Hurricanes @ Leafs - Oct. 17th, 7:00pm - LeafsTV, SN 590

[crazyperfectdevil]

Interesting to see the group of articles from the major papers today talking about how the Leafs need to improve their puck possession game and stop being heavily outshot and outchanced.  "Corsi" by any other name....

Even Carlyle's getting in on it.

I want the Leafs to improve upon games won, and cut down on lost games.  So far so good.

Then no point wading into any stats-based discussion then right?  Just watch for the end result and leave those of us who want to look deeper into how the team is playing to do that.  I am at wit's end at how to avoid people becoming increasingly defensive about something being discussed that isn't revolutionary.

How did you get that from what he said? 

Anyway I think all of us here want the same thing..the leafs to win.  Like as much as I know certain stats show as as being potentially in trouble.  I'm not sure I'm going to feel that much better tonight if the leafs have more puck possession and more shots than their opponent and lose the game....

 

[CarltonTheBear]

Interesting to see the group of articles from the major papers today talking about how the Leafs need to improve their puck possession game and stop being heavily outshot and outchanced.  "Corsi" by any other name....

Even Carlyle's getting in on it.

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

 

[GhostofPotvin29]

How did you get that from what he said? 

Not so much him as hockey fans in general - you get a lot of backhanded remarks like that.  His comment was perfectly congenial.  I'm sure it was light-hearted, but in a way still defensive about commenting on issues like that despite the team being 6-1.

I'm not sure I'm going to feel that much better tonight if the leafs have more puck possession and more shots than their opponent and lose the game....

No one is saying you should.  That's going to happen a bunch of times in a season (like it happened to Minnesota).  The idea is that, generally speaking, you have a better probability of winning if you are on the positive side of that ledger more often than not.  It's probability, not a definite.  That's why you still play the games, while you still watch the games.  It's just something that you want to be doing more than not doing it, because teams tend to win more when they do generally speaking.  There are always exceptions - strong special teams play + strong goaltending can buoy poor 5 on 5 possession teams up, too.  And it's not an exact science.

 

[TML fan]

Interesting to see the group of articles from the major papers today talking about how the Leafs need to improve their puck possession game and stop being heavily outshot and outchanced.  "Corsi" by any other name....

Even Carlyle's getting in on it.

I want the Leafs to improve upon games won, and cut down on lost games.  So far so good.

Then no point wading into any stats-based discussion then right?  Just watch for the end result and leave those of us who want to look deeper into how the team is playing to do that.  I am at wit's end at how to avoid people becoming increasingly defensive about something being discussed that isn't revolutionary.

How did you get that from what he said? 

Anyway I think all of us here want the same thing..the leafs to win.  Like as much as I know certain stats show as as being potentially in trouble.  I'm not sure I'm going to feel that much better tonight if the leafs have more puck possession and more shots than their opponent and lose the game....

It's not sustainable over the long term, and it won't work in the playoffs. We all saw how quickly it can turn around last year. I don't care to see it again. On a game by game basis, sure a win is a win. We're all happy about that, but I don't understand how one can think that getting massively outplayed every night is a recipie for long term success.

The Leafs record is flattering, and a testament to their goaltending, or in the case of Saturday's game, Edmonton's goaltending.

 

[bustaheims]

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

So, basically, if its names was a little more upfront about what it is? Yeah, I guess, though, I'd still argue that it doesn't do a good job of representing what it's being used to represent on an individual to individual level.

 

[Frank E]

Interesting to see the group of articles from the major papers today talking about how the Leafs need to improve their puck possession game and stop being heavily outshot and outchanced.  "Corsi" by any other name....

Even Carlyle's getting in on it.

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

I think it's pretty common for people to be skeptical about new things that they don't understand.

 

[GhostofPotvin29]

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

So, basically, if its names was a little more upfront about what it is? Yeah, I guess, though, I'd still argue that it doesn't do a good job of representing what it's being used to represent on an individual to individual level.

What's it being used to represent on an individual to individual level? 

