the "it's part possession, part efficiency" (sv% & shot%) camp. I agree that there's a lot of people who are misusing possession stats to make claims that they simply can't support on their own.

This.

I'm not a fan of the idea it's "all or nothing" with fenwick/Corsi.....sv% doesn't tell us a teams going to be good or not by itself, goals scored doesn't tell us if a team is going to be good or not by itself, and Corsi/Fenwick doesn't tell us whether a team is going to be good or not by itself.

Used in proper context though, and preferably together, they can paint a pretty damn accurate picture of how good a team is, and whether they're as good or bad as their record might indicate.

CH1
10-10-2013, 04:19 PM
Agreed... but people can probably stop using the word "luck" every 5 seconds to describe our team.

If my team wins the Cup, you can use the word luck every 2 seconds.

hockeylover
10-10-2013, 04:26 PM
If my team wins the Cup, you can use the word luck every 2 seconds.

Habsforever already does it every two seconds for us, so I couldn't imagine if we actually won.

CH1
10-10-2013, 04:28 PM
I personally wouldn't want to be within a 50 KM radius of habsforever if the Leafs ever won the Cup.

PlayerToBeNamedLater
10-11-2013, 11:59 AM
This.

I'm not a fan of the idea it's "all or nothing" with fenwick/Corsi.....sv% doesn't tell us a teams going to be good or not by itself, goals scored doesn't tell us if a team is going to be good or not by itself, and Corsi/Fenwick doesn't tell us whether a team is going to be good or not by itself.

Used in proper context though, and preferably together, they can paint a pretty damn accurate picture of how good a team is, and whether they're as good or bad as their record might indicate.

Yup.

There isn't one stat in any sport that can be used as the sole predictor of success.

PlayerToBeNamedLater
10-11-2013, 04:17 PM
By the way, the media has gone overboard with these giveaway stats. I know that they latch onto anything that makes the Leafs look bad, but that's what hockey is, a series of giveaways. The fact you give the puck away a lot, means you have it a lot, which is a good thing. Next to +/-, it's the most useless stat there is.

leafman101
10-15-2013, 09:53 PM
I've been thinking about it and this theory that fenwick/corsi is the basis and predictor for team success has a massive fundamental flaw that puts makes it completely worthless. They track 5 on 5. It flat out ignores special teams, which many who actually know hockey would say is the most important part of the game (besides maybe goaltending which it also completely ignores conveniently).

Didn't Scotty Bowman say you can tell how good a team is by adding its PP and PK rates together?

But these stats baffoons have rendered special teams completely worthless (and goaltending too).

5 on 5 corsi has absolutely no worth in judging or predicting the success of a team.

zeke
10-15-2013, 09:58 PM
well, that's not true entirely, at least in terms of predicting future success.

They don't ignore PP/PK willy-nilly - they concentrate on 5on5 because it is a much more repeatable and consistent number from year to year, while special teams are much more variable and unpredictable.

that being said, you are right that it's obviously foolish to ignore special teams.

leafman101
10-15-2013, 10:02 PM
Special teams AND goaltending. If there was an equation those two would be at least 50% of it. And it would probably be a good equation.

Montana
10-15-2013, 10:03 PM
This isn't anything ground breaking though........we've known from day one that no statistic singlehandedly predicts future success.




....still doesn't change the fact it's a valuable piece of a larger equation that does a damn good job at it.

leafman101
10-15-2013, 10:05 PM
The point was never that the stat doesn't have any value. Its that the people who are using it are doing it wrong. Really really wrong.

PlayerToBeNamedLater
10-15-2013, 10:05 PM
They do account for special teams and goaltending. Leafs high PP % and SV% last year were other numbers they believe are unsustainable.

Montana
10-15-2013, 10:06 PM
The point was never that the stat doesn't have any value. Its that the people who are using it are doing it wrong. Really really wrong.



Yeah, and we've known and been saying that from day one.........

hockeylover
10-15-2013, 10:06 PM
This isn't anything ground breaking though........we've known from day one that no statistic singlehandedly predicts future success.

....still doesn't change the fact it's a valuable piece of a larger equation that does a damn good job at it.

It's weird though that for us, we not only win despite being outpossessed, we actually lose more when out-possessing our opponents.

Can't quite figure that one out.

MindzEye
10-15-2013, 10:08 PM
Shots chart from tonight's game

http://www.extraskater.com/game/2013-10-15-wild-maple-leafs

Montana
10-15-2013, 10:09 PM
It's weird though that for us, we not only win despite being outpossessed, we actually lose more when out-possessing our opponents.

Can't quite figure that one out.


We're an elite goal scoring team, with great goaltending.....that had league worst possession last year.......so that's almost precisely what the numbers would predict would happen. (that we'd win games while losing possession).

IMO the fact we win the corsi/fenwick the odd time and lose the game, is more likely than not just an anomalous outcome that seems particularly weird because we win that battle so little.....so when we win it, and lose the game it really sticks out. But that's purely speculation on my part, based on the fact that the lower the sample size the more likely you can see something "odd" in the numbers.

hockeylover
10-15-2013, 10:15 PM
We're an elite goal scoring team, with great goaltending.....that had league worst possession last year.......so that's almost precisely what the numbers would predict would happen. (that we'd win games while losing possession).

IMO the fact we win the corsi/fenwick the odd time and lose the game, is more likely than not just an anomalous outcome that seems particularly weird because we win that battle so little.....so when we win it, and lose the game it really sticks out. But that's purely speculation on my part, based on the fact that the lower the sample size the more likely you can see something "odd" in the numbers.

Yes. I get this part completely.

I don't get why we lose more often than not when we outpossess our opponent. When I did it last night it was something like 110 point pace when we lose it and 68 point pace when we win it.

Montana
10-15-2013, 10:18 PM
I don't get why we lose more often than not when we outpossess our opponent. When I did it last night it was something like 110 point pace when we lose it and 68 point pace when we win it.


My guess (and I concede that's all it is) is that the sample size of games in which we've won possession and lost the game, is so small that it's likely not worthy of deriving much from, or indicative of anything substantial......but I could be completely wrong on that.

hockeylover
10-15-2013, 10:21 PM
My guess (and I concede that's all it is) is that the sample size of games in which we've won possession and lost the game, is so small that it's likely not worthy of deriving much from, or indicative of anything substantial......but I could be completely wrong on that.

I guess that could be it. I think it's 5-7. It'll be interesting to keep track of this over two seasons. If it doesn't get better though, I call shenanigans.

