Advanced Stats: Corsi, QoC, etc

[Turrican*]

Does it seem that these stats are used as crutches to propel a player with less points or ppg, seem better than a player with a higher point total?

I mean, if Stamkos or Giroux have weaker advanced stats than another player, would you take the other player?

Or are these stats only suitable in discussions when considering B-tier and C-tier players?

 

[NyQuil]

People use them around here like they are gospel, when are they are really telling you is about shots for/shots against.

They can add a dimension of analysis, but a lot of people mistakenly rely on them as a proxy for a player's overall effectiveness.

 

[Hockey Monkey]

The point of advanced stats is to suggest that there are a lot of different ways to judge a player's effect on a game rather than just point total. Like any other statistical measure, they have to be taken in context to be effective.

If you view this as a "crutch" or something that's your decision I guess.

 

[Arrch]

It's not about player X has better advanced stats than Player Y, therefore Player X is better. Advanced stats help provide context to a players typical stats like points and +/-.

 

[Micklebot]

Does it seem that these stats are used as crutches to propel a player with less points or ppg, seem better than a player with a higher point total?

I mean, if Stamkos or Giroux have weaker advanced stats than another player, would you take the other player?

Or are these stats only suitable in discussions when considering B-tier and C-tier players?

Advanced stats can be used to tease out certain variable that may effect the more traditional stats, like the system played, but it really can't be used on its own. It simply helps add some additional context to an argument, for example, if you have 2 players with similar pts/gm, but greatly different QOC, you might lean towards the guy with a higher QOC, but it's by no means perfect.

 

[Sureves]

It's not about player X has better advanced stats than Player Y, therefore Player X is better. Advanced stats help provide context to a players typical stats like points and +/-.

This. Statistics themselves are almost entirely useless. It is the discussion, application, and circumstantial consideration that brings value to them.

If used properly, advanced hockey metrics are extremely powerful in explaining a lot of things, IMO.

 

[MasterofGrond]

Who has actually used advanced stats as a way to justify another player over Stamkos or Giroux?

I'm a huge supporter of advanced stats. I happen to think they can be very illuminating about which players really contribute to winning hockey games, but if someone says Justin Williams is better than Malkin because he has a better Corsi, the problem isn't with advanced stats. The problem is with the guy who doesn't understand them.

 

[PumpkinBombX]

The thing about hockey is it's a team sport, and team stats are a better predictor for wins than individual stats. Stats in baseball are very useful because baseball is really an individual sport. It's always 1 vs 1.

A good example is: Let's say you have your number 1 D-man playing 30 min a night and he's awesome with those minutes. You also have a #3 playing 20 min a night and is also awesome. Good point totals, good +/- etc. If that top D gets injured and then the #3 has to cover the extra minutes, quite often you'll see his point totals dip along with the rest of the stats because he cant handle the extra minutes.

Jay Bouwmeester (I think) is a great example of this. I think if he played less minutes you'd see better point totals out of him because playing the minutes he does he can only really play defensively, and can't really be physical due to fatigue.

 

[TheHudlinator]

Of course you can't just use the advanced stats that would be like using point totals for defense men to determine who is the better player. But you can use them to get deeper understanding of a player is a good why to compare to similar players.

If you know what you are doing with statistics they can be very powerful.

 

[Freudian]

They are line stats more than player stats. But those that throw them around often don't understand this and use them as proof of how good a player is.

They are inferior to observation when it comes to telling us how good a player is.

 

[TOML]

For example, the Sedins lead the league in Corsi Relative (would've been healthy Crosby), but the Glencross-Iglina-Jokinen lead the league in Corsi Relative to QoC.

Does that mean Glencross-Iglina-Jokinen are better than the Sedins? Or does it mean the QoC stat is a bit arbitrary?


imo These stats reflect how the team plays. For example, NYR's excellent record from their reliance on Lundqvist combined with having a relatively injury-free season is highlighted by their low corsi ratings. And Tampa's 'surprise' success last season despite their shaky goaltending was shown by their high corsi ratings.

At the moment these are the teams with the highest corsi:

Pittsburgh
Detroit
Boston
StL
Van
SJ


So you can see tonight's battle between NYR (superior health + reliance on Lundy) vs. Pittsburgh (dominant corsi) being quite indicative of how the playoffs in the east might play out. Esp. when both teams aren't as banged-up as Boston. (Yes, you can't rule out Philly either, but they're a bit banged-up as well.) Cheers.