 

[CarltonTheBear]

So, basically, if its names was a little more upfront about what it is? Yeah, I guess, though, I'd still argue that it doesn't do a good job of representing what it's being used to represent on an individual to individual level.

Basically. The term corsi brings up the thoughts that it's an advanced stat and that seems to unease some people. But really what's so advanced about looking at how many shots a team allows vs. how many they produce.

It's obviously not a perfect stat. But I think most opponents of it say that people are trying to use it to predict seasons and they bring up the whole "the game is played on the ice not on a spreadsheet" line. But no supporter of corsi or advanced stats would ever argue against that. But history has shown that teams that produce more shots usually win games. Not always, but usually. Now we have a stat that can follow that. It's not perfect but neither is +\- or any of the RTSS stats. And most people aren't losing their minds every time they are brought into the discussion.

 

[crazyperfectdevil]

Interesting to see the group of articles from the major papers today talking about how the Leafs need to improve their puck possession game and stop being heavily outshot and outchanced.  "Corsi" by any other name....

Even Carlyle's getting in on it.

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

I think it's pretty common for people to be skeptical about new things that they don't understand.

is it that?  or is it the fact that people generally pick apart stats.  Corsi and the like may be new..but stats aren't.  I see people take shots at +/- all the time ..or say shots on goal don't tell the whole story..and so on.  Isn't this just an expression of the same thing that is basically people wanting to say..that doesn't tell the whole story..that you still have to watch the game?  maybe things seem extra defensive right now because this stuff is relatively new.. (and i'd say both sides are seemingly a little on edge about it) but over the long run i see this being the same deal. 

I even think that both "sides" would probably agree to this ...it's just a matter of emphasis ..and when you defend a position you make it seem more extreme.  So instead of hearing "corsi is a good way to gauge how a team is performing..like shots on goal ..time on attack ..etc."  the otherside hears "no team ever does anything without corsi ..it's corsi or bust"  ..likewise the stats people don't end up hearing "I don't feel like focusing on these numbers does much to add to my enjoyment of the game and they're not definitive.."  it comes across like "i hate all numbers!"

 

[GhostofPotvin29]

So, basically, if its names was a little more upfront about what it is? Yeah, I guess, though, I'd still argue that it doesn't do a good job of representing what it's being used to represent on an individual to individual level.

Basically. The term corsi brings up the thoughts that it's an advanced stat and that seems to unease some people. But really what's so advanced about looking at how many shots a team allows vs. how many they produce.

It's obviously not a perfect stat. But I think most opponents of it say that people are trying to use it to predict seasons and they bring up the whole "the game is played on the ice not on a spreadsheet" line. But no supporter of corsi or advanced stats would ever argue against that. But history has shown that teams that produce more shots usually win games. Not always, but usually. Now we have a stat that can follow that. It's not perfect but neither is +\- or any of the RTSS stats. And most people aren't losing their minds every time they are brought into the discussion.

Something worth noting through 6 games: http://www.extraskater.com/teams/2013/5v5?sort=pdo

For those that don't know, PDO number is just the addition of the teams even-strength SV% with their even-strength SH%.  Last season the Leafs had the highest "PDO" in the league at 103.  Typically, being too high or too low of 100 is a sign of good/bad luck.  So far this season the Leafs are 8th in the league at 102.7 - SH% of 9.4% and SV% of 93.3 (last season was 10.7% and 92.3%). 

If they continue to get the kind of goaltending these two goalies can give, I think they can be okay over the course of a season, although it's too early to take too much out of these numbers.  Their PP & PK have the potential to off-set any even-strength regression too.

 

[bustaheims]

What's it being used to represent on an individual to individual level?

People use it to represent the quality of an individual player's defensive play/influence on puck possession, and, well, just like +/-, there are some obvious flaws with that. Then, of course, it gets rolled into a bunch of other stats - like QoC and the various other tweaks that are used to represent Corsi on an individual basis - and, really, it just means those stats maintain that flaw (or potentially even exacerbate it).