PlayerToBeNamedLater
10-15-2013, 10:22 PM
Yes. I get this part completely.

I don't get why we lose more often than not when we outpossess our opponent. When I did it last night it was something like 110 point pace when we lose it and 68 point pace when we win it.

It's a pretty small sample size. I think it's only about 10 games.

hockeylover
10-15-2013, 10:24 PM
It's a pretty small sample size. I think it's only about 10 games.

Yeah, 12. Still... under .500? It'll be interesting to see if it continues.

PlayerToBeNamedLater
10-15-2013, 10:25 PM
It's tough right now missing Clarkson, Kulemin, JVR who are 3 of our best possession guys.

mbow30
10-15-2013, 10:49 PM
one thing to remember about 'possession' is that we aren't measuring it by actually... yknow... timing possession.

MindzEye
10-15-2013, 10:55 PM
one thing to remember about 'possession' is that we aren't measuring it by actually... yknow... timing possession.

Yeah, a few of us discussed that about a month ago. Corsi/Fenwick is a proxy for possession, not actual possession.

Montana
10-15-2013, 10:58 PM
one thing to remember about 'possession' is that we aren't measuring it by actually... yknow... timing possession.



I'm not sure that's something that requires remembering, as I highly doubt that's a mistake anyone around here has made in the first place.......

Montana
10-15-2013, 10:59 PM
Yeah, a few of us discussed that about a month ago. Corsi/Fenwick is a proxy for possession, not actual possession.



Next you'll be telling me something crazy like....homeruns aren't "the amount of runs a player scored at home".

PlayerToBeNamedLater
10-15-2013, 11:02 PM
There's been a lot of work done and shots attempts toward goal is a very good proxy for possession. That doesn't mean it's the only stat that matters.

mbow30
10-15-2013, 11:09 PM
I'm not sure that's something that requires remembering, as I highly doubt that's a mistake anyone around here has made in the first place.......

and yet it is very apparent that people are, in fact, doing that.

mbow30
10-15-2013, 11:11 PM
There's been a lot of work done and shots attempts toward goal is a very good proxy for possession. That doesn't mean it's the only stat that matters.

a lot of work by whom?

again, it doesn't take an analytical mind to figure out that teams that take a lot more shots than they allow tend to have possession of the puck more often than their opponents.

where the stat runs into problems is from the opposite end of the spectrum. a team that doesn't take a lot of shots isn't going to post a very strong corsi. but that doesn't mean that they aren't in possession of the puck less than their opponents.

and i don't understand why you need a 'proxy' for a stat that can pretty easily be empirically tracked.

PlayerToBeNamedLater
10-15-2013, 11:17 PM
The team that just hangs on to the puck, but doesn't shoot is a statistical anomaly. If you're going enough to possess a lot of the puck, you're good enough to generate chances.

Montana
10-15-2013, 11:18 PM
and yet it is very apparent that people are, in fact, doing that.


I've yet to see a single person confuse Corsi/Fenwick with being a literal representation of time of possession.

Have never seen it happen once, let alone on this board.

mbow30
10-15-2013, 11:19 PM
15% of nhl teams currently employ somebody to study advanced stats.

that's 4-5 teams.

these advances aren't coming from the teams.

MindzEye
10-15-2013, 11:20 PM
There's been a lot of work done and shots attempts toward goal is a very good proxy for possession. That doesn't mean it's the only stat that matters.

Define "very good" though. How much correlation are we talking here? Because Corsi/Fenwick only has a 55% correlation with winning, yet the same crowd considers that a very good proxy for team quality.

mbow30
10-15-2013, 11:21 PM
I've yet to see a single person confuse Corsi/Fenwick with being a literal epresentation of time of possession.

Have never seen it happen once, let alone on this board.

you're acting pretty irrationally here.

look at this thread.

all people are talking about is the leafs' possession.

we don't know what the leafs' possession is. we just assume corsi/fenwick tells us the leafs' possession. but we don't know that for a fact.

we do know, though, that the leafs coaches care about possession enough to time possession, and have intimdated that the team's possession stats last year didn't exaclty correlate with the team's corsi.

PlayerToBeNamedLater
10-15-2013, 11:21 PM
Define "very good" though. How much correlation are we talking here? Because Corsi/Fenwick only has a 55% correlation with winning, yet the same crowd considers that a very good proxy for team quality.

I'm saying it's a good proxy for possession, not in terms of winning/losing.

PlayerToBeNamedLater
10-15-2013, 11:23 PM
you're acting pretty irrationally here.

look at this thread.

all people are talking about is the leafs' possession.

we don't know what the leafs' possession is. we just assume corsi/fenwick tells us the leafs' possession. but we don't know that for a fact.

we do know, though, that the leafs coaches care about possession enough to time possession, and have intimdated that the team's possession stats last year didn't exaclty correlate with the team's corsi.

A dude from pension puppets, actually tracked the Leafs possession last year with a stop watch (I think I posted the article here) and his numbers came very close to the fenwick numbers.

MindzEye
10-15-2013, 11:24 PM
I'm saying it's a good proxy for possession, not in terms of winning/losing.

I know what you're saying, but follow me for a second.

The same people who say that Corsi is a very good proxy for possession (the people who do the statistical work itself) are the same people who make the Corsi = team quality leap of faith based on a statistical 55% correlation between Corsi and winning.

So how good is Corsi as a possession proxy? Quantitatively? 80%? 90? Better?

PlayerToBeNamedLater
10-15-2013, 11:27 PM
I know what you're saying, but follow me for a second.

The same people who say that Corsi is a very good proxy for possession (the people who do the statistical work itself) are the same people who make the Corsi = team quality leap of faith based on a statistical 55% correlation between Corsi and winning.

So how good is Corsi as a possession proxy? Quantitatively? 80%? 90? Better?

Those people are doing it wrong.

Possession is just one part of a very complex game.

Montana
10-15-2013, 11:28 PM
you're acting pretty irrationally here.

look at this thread.

all people are talking about is the leafs' possession.


Yeah, cause we're all smart enough to know what the f*ck is being referred to when we say it.

When we say "Player X has 30 homeruns" in the baseball thread no one confuses it for someone meaning Player x has scored 30 runs at home.


Literally everyone here knows what we're talking about when we use the term "possesion".....you're fabricating a problem that doesn't exist.



we don't know what the leafs' possession is. we just assume corsi/fenwick tells us the leafs' possession. but we don't know that for a fact.

