• Moderators, please send me a PM if you are unable to access mod permissions. Thanks, Habsy.

Leafs - Habs - Sens

yeah, I get that they're tougher to understand, but I don't think they're near as tough to understand as you guys think. Right now the biggest problem is you guys don't have a reference to what a good or bad p60 is, and I'm not sure that there's any way to fix that aside from experience. One of the biggest things here was splitting out ES and PP scoring. It just seems way too important when trying to compare players to each other not to factor that in.

I do think the possesion numbers should be pretty simple to understand - one is a percentage where 50% = average and the other is a +/- relative number where average = 0, so I don't think I can make that any easier to understand. That's no more complicated than understanding regular old +/-.

and I gotta be honest I have fun wasting time with these numbers but only because I think they're useful. would be tough for me to waste that time on numbers I don't think are actually very good anymore (last couple years were a bit of a chore). maybe you guys could just trust me that once you get used to them they'll make you way more informed than you used to be?

that being said I do love the letter grade idea I came up with (with an assist to beleafer) because that makes it easier for everyone to understand. and I did use pace instead of ppg originally ibecause it was easier to understand. So maybe I can come up with something easier.
 
That's ultimately what it comes down to....how informed do you want to be.

I have family members who love the Leafs but don't care about points per game, cap hit, or save percentage.....they care about wins and loses and what their eye test tells them on game night.

...and that's perfectly fine if that's what you enjoy. It's ultimately all just for entertainment, so whatever flavor you prefer is up to you.

Sticking to simple PPG because it's what you know & can easily process because of past experience is fine too......but as zeke says, if you wish to be as informed as possible, and take a small amount of time to familiarize yourself with the finer details of some of these advanced stats, you have an opportunity to become way more informed.


If that isn't your flavor of engagement....all the power to you. Just know it's there, and concede you are less informed than you could be.

Ultimately it's whatever floats your boat.....but speaking from experience. I'm real glad I spent the time to inform myself. Well worth the minimal effort....and has made Zeke's pre-game posts exponentially more informative than they were in the old PPG days.
 
Montana said:
That's ultimately what it comes down to....how informed do you want to be.

I have family members who love the Leafs but don't care about points per game, cap hit, or save percentage.....they care about wins and loses and what their eye test tells them on game night.

...and that's perfectly fine if that's what you enjoy. It's ultimately all just for entertainment, so whatever flavor you prefer is up to you.

Sticking to simple PPG because it's what you know & can easily process because of past experience is fine too......but as zeke says, if you wish to be as informed as possible, and take a small amount of time to familiarize yourself with the finer details of some of these advanced stats, you have an opportunity to become way more informed.


If that isn't your flavor of engagement....all the power to you. Just know it's there, and concede you are less informed than you could be.

Ultimately it's whatever floats your boat.....but speaking from experience. I'm real glad I spent the time to inform myself. Well worth the minimal effort....and has made Zeke's pre-game posts exponentially more informative than they were in the old PPG days.
Click to expand...
This is a good time to humble myself and ask (anyone can answer this question), is there a good place to learn advanced stats so I can understand and discuss them? I'm pretty much still a beginner right now.
 
Don't think anyone said informing yourself, or advanced stats are bad, or that zeke should include less information. Opposite actually. More info.

For example, according to the numbers Martin is 3rd on the team among forwards in p/60. But he is 3rd worst in actual production, and 4th worst in ES production and only has one goal. It would just be nice to have the complete data set so nothing needs to be read between the lines.

The other thing is p/60 weighs goals, primary and secondary assists equally.
 
Last edited:
putting the pace points in kind of screws up my whole format though.

but on the other hand I want it to be accessible so no need for me to be stubborn.

i'll try to work something out.
 
zeke said:
putting the pace points in kind of screws up my whole format though.

but on the other hand I want it to be accessible so no need for me to be stubborn.

i'll try to work something out.
Click to expand...

I'd just throw them on the end, as a simple 25-35-60.
 
Jeremy said:
This is a good time to humble myself and ask (anyone can answer this question), is there a good place to learn advanced stats so I can understand and discuss them? I'm pretty much still a beginner right now.
Click to expand...

advanced stats are simple. don't ever think they're not.