On a macro basis, Corsi has some value, but on a micro basis, the other players on the ice have too much influence on it for me to see it as a valuable measurement.

 

[GhostofPotvin29]

What's it being used to represent on an individual to individual level?

People use it to represent the quality of an individual player's defensive play/influence on puck possession, and, well, just like +/-, there are some obvious flaws with that. Then, of course, it gets rolled into a bunch of other stats - like QoC and the various other tweaks that are used to represent Corsi on an individual basis - and, really, it just means those stats maintain that flaw (or potentially even exacerbate it).

On a macro basis, Corsi has some value, but on a micro basis, the other players on the ice have too much influence on it for me to see it as a valuable measurement.

I think it has potential if you view a player over multiple seasons and start to see a trend.  If a player, with different linemates etc, season to season to consistently having a positive number then it could be the basis to say this player is a key to driving possession on this line.  It's hard to use it usefully in a pointed context though.

So overall, agreed.

 

[GhostofPotvin29]

JVR has back spasms.  Won't play tonight and is day-to-day.

 

[crazyperfectdevil]

JVR has back spasms.  Won't play tonight and is day-to-day.

i hope we're not playing the role of the ottawa senators this year

 

[freer]

JVR has back spasms.  Won't play tonight and is day-to-day.

i hope we're not playing the role of the ottawa senators this year

I know that not having regulars in the line is not good. But I am not bitching if we are winning with the farm team up either. And the OTT role last season working very well for them. Just saysing

 

[proteus2000]

Pretty amazing that the Leafs have obvious spots that they need to work on, yet have won 6 of 7 games.

I think corsi and stats like puck possession/SOG statistically do make sense for the most part, but the Leafs buck that trend as the team can snipe, they have great special teams, and their goaltending duo is potentially the best in the league. These flaws will matter more come playoff time but they have 74 games to prepare.

 

[Kin]

There's an argument that pretty much says if Corsi was called shot +\- it would be a lot more accepted. I can't disagree with that.

But if the argument is as simple as "it's better to outshoot your opponent than it is to be outshot by them" in a sort of over-arching sense then why would you need to come up with any name for it outside of shots for and shots against which have been tracked forever by the NHL? Unless, again, it was to try and represent something on an individual level which, as mentioned, is where it begins to rub up against its flaws. 

 

[Bonsixx]

Was kinda hoping Bernier would get this one. Reimer wasn't very good against Carolina last year (3.50 GAA, .885 Save% in two games), not that that matters now, I guess.

 

[proteus2000]

Was kinda hoping Bernier would get this one. Reimer wasn't very good against Carolina last year (3.50 GAA, .885 Save% in two games), not that that matters now, I guess.

It feels like even though CAR has been pretty bad the last couple of yrs the Leafs really struggle against them. I don't think it will be an easy game.

 

[Kin]

maybe things seem extra defensive right now because this stuff is relatively new.. (and i'd say both sides are seemingly a little on edge about it) but over the long run i see this being the same deal. 

But I think the problem, from my perspective anyway, is that while this stuff is relatively new in a hockey sense the idea of using stats to look at sports in a fresh way isn't. The people advocating Corsi aren't Bill James in 1983, we live in a post-Moneyball world. Hell we live in a post Moneyball: The Movie world. The idea that people are inherently resistant to that kind of analysis doesn't really ring true.

And for someone like me, the problem I have with it isn't that it's new or that I don't understand it or I don't like numbers, it's that because I know about the ongoing development in using numbers to analyze sports it seems like the people who are going in whole hog in this one area are essentially ignoring lessons learned in other sports on the matter. Things like Corsi has the exact same flaws that a lot of the early basketball metrics had where they tried to baseball-ize analysis and come to pretty specific conclusions from the raw data in spreadsheets which doesn't really work for a sport where five people are constantly involved in the play on both offense and defense. It's only when analysis in that sport moved away from that and towards the study of tape and grading plays and accounting for each individual action on the court that things really began to take hold.

That revolution happened already. What I'm seeing in hockey stats doesn't seem problematic because it's so new, it's because it seems like a regression.