We don't do that at all......in fact we lambast those that do. It's a proxy, not an definitive measure.

Nor does anyone here confuse it as such...



we do know, though, that the leafs coaches care about possession enough to time possession, and have intimdated that the team's possession stats last year didn't exaclty correlate with the team's corsi.


We also know we don't have that data available to us....hence using what we do have. Corsi/Fenwick.....and we attempt to properly contextualize it as much as possible, so it's not miss used (in the manner that many have miss used it, Mirtle, etc)

Montana
10-15-2013, 11:29 PM
Those people are doing it wrong.

Possession is just one part of a very complex game.


You're damn near going to have to put this in your sig soon, with the amount you have to keep repeating it.

mbow30
10-15-2013, 11:33 PM
Yeah, cause we're all smart enough to know what the f*ck is being referred to when we say it.

When we say "Player X has 30 homeruns" in the baseball thread no one confuses it for someone meaning Player x has scored 30 runs at home.


Literally everyone here knows what we're talking about when we use the term "possesion".....you're fabricating a problem that doesn't exist.




We don't do that at all......in fact we lambast those that do. It's a proxy, not an definitive measure.

Nor does anyone here confuse it as such...



We also know we don't have that data available to us....hence using what we do have. Corsi/Fenwick and attempting to properly contextualize it, so it's no misused ( as many have done, Mirtle, etc)

except 'home run' has a clearly defined meaning.

just like corsi can have, without calling it something that it isn't.

PlayerToBeNamedLater
10-15-2013, 11:33 PM
People just say "possession" because it's easy. Writing "proxy for possession" every single time isn't necessary imo. We all know what we're referring to.

Fenwick isn't a true possession stat, just a proxy. But the creators of the stat wouldn't suggest any different.

mbow30
10-15-2013, 11:36 PM
'corsi doesn't mean possession. we just call it possession'.

why would you do that?

we don't call a home run a home run for the sake of it. we call it a home run because a name was assigned to a hit to describe what the hit is. a hit where the batter scores. which is a lot easier than saying 'he got the hit that allowed him to run around all the bases without anybody getting him out, or anybody else getting a hit to advance him to another base'. just like it's a lot easier to say 'corsi' than 'the differential between shot attempts for and shot attempts against'.

Montana
10-15-2013, 11:36 PM
except 'home run' has a clearly defined meaning.

just like corsi can have, without calling it something that it isn't.

Ummm....everyone knows what "possession" means in this context dude. It's a pretty defined statement in these discussions.

Literally never seen anyone get confused by it before now...

mbow30
10-15-2013, 11:37 PM
People just say "possession" because it's easy. Writing "proxy for possession" every single time isn't necessary imo. We all know what we're referring to.

Fenwick isn't a true possession stat, just a proxy. But the creators of the stat wouldn't suggest any different.

lots of people do that all the time.

'the leafs are a bad puck possession team'. that's what we hear, time and again. and it' sbecause of their corsi.

it would be just as easy to say 'the leafs corsi isn't very good', except a lot more factually correct.

Montana
10-15-2013, 11:37 PM
'corsi doesn't mean possession. we just call it possession'.

why would you do that?


People just say "possession" because it's easy. Writing "proxy for possession" every single time isn't necessary imo. We all know what we're referring to.

.

PlayerToBeNamedLater
10-15-2013, 11:37 PM
This is just semantics.

You can call it whatever you want, but we all know what it means.

leafman101
10-15-2013, 11:41 PM
What a load of BS. When James Mirtle or these other fake stats guys talk about the Leafs possession game he is talking about their possession game.

James Mirtle ‏
Wild the NHL's top possession team after tonight's game - small consolation but they look like team on the rise http://www.extraskater.com/teams/2013/5v5


He's blatantly talking about possession not pucks directed at the net. They are using it wrong. That is the whole issue here.

PlayerToBeNamedLater
10-15-2013, 11:43 PM
But there's a strong correlation between pucks directed at the net and possession. Thus it being a proxy for possession.

I don't know why you guys are upset about this. As the Leafs are showing, you can still win at good clip without being great at this particular stat.

leafman101
10-15-2013, 11:45 PM
We're criticizing the mainstream media for blatantly using advanced stats wrong. Why wouldn't we do that?

You always criticize them for being stupid.

hockeylover
10-15-2013, 11:46 PM
Sorry we don't win the right way for you, James.

PlayerToBeNamedLater
10-15-2013, 11:47 PM
We're criticizing the mainstream media for blatantly using advanced stats wrong. Why wouldn't we do that?

You always criticize them for being stupid.

I'm with you on this.

That doesn't mean I think Fenwick is a useless/overrated stat. I think it's a really good stat.

leafman101
10-15-2013, 11:50 PM
No one said it has no value. it definitely overrated though. It doesn't have anywhere near the weight these so called "advanced stats proponents" think it does. Nor does it tell them what they think it does.

MindzEye
10-15-2013, 11:51 PM
But there's a strong correlation between pucks directed at the net and possession. Thus it being a proxy for possession.

And we return to my question. What exactly is the correlation? The same claim is made about Corsi and winning, yet the correlation is only 55%

PlayerToBeNamedLater
10-15-2013, 11:51 PM
Hell, I argue with Mirtle on twitter all the time about how it's being used.

Montana
10-15-2013, 11:52 PM
No one said it has no value. it definitely overrated though. It doesn't have anywhere near the weight these so called "advanced stats proponents" think it does. Nor does it tell them what they think it does.

...and no one here has argued otherwise.

Hell, the only "argument" is msun incorrectly assuming people here don't know what they're referring to when they use the term "possession".....when they quite clearly do.

Montana
10-15-2013, 11:59 PM
And we return to my question. What exactly is the correlation? The same claim is made about Corsi and winning, yet the correlation is only 55%

What's the correlation between goal ranking and final season standing?

What's the correlation between sv% ranking and final season standing?



None are great predictors on their own, (and people would be just as incorrect to try and use them as such) but all three together trump any of them individually. Adding Corsi/Fenwick improves the overall equation.

It's not a be all/end all variable....but it's a variable that when used right, is a worthwhile one.

PlayerToBeNamedLater
10-16-2013, 12:13 AM
Of the top 16 save percentage teams last year, 12 made the playoffs.
Of the top 16 goal scoring teams last year, 10 made the playoffs.
Of the top 16 fenwick teams last year, 12 made the playoffs.