TOML

 

[tarheelhockey]

If two players have the same stat line, but one of them has much higher quality of competition and far more starts at the defensive end of the ice, then who is performing better?

That's the logic behind advanced stats. You still need to know those players' basic numbers before the advanced numbers can really be useful.

 

[TheHudlinator]

For example, the Sedins lead the league in Corsi Relative (would've been healthy Crosby), but the Glencross-Iglina-Jokinen lead the league in Corsi Relative to QoC.

Does that mean Glencross-Iglina-Jokinen are better than the Sedins? Or does it mean the QoC stat is a bit arbitrary?


imo These stats reflect how the team plays. For example, NYR's excellent record from their reliance on Lundqvist combined with having a relatively injury-free season is highlighted by their low corsi ratings. And Tampa's 'surprise' success last season despite their shaky goaltending was shown by their high corsi ratings.

At the moment these are the teams with the highest corsi:

Pittsburgh
Detroit
Boston
StL
Van
SJ


So you can see tonight's battle between NYR (superior health + reliance on Lundy) vs. Pittsburgh (dominant corsi) being quite indicative of how the playoffs in the east might play out. Esp. when both teams aren't as banged-up as Boston. (Yes, you can't rule out Philly either, but they're a bit banged-up as well.) Cheers.


TOML

This is what I was talking about you can't use one stat to say players are better you must use all of them including ppg. Or else end up this stupid arguments like above.

 

[BillDineen]

Here is a 1hr10min hockey analytics conference at MIT that includes some team reps including Stan Bowman.

http://video.mit.edu/watch/hockey-analytics-7187/

Corsi's weakness is that it treats all shots, missed shots and blocked shots equally and defines QoC by the same metric. Teams appear to keep their own stats which are better than the public stats used in the blogosphere's articles.

For instance if you kept possession time by zone as the NHL did in the 90s, then you could go through the same analysis by defining QoC in terms of other player's relative possession time by zone etc, etc and deriving conclusions based on that. So the method may be sound, but the underlying stat (Corsi) is not close to perfect.

 

[Micklebot]

Here is a 1hr10min hockey analytics conference at MIT that includes some team reps including Stan Bowman.

http://video.mit.edu/watch/hockey-analytics-7187/

Corsi's weakness is that it treats all shots, missed shots and blocked shots equally and defines QoC by the same metric. Teams appear to keep their own stats which are better than the public stats used in the blogosphere's articles.

For instance if you kept possession time by zone as the NHL did in the 90s, then you could go through the same analysis by defining QoC in terms of other player's relative possession time by zone etc, etc and deriving conclusions based on that. So the method may be sound, but the underlying stat (Corsi) is not close to perfect.

True, something that measured scoring chances in the same fashion, rather then shots, would likely be better, or at very least a good compliment.

 

[Thirty One]

For example, the Sedins lead the league in Corsi Relative (would've been healthy Crosby), but the Glencross-Iglina-Jokinen lead the league in Corsi Relative to QoC.

Does that mean Glencross-Iglina-Jokinen are better than the Sedins? Or does it mean the QoC stat is a bit arbitrary?


imo These stats reflect how the team plays. For example, NYR's excellent record from their reliance on Lundqvist combined with having a relatively injury-free season is highlighted by their low corsi ratings. And Tampa's 'surprise' success last season despite their shaky goaltending was shown by their high corsi ratings.

At the moment these are the teams with the highest corsi:

Pittsburgh
Detroit
Boston
StL
Van
SJ


So you can see tonight's battle between NYR (superior health + reliance on Lundy) vs. Pittsburgh (dominant corsi) being quite indicative of how the playoffs in the east might play out. Esp. when both teams aren't as banged-up as Boston. (Yes, you can't rule out Philly either, but they're a bit banged-up as well.) Cheers.


TOML

The Rangers often don't pursue the puck hard in the defensive zone. They just clog the middle of the ice, letting the opponents cycle the puck along the perimeter while they clog the shooting lanes, to ensure the shots are either blocked or wide. It's not surprising they have a poor Corsi, it doesn't, in my opinion, indicate they are any poorer of a team, it will just take 5 years of Dan Girardi's life.

Tying lack of injuries to Corsi is a point that needs further explanation.

 

[Unaffiliated]

People use them around here like they are gospel

Never seen that here. Maybe I don't participate in PlayerX vs PLayerY debates enough

 

[Les Wynan*]

Here is a 1hr10min hockey analytics conference at MIT that includes some team reps including Stan Bowman.

http://video.mit.edu/watch/hockey-analytics-7187/

Corsi's weakness is that it treats all shots, missed shots and blocked shots equally and defines QoC by the same metric. Teams appear to keep their own stats which are better than the public stats used in the blogosphere's articles.