You know back in the old days when we'd say "nice win but we can't get outshot like that"? well, that was pretty much us using "advanced" stats to diagnose the game more accurately than the score may have told us. This is actually funny - a guy like don cherry will rant against advanced stats but then say "oh man the leafs can't get outshot like that".....without realizing that's really all that advanced stats are trying to say.

But now we don't just use shots that hit the net, but also use shots that miss the net and that are blocked - i.e. shot attempts. This is all "corsi" is - it is shot attempts.

Basically the math tells us that the ability to out-shot-attempt the other team at even strength is the most reliable indicator of a good team. The more a team does that, the better they are. And this is a better stat to look at in a smaller sample than just goals for/against (which we all already knew - going back to our "oh man the win is nice but we can't keep getting outshot like this" arguments before).

And likewise, instead of using plus/minus for individual players, we can use a +/- of shot attempts instead - or in this case, we make it a percentage instead of a +/-. (we could have done this with ordinary +/- too....if a player had been on the ice for 5 goals for and 3 against, we could say he was +2, or we could say he was on the ice for 62.5% of the goals for - or 62.5gf%).

So when we say that a player has a 55cf% - all we are saying is that that player is on the ice, 55% of the total shot attempts on the ice both ways, were shot attempts for his team. That's all. So if he's on the ice for 11 shot attempts for and 9 shot attempts against, he has a 55cf%.

That's the basics. It's really simple.


where it does get more complicated is when we start adjusting the numbers.....but, at the same time, that's where all the power of the advanced stats really comes into play.

you know how we used to have to say "sure that guy is a +32, but he just plays easy bottom pair minutes for an elite team, so that doesn't mean as much as we think"? well now we don't have to say that (or, at least not as much). We can now see how the entire league does in certain situations on averaage - i.e. how players or teams on average perform on shot attempts with a 2 goal lead in the 3rd period when they start the shift in the defensive zone - we can look at the league average performance in all those different situations and then compare one player's performance to each of those averages, and adjust out most of the stuff that used to make the old numbers useless.

but the good news is that none of us really have to understand those adjustments - we can just get that they're levelling out the playing field for players being used in different roles, so that we can make a fair comparison between a sheltered offensive defenseman and a tough defensive dman taking all the tough shifts.


there is one major problem with all the advanced stats, though, and that's that they haven't really figured out how to adjust for quality of compeition as of yet - which is why for now I just include the quality of competition as a stand alone (the letter grades), which we just have to keep in mind when looking at the other numbers and try to get an idea over time of how much it matters.


but remember, at the base of it all, is just saying that being on the ice for more shot attempts for your team than you allow the other team is a good thing. That's all.

So if you're up over 50% in corsi, that's a good thing.
 
UWHabs said:
I'd just throw them on the end, as a simple 25-35-60.
Click to expand...

the lines are already too long though. i'm gonna have to go with full spreadsheet format I think so I can drop all the letters. I should have done it long ago but i had my system already in place. I'll figure it out.
 
zeke said:
yeah, I get that they're tougher to understand, but I don't think they're near as tough to understand as you guys think. Right now the biggest problem is you guys don't have a reference to what a good or bad p60 is, and I'm not sure that there's any way to fix that aside from experience. One of the biggest things here was splitting out ES and PP scoring. It just seems way too important when trying to compare players to each other not to factor that in.

I do think the possesion numbers should be pretty simple to understand - one is a percentage where 50% = average and the other is a +/- relative number where average = 0, so I don't think I can make that any easier to understand. That's no more complicated than understanding regular old +/-.

and I gotta be honest I have fun wasting time with these numbers but only because I think they're useful. would be tough for me to waste that time on numbers I don't think are actually very good anymore (last couple years were a bit of a chore). maybe you guys could just trust me that once you get used to them they'll make you way more informed than you used to be?

that being said I do love the letter grade idea I came up with (with an assist to beleafer) because that makes it easier for everyone to understand. and I did use pace instead of ppg originally ibecause it was easier to understand. So maybe I can come up with something easier.
Click to expand...
I think the reference point is the hardest thing to get.

Also when talking with the general public it is hard to explain this stuff to them as they still reference plus minus as a stat that means you are good at defense.