All these stats are good indicators of a successful team, but none of them on their own can consistently predict success from year to year. But combined, you might have something.

zeke
10-16-2013, 12:37 AM
What's the correlation between goal ranking and final season standing?

What's the correlation between sv% ranking and final season standing?



None are great predictors on their own, (and people would be just as incorrect to try and use them as such) but all three together trump any of them individually. Adding Corsi/Fenwick improves the overall equation.

It's not a be all/end all variable....but it's a variable that when used right, is a worthwhile one.

actually, goal differential is by far and away the best predictor, with a correlation of over 80%.

obviously.

it's just not as repeatable as corsi.

Montana
10-16-2013, 12:50 AM
actually, goal differential is by far and away the best predictor, with a correlation of over 80%.

obviously.

it's just not as repeatable as corsi.


not sure why you'd word it like that, when I made zero mention of goal differential........but to your point, I agree entirely, goal differential has a strong correlation, not unlike run differential in baseball.

zeke
10-16-2013, 12:51 AM
which is probably worth talking about, actually.

because many of the teams that Corsi claims it is able to spot "overachieving" and who then regress, are also teams that seem to be obviously overachieving based on other metrics like goal differential - that you can predict for obvious regression without looking at the possession stats.

In fact, I could point out examples of where goal differential showed an obvious future regression that somehow Fenwick failed to see entirely.


I'm thinking of the 11-12 Florida Panthers.

They surprised everyone by finishing in first place in the southeast with 94pts. It wasn't a tremendous point total, being 6th best in the East and 14th overall, but it was a solid season.

Looking at their Fenwick Close, they ranked 13th overall in the league, and 7th in the East - Fenwick told us that Florida was full marks for their 94pts, and that it was totally sustainable.

Of course, if we took a gander at the much more strongly win correlated Goal Differential (like I remember doing at the time), we saw that they had one of the worst goal differentials in the league at -0.23, which ranked 10th in the east and 21st overall.

The goal differential pretty clearly told us that team was a fluke who couldn't possibly keep winning, even though Fenwick indicated they should be able to.

And the next year, of course, the Panthers went right back to sucking something fierce, as we all expected they would based on the pre-possession stats we had.


So there's one example, at least, of where the posession stats were lying to us, and the much more strongly correlated stat trumped it quite easily in predicting future wins.

At some point I think I'm going to have to line up some sort of comparison between fenwick and goaldif predictions, and see which one actually works better in real life, and how many examples Corsi/fenwick is taking credit for that goal differential would explain even better, and how many times fenwick completely misses a future regression that goal differential makes obvious.

zeke
10-16-2013, 12:52 AM
not sure why you'd word it like that, when I made zero mention of goal differential........but to your point, I agree entirely, goal differential has a strong correlation, not unlike run differential in baseball.

you said "goal ranking".....I thought you were talking aboiut both goals for and against.

Montana
10-16-2013, 12:55 AM
you said "goal ranking".....I thought you were talking aboiut both goals for and against.

nope....just meant goal rankings.



...but I think you touch on an accurate point. A metric that combined Sv%, goal differential, and possession would likely be the single most accurate predictor of success than anything we're currently using/discussing.

edit: although given a second though, while it may work on it's own as a simple tool.....I think I'd prefer sv%/GF/Corsi. They might come out with somewhat similar results, with the latter being a little more accurate.

soco22
10-16-2013, 12:56 AM
Why don't y'all try and come up with a better advanced stat? Seems like this board has the right ppl to do this sort of thing. The effort to just get the fantasy points to an exact measure shows level of detail ppl would put in.

Ps: if this happens, def should be partly named after me for throwing out the idea haha

Montana
10-16-2013, 01:03 AM
Looking at their Fenwick Close, they ranked 13th overall in the league, and 7th in the East - Fenwick told us that Florida was full marks for their 94pts, and that it was totally sustainable.

Fenwick doesn't "tell" us that though.......idiots in the media might have tried to argue that, but everyone here is long in agreement that Fenwick and Fenwick alone doesn't really tell us anything that strongly all by itself.

I feel almost like people are constantly arguing that SV% is better than GAA, when there's no one on the board ever arguing that GAA is better than SV% anyway......in fact the two of us that argue most consistently to not throw the baby out with the bath water, when it comes to these possessions stats, also point to sv% and goals are just as necessary to paint a truly accurate picture.

That Panthers team was 27th in the NHL in goals for.....and middle of the pack in SV%.


Middle of the pack in two categories, and near the bottom of the barrel of another, is a pretty good indication they weren't as good as their record.



That said, I would certainly be intrigued to see the predictive abilities of (Sv% + GF + Corsi/Fenwick) VS Goal Differential all by itself, and see which was more accurate.

Montana
10-16-2013, 01:05 AM
Why don't y'all try and come up with a better advanced stat? Seems like this board has the right ppl to do this sort of thing. The effort to just get the fantasy points to an exact measure shows level of detail ppl would put in.

Ps: if this happens, def should be partly named after me for throwing out the idea haha


We've basically already come up with two.......(Sv%/GF/Corsi) and SV% that normalizes PP opportunities against.


We've just been too lazy to do all the number crunching required and the requisite naming of said brilliant metrics.

PlayerToBeNamedLater
10-16-2013, 09:24 AM
Goal differential isn't a very good projection stat because it's a team stat that isolates one specific year. Teams change from year to year and it doesn't account for that. If Team A has a +20 GD but loses it's best goalie and top C the next year, GD is pretty meaningless now. You need to use player stats to account for player movement from year to year. That's what I don't like about what some of these stats guys have done this year, they're focusing on team models instead of player models, which don't account for team changes.

If you create some sort of formula using goals for, sv%, corsi per player, for every team, I think you can create a pretty nice tool. But I would take some time to do it properly.

MindzEye
10-16-2013, 09:25 AM
We've basically already come up with two.......(Sv%/GF/Corsi) and SV% that normalizes PP opportunities against.


We've just been too lazy to do all the number crunching required and the requisite naming of said brilliant metrics.

Yeah, that shit sounds a lot like work.

BeLeafer
10-16-2013, 10:01 AM
which is probably worth talking about, actually.

because many of the teams that Corsi claims it is able to spot "overachieving" and who then regress, are also teams that seem to be obviously overachieving based on other metrics like goal differential - that you can predict for obvious regression without looking at the possession stats.

In fact, I could point out examples of where goal differential showed an obvious future regression that somehow Fenwick failed to see entirely.


I'm thinking of the 11-12 Florida Panthers.