For instance if you kept possession time by zone as the NHL did in the 90s, then you could go through the same analysis by defining QoC in terms of other player's relative possession time by zone etc, etc and deriving conclusions based on that. So the method may be sound, but the underlying stat (Corsi) is not close to perfect.

Corsi has been shown to be highly correlated with zone time.

 

[BillDineen]

Never seen that here. Maybe I don't participate in PlayerX vs PLayerY debates enough

There are some Flyers bloggers that claim they prove their points because of corsi based advanced stats. Those opinion sometimes get quoted and reach the Flyers forum.

 

[billybudd]

Does it seem that these stats are used as crutches to propel a player with less points or ppg, seem better than a player with a higher point total?

I mean, if Stamkos or Giroux have weaker advanced stats than another player, would you take the other player?

Or are these stats only suitable in discussions when considering B-tier and C-tier players?

They're just additional datapoints, which are always useful. However, they don't always mean what people think they mean because they exist without context. Raw Corsi, in particular, means almost nothing by itself when looking at an individual player and comparing him to one on another team (strong systemic component to raw Corsi).

Zone starts, to use another stat requiring context, tell about a coach's preferred usage, but not why. Henrik Sedin doesn't get all those offensive zone starts because Vigneault thinks he's reckless/risky with the puck, for example, while another highly skilled guy with high offensive zone starts (say, Mike Ribeiro) might absolutely get starts in the offensive zone for that reason..

 

[TOML]

The Rangers often don't pursue the puck hard in the defensive zone. They just clog the middle of the ice, letting the opponents cycle the puck along the perimeter while they clog the shooting lanes, to ensure the shots are either blocked or wide. It's not surprising they have a poor Corsi, it doesn't, in my opinion, indicate they are any poorer of a team, it will just take 5 years of Dan Girardi's life.

Tying lack of injuries to Corsi is a point that needs further explanation.

Lack of injuries will boost any teams' standings, regardless of Corsi. Like you said, Girardi surviving thus far is remarkable.

The other prominent high-standings, low-corsi team: Nashville

imo A NYR-Nashville finals might seal up the official return of deadpuck hockey. Cheers.


TOML

 

[Les Wynan*]

If you remove October, the Rangers' possession numbers are much improved. Still not the elite team the standings paint them as and would probably lose a playoff series to Boston, Pittsburgh or Philadelphia but better than league average for sure.

The Preds have completely lived off a hot power play and a hot goaltender. If the Preds are favorites in the first round against Detroit, Chicago or St. Louis, bet against them and win big.

 

[billybudd]

The Rangers often don't pursue the puck hard in the defensive zone. They just clog the middle of the ice, letting the opponents cycle the puck along the perimeter while they clog the shooting lanes, to ensure the shots are either blocked or wide. It's not surprising they have a poor Corsi, it doesn't, in my opinion, indicate they are any poorer of a team, it will just take 5 years of Dan Girardi's life.

Exactly. Convergence isn't necessarily a less effective thing to do than pressuring for turnovers all over the ice, but raw Corsi treats the approach as if the "pack the house" team is getting dominated, which isn't always the case.

 

[Les Wynan*]

Exactly. Convergence isn't necessarily a less effective thing to do than pressuring for turnovers all over the ice, but raw Corsi treats the approach as if the "pack the house" team is getting dominated, which isn't always the case.

NYR's scoring chances are being tracked and they're more or less in line with their Corsi differential - which again is actually pretty solid if you remove October when the Rangers were playing like an AHL team but getting bailed out by Lundqvist.

 

[Freudian]

The Rangers often don't pursue the puck hard in the defensive zone. They just clog the middle of the ice, letting the opponents cycle the puck along the perimeter while they clog the shooting lanes, to ensure the shots are either blocked or wide. It's not surprising they have a poor Corsi, it doesn't, in my opinion, indicate they are any poorer of a team, it will just take 5 years of Dan Girardi's life.

Tying lack of injuries to Corsi is a point that needs further explanation.

That's the same reason people don't think Preds are good. Rangers and Preds are all about preventing good scoring chances and allowing shots they know their goaltenders will have a good chance stopping.

Any system that treats every shot the same will be mostly blind to why these teams are succeeding. Not all shots are created equal.