But do as you want. I enjoyed the pace stats but also appreciate the current state.

Sent from my Pixel using Tapatalk
 
I'd toss out the corsi and throw in either ppg or SCF%. I think corsi is a pretty useless stat on an individual level, as it is heavily affected by team play and situational play. It's a glorified +/-, as far as I'm concerned.
 
I think the data shows CF% is actually a better predictor on an individual level than SCF or even xGF simply because the sample size is much bigger. But they do also track individual CF which could be interesting to use.

Also they have "catch-all" stats now like WAR, GAR and Game Score which might be interesting to track. Although as far as the analytics go I think zeke's table does a good job of capturing most key events.
 
zeke said:
advanced stats are simple. don't ever think they're not.

You know back in the old days when we'd say "nice win but we can't get outshot like that"? well, that was pretty much us using "advanced" stats to diagnose the game more accurately than the score may have told us. This is actually funny - a guy like don cherry will rant against advanced stats but then say "oh man the leafs can't get outshot like that".....without realizing that's really all that advanced stats are trying to say.

But now we don't just use shots that hit the net, but also use shots that miss the net and that are blocked - i.e. shot attempts. This is all "corsi" is - it is shot attempts.

Basically the math tells us that the ability to out-shot-attempt the other team at even strength is the most reliable indicator of a good team. The more a team does that, the better they are. And this is a better stat to look at in a smaller sample than just goals for/against (which we all already knew - going back to our "oh man the win is nice but we can't keep getting outshot like this" arguments before).

And likewise, instead of using plus/minus for individual players, we can use a +/- of shot attempts instead - or in this case, we make it a percentage instead of a +/-. (we could have done this with ordinary +/- too....if a player had been on the ice for 5 goals for and 3 against, we could say he was +2, or we could say he was on the ice for 62.5% of the goals for - or 62.5gf%).

So when we say that a player has a 55cf% - all we are saying is that that player is on the ice, 55% of the total shot attempts on the ice both ways, were shot attempts for his team. That's all. So if he's on the ice for 11 shot attempts for and 9 shot attempts against, he has a 55cf%.

That's the basics. It's really simple.


where it does get more complicated is when we start adjusting the numbers.....but, at the same time, that's where all the power of the advanced stats really comes into play.

you know how we used to have to say "sure that guy is a +32, but he just plays easy bottom pair minutes for an elite team, so that doesn't mean as much as we think"? well now we don't have to say that (or, at least not as much). We can now see how the entire league does in certain situations on averaage - i.e. how players or teams on average perform on shot attempts with a 2 goal lead in the 3rd period when they start the shift in the defensive zone - we can look at the league average performance in all those different situations and then compare one player's performance to each of those averages, and adjust out most of the stuff that used to make the old numbers useless.

but the good news is that none of us really have to understand those adjustments - we can just get that they're levelling out the playing field for players being used in different roles, so that we can make a fair comparison between a sheltered offensive defenseman and a tough defensive dman taking all the tough shifts.


there is one major problem with all the advanced stats, though, and that's that they haven't really figured out how to adjust for quality of compeition as of yet - which is why for now I just include the quality of competition as a stand alone (the letter grades), which we just have to keep in mind when looking at the other numbers and try to get an idea over time of how much it matters.


but remember, at the base of it all, is just saying that being on the ice for more shot attempts for your team than you allow the other team is a good thing. That's all.

So if you're up over 50% in corsi, that's a good thing.
Click to expand...
Thx Zeke! U da man. Will try to catch on little by little
 
I know a lot of you have poo-poo'd on the +/- stat over the years ... but Auston is leading the league with a +16
 
Cause whatever it is that it says... you can say the same for Jason Zucker last year. Except it's more definitive about him cause he did it for the entire season.
 
I'm just saying that if it were completely random and baseless, that's one thing, but it still is a positive stat that is better than every other player's rating.

I don't like the stat personally. It has obvious flaws. But there's some comparative value in being the leader in the category, flaws aside.
 
What might that be though? What can you say about Jason Zucker based on him being having the best +/- in the league last season?

That's a legitimate question... cause I honestly don't know what it says about a player.
 
You must log in or register to reply here.
Back
Top