They surprised everyone by finishing in first place in the southeast with 94pts. It wasn't a tremendous point total, being 6th best in the East and 14th overall, but it was a solid season.

Looking at their Fenwick Close, they ranked 13th overall in the league, and 7th in the East - Fenwick told us that Florida was full marks for their 94pts, and that it was totally sustainable.

Of course, if we took a gander at the much more strongly win correlated Goal Differential (like I remember doing at the time), we saw that they had one of the worst goal differentials in the league at -0.23, which ranked 10th in the east and 21st overall.

The goal differential pretty clearly told us that team was a fluke who couldn't possibly keep winning, even though Fenwick indicated they should be able to.

And the next year, of course, the Panthers went right back to sucking something fierce, as we all expected they would based on the pre-possession stats we had.


So there's one example, at least, of where the posession stats were lying to us, and the much more strongly correlated stat trumped it quite easily in predicting future wins.

At some point I think I'm going to have to line up some sort of comparison between fenwick and goaldif predictions, and see which one actually works better in real life, and how many examples Corsi/fenwick is taking credit for that goal differential would explain even better, and how many times fenwick completely misses a future regression that goal differential makes obvious.

If you ran a proper regression model and looked at colinearity, you'd in all likelihood find a high degree of colinearitty between goal differential and wins or winning percentage. Colinearity is a measure of sameness ... do the two variables measure much the same thing? The Panthers were a truly lucky team to post that record. It should not have happened, but the goals for/against shook out strangely over the course of the season.

I think the reason people have issue with all this possession/shooting percentage stuff is that Mirtle concluded that the Leafs are lucky. He put way too much weight on the indicators ... much more than he currently professes they should have.

leafman101
10-16-2013, 11:01 AM
Here is something i tested out. PP% + PK% + SV% + Fenwick% + SH%. Its crude here are the results:

2012-13 ForumIce Rating (regular season finish/playoff finish)
1. Chicago - 2.59 (1st/Cup Winner)
2. LA - 2.57 (7th/ 3rd round)
3. Ottawa - 2.57 (12th/2nd round)
4. Boston - 2.56 (5th/Cup Finals)
5. San Jose - 2.55 (11th/2nd round)
6. St. Louis - 2.54 (6th/1st round)
7. Pittsburgh - 2.54 (2nd/3rd round)
8. Washington - 2.54 (9th/1st round)
9. Toronto - 2.53 (9th/1st round)
10. Philly - 2.53 (20th/DNQ)
11. Montreal - 2.53 (4th/1st round)
12. Detroit - 2.52 (12th/2nd round)
13. Vancouver - 2.51 (7th/1st round)
14. Anaheim - 2.51 (3rd/1st round)
15. New Jersey - 2.48 (21st/DNQ)
16. NYRangers - 2.48 (12th/2nd round)
17 NYIslanders - 2.48 (15th/1st round)
18. Edmonton - 2.48 (24th/DNQ)
19. Dallas - 2.47 (21st/DNQ)
20. Tampa - 2.46 (28th/DNQ)
21. Minnesota - 2.46 (15th/1st round)
22. Calgary - 2.46 (25th/DNQ)
23. Columbus - 2.45 (15th/DNQ)
24. Phoenix - 2.41/DNQ)
25. Winnipeg - 2.41 (18th/DNQ)
26. Colorado - 2.41 (29th/DNQ)
27. Carolina - 2.41 (25th/DNQ)
28. Florida - 2.38 (30th/DNQ)
29. Buffalo - 2.38 (21st/DNQ)
30. Nashville - 2.37 (27th/DNQ)


It gets 14 of the playoff teams (technically 15 as NY rangers and Islanders are tied for 15th with Jersey). It catches teams that were better in the playoffs than the regular season (LA/Ottawa/San Jose). It also catches teams that overachieved in the regular season (Montreal/Anaheim/Vancouver). Three of the final four teams ended up in the top 4.

Philly is the big outlier. Save% may need to be weighted more to correct that. Their other numbers were good.

Its rough, but at least its a complete picture.

leafman101
10-16-2013, 11:24 AM
One other note on the Flyers, they finished last season 12-8-2 (97 point pace). They might have made a playoff push in an 82 game season.

Montana
10-16-2013, 11:40 AM
Is that an average of their rankings in each category?.....if so, how does it compare if you just average teams rankings in GF, SV% and Fenwick?

leafman101
10-16-2013, 11:47 AM
Its straight up addition of the raw data. Eg Last year the Leafs were 18.7% on the PP, 87.9% on the PK, .917 save %, 44.1 Fenwick% and 10.56 shooting%. .187 + .879 + .917 + .441 + .1056 = 2.53.

UWHabs
10-16-2013, 11:49 AM
Here is something i tested out. PP% + PK% + SV% + Fenwick% + SH%. Its crude here are the results:

2012-13 ForumIce Rating (regular season finish/playoff finish)
1. Chicago - 2.59 (1st/Cup Winner)
2. LA - 2.57 (7th/ 3rd round)
3. Ottawa - 2.57 (12th/2nd round)
4. Boston - 2.56 (5th/Cup Finals)
5. San Jose - 2.55 (11th/2nd round)
6. St. Louis - 2.54 (6th/1st round)
7. Pittsburgh - 2.54 (2nd/3rd round)
8. Washington - 2.54 (9th/1st round)
9. Toronto - 2.53 (9th/1st round)
10. Philly - 2.53 (20th/DNQ)
11. Montreal - 2.53 (4th/1st round)
12. Detroit - 2.52 (12th/2nd round)
13. Vancouver - 2.51 (7th/1st round)
14. Anaheim - 2.51 (3rd/1st round)
15. New Jersey - 2.48 (21st/DNQ)
16. NYRangers - 2.48 (12th/2nd round)
17 NYIslanders - 2.48 (15th/1st round)
18. Edmonton - 2.48 (24th/DNQ)
19. Dallas - 2.47 (21st/DNQ)
20. Tampa - 2.46 (28th/DNQ)
21. Minnesota - 2.46 (15th/1st round)
22. Calgary - 2.46 (25th/DNQ)
23. Columbus - 2.45 (15th/DNQ)
24. Phoenix - 2.41/DNQ)
25. Winnipeg - 2.41 (18th/DNQ)
26. Colorado - 2.41 (29th/DNQ)
27. Carolina - 2.41 (25th/DNQ)
28. Florida - 2.38 (30th/DNQ)
29. Buffalo - 2.38 (21st/DNQ)
30. Nashville - 2.37 (27th/DNQ)


It gets 14 of the playoff teams (technically 15 as NY rangers and Islanders are tied for 15th with Jersey). It catches teams that were better in the playoffs than the regular season (LA/Ottawa/San Jose). It also catches teams that overachieved in the regular season (Montreal/Anaheim/Vancouver). Three of the final four teams ended up in the top 4.

Philly is the big outlier. Save% may need to be weighted more to correct that. Their other numbers were good.

Its rough, but at least its a complete picture.

But adding those together, you're roughly just duplicating a GF/GA stat.
PP% + Fenwick% + SH% ~= GF
SH% + SV% ~= GA

It is still interesting to see, and might do a better job overall than simply GF/GA.

Going back to points before, it'd be great if we could find some sort of individual WAR stat. The problem will come down to how you treat something like shooting pct. Ideally it should be something like a BABIP type stat, where individual players should have "expected" ranges to see deviations from, but I really don't know how to account for that. Really, until someone finds a way to score each shot (ie. speed and distance from goal). Maybe that could be a better way to play things - divide the shots into a few categories (like baseball has GB/FB/LD), and then we might be able to see more trends on certain types of shots.

leafman101
10-16-2013, 11:55 AM
But adding those together, you're roughly just duplicating a GF/GA stat.
PP% + Fenwick% + SH% ~= GF
SH% + SV% ~= GA

It is still interesting to see, and might do a better job overall than simply GF/GA.

Sure to a certain extent, but isn't that what we are trying to do. Obviously GF/GA decides hockey games. But we are trying to delve deeper to see what causes a team to score more/give up less goals.

worm
10-16-2013, 12:02 PM
Here is 5-5 GF/GA
1 CHICAGO
2 PITTSBURGH
3 NY RANGERS
4 BOSTON
5 ANAHEIM
6 MONTREAL
7 DETROIT
8 LOS ANGELES
9 TAMPA BAY
10 WASHINGTON
11 PHOENIX
12 TORONTO
13 ST LOUIS
14 COLUMBUS
15 OTTAWA
16 VANCOUVER

zeke
10-16-2013, 12:32 PM
I still think the entire team stat approach is backwards from the start.

OFFENSE

- shots are simply not a good proxy for team offense. At all.
- offense is a matter of indi idual talent
- the only useful way to predict offense is to create a useful projection for each individual player and sum it all up in an intelligent way. My 2yr stats do this in a very basic way but its the right kind of idea.

DEFENSE

- most importantly, goalie performance should only be projected based on the goalies' own performance track record. Like offense, this is an individual stat not a team stat.

- shots against ARE a useful proxy for team defense. Not perfect by any means, but useful.
- adjusting shots against for quality might be useful, but i'm not sold on that
- still, i would use recent team shots against totals to project future team defense

- overall defense, then, would be projectrd based on recent team shots against totals and recent individual goalie save percentage


The more i think about it the more i realize that i think the biggest problem with the posession stats is that they treat shots for and against as having the same meaning and value, when i don't think they do.

Shots for are largely under your control, and its quite easy for talented players to pass up easy shots in favor of trying for better scoring chances, and thus beating the odds. Strategy can quite easily effect the quality distribution of shots for - some coaches quite clearly embrace the 'shoot
from everywhere and crash the crease" approach, others clearly embrace the opposite approach.

Shots against don't work the same way - it is very hard for any team to significantly effect the quality distribution of shots against compared to other teams, since every team has the same basic idea of trying to prevent shots against from in close.

I don't know for sure but i'd wager that shots for have a much weaker correlation with goals for than shots against have with goals against (though then again save percentage might **** with that a bit).

One thing i'm not convinced needs to be factored in separately is special teams. I think that would already be accounted for in the other stats.

So yeah, my projection model would be:

1.individual production projections totalled up for team offense, with shots being irrelevant
2.team shots against totals based on recent trends for team defense
3.individual save percentage projections factored into the shots against for team defense

As a sidenote, it's obviously much easier to evaluate the impact of offensive and goale roster changes than defensi e changes.

hockeylover
10-16-2013, 01:05 PM
You know what's going to be really annoying? You don't need any advanced stats to tell you a team can't play at a 140+ point pace for very long. But we're really going to hear it from the metrics guys as soon as we lose a few. Regression and shit.

zeke
10-16-2013, 06:17 PM
Interesting discussion going on on twitter the past coupla days, thanks to a leafs-supporting post from a respected sabermatrician a couple days ago.

The original post: http://blog.philbirnbaum.com/2013/10/corsi-shot-quality-and-toronto-maple.html


Corsi, shot quality, and the Toronto Maple Leafs

he Toronto Maple Leafs had a decent season in 2012-13, finishing fifth in the conference and making the playoffs for the first time since 2004. But, perhaps, we fans of God's Team shouldn't get too optimistic. For months now, hockey sabermetricians have been arguing that the Leafs were still a bad team -- a bad team that just happened to get exceptionally lucky.

But ... I've been fiddling a bit with the numbers, and I'm not sure I agree.

Before I get to my own case, though, let me tell you why the consensus says what it says. First, Sean McIndoe has an excellent Grantland article that summarizes the issue. Second, when you're done that, here's my own summary, which is a bit more statistical......

and a good response: http://www.sbnation.com/nhl/2013/10/16/4842102/nhl-stats-corsi-predictive-value-shot-quality


Is Corsi really as predictive as we think?

Most analysts rely heavily on shot differential, or Corsi, to make predictions for teams. When they do that, they're largely ignoring shooting percentages -- a decision they make based on evidence that shooting percentages have a lot of random variance, and that shot differential is actually a better predictor than the number of goals scored or wins compiled.

But Phil Birnbaum had an interesting post Tuesday challenging that idea, proposing that shooting a lot might be correlated with taking lower-quality shots.

He pointed to some pretty well-established score effects, or changes teams make depending on the score of a game, wherein teams adopt a defensive shell with a lead, taking fewer shots but for a higher percentage. He additionally noted that on some plays a player might pass on a low-percentage shot for the chance at a higher-percentage one, and wondered if there weren't strategies and/or random fluctuations that could result in both a low shot total and a high shooting percentage.....


and another response that is not quite as good: http://www.pensionplanpuppets.com/2013/10/16/4842716/rebutting-phil-birnbaum-on-shot-quality


Rebutting Phil Birnbaum On Shot Quality

I'm not a big believer in shot quality at the team level, as you may already know. But I'm not going to argue about shot quality today. I'm also not a big believer in the ability of 48 game sample sizes to settle team-level statistical debates, but I'm going to try and stay away from that angle too. Instead I'm going to assume that the central premise of Phil's argument is true: that a team could play in a defensive shell that lowered their Corsi but increased their SH% dramatically. Even if that's true, I think you still need to demonstrate that last year's Leafs team did in fact deploy that strategy with that result. Phil doesn't do that (not that I can see, anyway) but I think I can pretty conclusively argue that the Leafs didn't work that way.......

BeLeafer
10-17-2013, 09:24 AM
It's probably been pointed out previously but the weakness of Corsi can be demonstrated by looking at the results for the Leafs under Maurice vs. Carlyle.

2006-07: +4.3, 49%
2007-08: +1.7, 44%
2008-09: +1.4, 41%

Interestingly, win percentage tracks remarkably well with shot differential for Maurice's teams. He was a major proponent of fire at will, quantity over quality.

However, when you compare this against another coach and system, the correlation breaks down. Here's Carlyle's nos. with the Leafs the last two seasons:

2012-13: -6.0, 54%
2013-14: -6.2, 86%*

This is a notably a small dataset (the use of the term "sample size" would be incorrect and its common misuse is a personal pet peeve of mine!). However, you can find the same trend with Carlyle's Anaheim teams.

2011-12: -1.0, 42%
2010-11: -3.8, 57%
2009-10: -3.2, 48%
2008-09: -0.2, 51%
2007-08: -0.7, 57%

His teams consistently show a negative shot differential and, if anything, it correlates negatively with win percentage (e.g. same results in 2011 and 2008 but vastly different shot differential). Clearly, Carlyle uses a coaching approach that focuses on shot quality rather than quantity, unlike the two coaches that preceded him in Toronto. The results belie to the Corsi hypothesis - no doubt created by a coach that thinks shot quantity improves winning chances.

zeke
10-17-2013, 11:11 AM
well we can't push that too far - what the Leafs are doing under Carlyle is quite exceptional, and there is a very large amount of data which does indicate that, at the very least, these kind of exceptions are very rare, and maybe even unsustainable.

There's another descriptions that might be more apt: "The Leafs are a very talented team playing very badly, and winning just based on skill alone".

Of course, nobody in the media wants to actually formulate that argument, because then they would have to admit that the Leafs are playing badly.


But it is funny that when the Ron Wilson coached leafs had a top-5 Corsi/Fenwick and finished 29th in the league, nobody, and I mean nobody, suggested they were a team that was just underachieving and was bound to improve the next year simply due to regression to the mean.

BeLeafer
10-17-2013, 01:45 PM
well we can't push that too far - what the Leafs are doing under Carlyle is quite exceptional, and there is a very large amount of data which does indicate that, at the very least, these kind of exceptions are very rare, and maybe even unsustainable.
This is why I went to Carlyle's numbers in Anaheim. For the coach, it's not exceptional. This is, as the numbers suggest, a coaching issue. Carlyle is clearly telling his teams that firing at will is not the way he likes to go.

In terms of their current record, it's obviously unsustainable. The goaltending is the difference between this 6-1 and a 3-3 or worse, imo. They aren't playing well at all. But they are executing on the offense well and it fits with Carlyle's general approach.


But it is funny that when the Ron Wilson coached leafs had a top-5 Corsi/Fenwick and finished 29th in the league, nobody, and I mean nobody, suggested they were a team that was just underachieving and was bound to improve the next year simply due to regression to the mean.

I did not know that! Just more grist for the advanced stats mill, I guess.

zeke
10-17-2013, 01:53 PM
well we can't push that too far - what the Leafs are doing under Carlyle is quite exceptional, and there is a very large amount of data which does indicate that, at the very least, these kind of exceptions are very rare, and maybe even unsustainable.

There's another descriptions that might be more apt: "The Leafs are a very talented team playing very badly, and winning just based on skill alone".

Of course, nobody in the media wants to actually formulate that argument, because then they would have to admit that the Leafs are playing badly.


But it is funny that when the Ron Wilson coached leafs had a top-5 Corsi/Fenwick and finished 29th in the league, nobody, and I mean nobody, suggested they were a team that was just underachieving and was bound to improve the next year simply due to regression to the mean.

totally screwed up that sentence. Meant to end it with "admit that the Leafs are very talented".

BeLeafer
10-17-2013, 01:54 PM
Funny, even so, I'd read it that way.

PlayerToBeNamedLater
10-25-2013, 01:51 PM
The Leafs don't play very good team defense. Part of it is they play a more open game compared to a lot of teams, part of it is a lack of possession, part of it is personnel.

The Leafs very much rely on their elite offensive talent and elite goaltending to win them games, which isn't a bad thing at all. It's actually a good thing, because you can teach team defense, you can't teach goal scoring.

The Leafs overall shooting % the last 3 years: 12.3% (2nd), 11.5% (1st), 9.8% (6th). This is a stretch of 140 games of elite shooting %. At some point, you just have to admit that they are the exception to the rule.

PlayerToBeNamedLater
10-25-2013, 02:11 PM
Sure to a certain extent, but isn't that what we are trying to do. Obviously GF/GA decides hockey games. But we are trying to delve deeper to see what causes a team to score more/give up less goals.

If you create some forumula that prorates a combination of goals/corsi for every position player, sv% for goalies for every roster based on 100-150ish game sample, maybe with some exceptions that factor in decline for older players, growth for younger players, I think you'd get a very accurate picture of the quality of each team.

Montana
10-25-2013, 02:43 PM
I know you would.

PlayerToBeNamedLater
10-25-2013, 04:02 PM
A good book is "The Numbers Game: Why Everything You Know About Soccer Is Wrong"

Not about hockey, but there are some intersting numbers and concepts that could be applied to hockey because both games are similar in a sense that both aren't a series of independent events, but continuous play, which is a lot harder to breakdown.

One interesting point made was that nowadays everyone is focusing on possession and how to score goals (in both hockey and soccer), but the best approach to counter this would be to focus on how to prevent them. It's not about scoring a lot of goals, it's just about outscoring your opponent.

hockeylover
10-26-2013, 02:13 PM
Pretty close to just abandoning the advanced stats debate altogether. It's pretty exhausting. If I hear regression to the mean one more time... some non-Leaf fans can't even acknowledge the Leafs have a good scoring team.

MindzEye
10-26-2013, 03:05 PM
It's really not worth it. When the Leafs don't suck this year (and I'm pretty much convinced of that now that we're well into the season, have looked terrible imo, yet manage to out talent/skill most teams by such a wide margin we're winning games), they'll just scream "outlier" and we're going to play this same game with them again.

Montana
10-26-2013, 03:35 PM
I dropped it a long time ago, so hearing people complain about the debate has become more annoying than anything, as I don't follow anyone that brings up the other side of the debate.

It's almost like hearing about the Westborro Baptist Church.....on they're own they're completely irrelevant and I'd never hear about them, if it wasn't for people feeding their hype machine, by complaining about them all the time.


The more people ignore the nonsense, the better off everyone is......in the end, the truth has a way of sticking up for itself.

Bleedsblue&white
10-27-2013, 01:14 PM
Enjoying this immensely, glad to see there is good debate going on.

Just for fun here:

The Leafs could be an outlier. Not their play last season...The Team itself, coached a certain way by a certain coach. I mean that in every example ("7 out of 10 teams that win game one win...")there are the exceptions. We say luck, randomness, whatever. I am saying that maybe the numbers ARE predictive of wins, but in no way a guarantee, and we are the team(s) that really do not make "sense," in the Corsi world.

We like to break things into seasons, but with Carlyle coaching, we need to break this team into...generations maybe? Iterations?
As long as Randy is coaching this style, and the team signs the players it does, we will continue to fall outside the norm, and win doing it.

zeke
10-27-2013, 04:15 PM
It's really not worth it. When the Leafs don't suck this year (and I'm pretty much convinced of that now that we're well into the season, have looked terrible imo, yet manage to out talent/skill most teams by such a wide margin we're winning games), they'll just scream "outlier" and we're going to play this same game with them again.

It's true. We have to realize there's literally no sample size big enough for them to ever consider non-corsi/fenwick success as legit.

MindzEye
10-27-2013, 04:40 PM
On a related note, does anyone actually know how to pull those shot location charts off of NHL.com's icetracker? I had some tracking ideas I wanted to play around with but can't manage to find the shot charts.

zeke
10-27-2013, 04:40 PM
I think those are actually ESPN charts.

zeke
10-27-2013, 04:41 PM
If you're going to play around with them, I'd suggest using "touching or within the circles" as a dividing line.

MindzEye
10-27-2013, 04:43 PM
It's true. We have to realize there's literally no sample size big enough for them to ever consider non-corsi/fenwick success as legit.

I'm wondering how many actual statisticians are among them, because it's my laymen's understanding that any individual statistic that shows less than a 90% correlation with real world results is pretty much considered entirely invalid across most disciplines.

I could be wrong, but that's the threshold I've heard quoted before.

hockeylover
10-27-2013, 04:45 PM
On a related note, does anyone actually know how to pull those shot location charts off of NHL.com's icetracker? I had some tracking ideas I wanted to play around with but can't manage to find the shot charts.

http://scores.espn.go.com/nhl/gamecast?gameId=400484399

Go to the gamecast option in each of the Game Summaries.

MindzEye
10-27-2013, 04:45 PM
If you're going to play around with them, I'd suggest using "touching or within the circles" as a dividing line.

Yeah, I was thinking of dividing the slot a few different ways, separating premium area point shots from the corners, etc.

zeke
10-27-2013, 04:47 PM
I'm wondering how many actual statisticians are among them, because it's my laymen's understanding that any individual statistic that shows less than a 90% correlation with real world results is pretty much considered entirely invalid across most disciplines.

I could be wrong, but that's the threshold I've heard quoted before.

I'm not sure the exact numbers but I think once a single stat can get ~60% correlation that's pretty good. But I could be wrong on that too.

MindzEye
10-27-2013, 04:47 PM
http://scores.espn.go.com/nhl/gamecast?gameId=400484399

Go to the gamecast option in each of the Game Summaries.

They are, you're a beast.

MindzEye
10-27-2013, 04:49 PM
I'm not sure the exact numbers but I think once a single stat can get ~60% correlation that's pretty good. But I could be wrong on that too.

Oh, 55% is pretty good. It's just not "pretty good" enough to draw concrete conclusions imo, which is exactly what the analytics community seems to have done.

I kind of taking the Fontana meme from Anchorman and applying it to his sex panther quote. It works 55% of the time, every time.

zeke
10-27-2013, 05:00 PM
The thing is, I'm still thinking that there's a bit of conflating of correlation with causation going on with these numbers right now.

Basically, I'd be willing to bet that in the vast majority of cases, a team's Corsi/Fenwick numbers would correlate quite strongly with other methods of projection - i.e. player-by-player PPG and SV% projections. So in most cases, I'd bet that when Corsi/fenwick sees an over or underachievement, then the individual projections would see that same over or underachievment (i.e you'd be able to spot clear unsustainable outlier PPG and SV% performances from career norms or reasonable expectaions).

What I'd be more interested in looking at is the subset of teams in which the Corsi/Fenwick numbers strongly disagree with the individual projections numbers, like this leafs team. I'd wager there aren't that many teams that have been in the Leafs situation - where their performance in the standings is 100% justified by their individual career track records, but completely unjustified by the Corsi/Fenwick numbers. Basically, the subset of teams where the two methods of team evaluation contradict each other this strongly. And I'd like to see which method turns out to be more relevant amongst that subset of teams.

MindzEye
10-27-2013, 05:33 PM
Okay...just had a chance to dig through the shot charts from this season and had a few things to share. Basically what I did was took a big rectangle from the outside of the faceoff dots in, from the goal line to the top of the circles. Then tapered it in slightly from the bottom of the circles to the goal line so , basically making a retarded looking hexagon.

- First observation...other than the 2 EN goals (Bolland & McClement), every single Leafs goal this season has been scored from that area. All of them, from the most dangerous area of the ice

- Only 6 goals against have come from outside of this zone

- The only win in which the Leafs were "badly" outshot within this zone was the win against Philly when Bernier was superhuman. We were outshot 8-17 from within that high leverage zone in that game. In no other win were the Leafs outshot within this zone by more than 4 shots. In their 4 losses, the Leafs were outshot by a spread of 8 in 2 of the 4 games lost.

- Overall the Leafs the average 15.5 shots from within that area, and give up 16.9

- In Leaf wins the shot spread within that area was 15.6/16.0, in losses it was 15.25/18.75

- 58% of Leafs shots for, were produced in this zone. 48% of shots against originated in this zone.