Puck possession vs Shot attempts

[hutter]

http://www.inlouwetrust.com/2014/7/28/5901789/efficiency-and-transition-offense

Here's a pretty interesting article/volume. The Devils, for example, have been a high corsi but low event team for a few years now. We've basically seen the same thing out of them every season.

This guy gets into why the way the Devils generate shooting attempts may be cutting the corners and negatively effecting the offense. He tracked every Devils' and opposition pass last year.

If you follow his articles, it turns out that shots generated from neutral zone passes have a higher chance of going in, so when the Devils hold it in the offensive zone for ages (their current strategy), they're actually hurting their offense, and giving the opposition more opportunities to make transition passes up. Sounds counterintuitive, but it seems like there are ways to have possession unproductively. The more I've thought about this, their system seems to exist for a defensive purpose rather than an offensive purpose. It allows a low number of events, limiting the opposition's number of opportunities even if it increases the quality (and yes, this guy has evidence) of a lot of them.

I think he argues that it's possible to play a way that attempts a ton of possession, but doesn't necessarily correlate to winning. The LA Kings play a very similar system, so i'm not so sure about that, but maybe it'd be something along the lines of crossing a threshold of talent in order to win.

 

[TOML]

http://www.inlouwetrust.com/2014/7/28/5901789/efficiency-and-transition-offense

Here's a pretty interesting article/volume. The Devils, for example, have been a high corsi but low event team for a few years now. We've basically seen the same thing out of them every season.

This guy gets into why the way the Devils generate shooting attempts may be cutting the corners and negatively effecting the offense. He tracked every Devils' and opposition pass last year.

If you follow his articles, it turns out that shots generated from neutral zone passes have a higher chance of going in, so when the Devils hold it in the offensive zone for ages (their current strategy), they're actually hurting their offense, and giving the opposition more opportunities to make transition passes up. Sounds counterintuitive, but it seems like there are ways to have possession unproductively. The more I've thought about this, their system seems to exist for a defensive purpose rather than an offensive purpose. It allows a low number of events, limiting the opposition's number of opportunities even if it increases the quality (and yes, this guy has evidence) of a lot of them.

I think he argues that it's possible to play a way that attempts a ton of possession, but doesn't necessarily correlate to winning. The LA Kings play a very similar system, so i'm not so sure about that, but maybe it'd be something along the lines of crossing a threshold of talent in order to win.

Considering both teams have been in the finals and one has won two cups recently, perhaps they're onto something.

Since being in the finals, NJ lost their best two stars, so winning wasn't exactly in the cards, yet they still stuck with their long-proven system. Now they're are off to a hot start so far this season, so it looks like the talent they've added in the summer is helping.

I think that overall talent > systems when it comes to winning though. If you don't have the horses...

 

[The Mighty Duck Man]

Corsi and Fenwick may originally have been developed as a possession proxy, but I also see Fenwick being used as a proxy for scoring chances. Obviously, some adjustment is going to be required to compensate for teams e.g. Boston who are good at keeping shots against to the outside. And yes, shots on goal might be even better for a scoring chance proxy. (Although some will say that if you leave someone open 10 feet in front of the net and they blow it, you should still get dinged ... so there's a case that missed shots should still be counted)

I know Extra Skater last season had a stat for SOG% which revealed, for instance, that Anaheim's low Corsis were largely misleading because they did a lot of shot-blocking.



I don't think we're debating that possession time measures possession time better than Corsi, that's sort of obvious. I'm just not sure that having offensive zone possession time is the most important factor in winning hockey games - it's definitely important, but you still have to do something with the possession.

Obviously you have to do some things with possession, but that's something anti-corsi posters have argued as well. You need context, that's obvious, and the data of how many shot attempts occur isn't that IMO.

I was harsh earlier, but to me, guys saying they prefer Corsi to actual possession seems nuts. You obviously need context, but you're not getting that with Corsi. Adding those kinds of percentages probably helps a lot, but it's still not optimal IMO.

 

[The Mighty Duck Man]

Considering both teams have been in the finals and one has won two cups recently, perhaps they're onto something.

Since being in the finals, NJ lost their best two stars, so winning wasn't exactly in the cards, yet they still stuck with their long-proven system. Now they're are off to a hot start so far this season, so it looks like the talent they've added in the summer is helping.

I think that overall talent > systems when it comes to winning though. If you don't have the horses...

Absolutely. Obviously it would be great to play a puck possession game, but most teams probably can't. Same thing with the dump and chase, it's a necessity for a lot of teams.

 

[NikF]

http://www.inlouwetrust.com/2014/7/28/5901789/efficiency-and-transition-offense

Here's a pretty interesting article/volume. The Devils, for example, have been a high corsi but low event team for a few years now. We've basically seen the same thing out of them every season.

This guy gets into why the way the Devils generate shooting attempts may be cutting the corners and negatively effecting the offense. He tracked every Devils' and opposition pass last year.

If you follow his articles, it turns out that shots generated from neutral zone passes have a higher chance of going in, so when the Devils hold it in the offensive zone for ages (their current strategy), they're actually hurting their offense, and giving the opposition more opportunities to make transition passes up. Sounds counterintuitive, but it seems like there are ways to have possession unproductively. The more I've thought about this, their system seems to exist for a defensive purpose rather than an offensive purpose. It allows a low number of events, limiting the opposition's number of opportunities even if it increases the quality (and yes, this guy has evidence) of a lot of them.

I think he argues that it's possible to play a way that attempts a ton of possession, but doesn't necessarily correlate to winning. The LA Kings play a very similar system, so i'm not so sure about that, but maybe it'd be something along the lines of crossing a threshold of talent in order to win.

Textbook Terry Murray era Kings, especially the defensive purpose part. The switch to Sutter made the Kings more even-keel in that respect, but that was the foundation, not surprised since we know Lombardi has pretty strong ties to Lamoriello. Carter and Gaborik additions were both moves to address this as well. The development of Pearson and Toffoli helped too. They needed more quick-strike players who can sting with minimal possession to beat Chicago, but the foundation is still the same.

 

[Phenomenon13]

Wouldn't this be a prime example of puck possession vs shot attempts?


http://www.youtube.com/watch?v=hjY31kGVkVQ

 

[Dr Quincy]

According to some guy on NHL network, the Avs had the 6th best offensive zone time in the NHL, yet of course ranked at the opposite end of the spectrum in shots. Then they had some quotes from Matt Duchene noting that their system doesn't include chucking the puck towards the net and going to go get it after a rebound or a chip in. MacKinnon ranked 2nd in the league in offensive zone entry by carrying the puck (at around 79% of all of his zone entries) and Duchene was 3rd in the league at around 77%. This type of analysis, combined with the obvious fact that the Avs' 'Corsi' and 'Fenwick' ratings did not correlate at all with their winning percentage, leads me to believe that the fact they were such an outlier was not all about luck - and the 'abnormally high' shooting percentage (which wasn't even first in the league) was part luck, part system.

This could all be true and is worth considering, but many of these same claims were made by Leafs fans last year. It's possible that it is indeed true and that the Avs system is a Corsi-bucking style, but that remains to be seen.

 

[Lafleurs Guy]

Not necessarily. It depends on what shots you take. Avs take fewer shots from the points than other teams. Those are usually low percentage shots and furthermore much more likely to create corsi/fenwick events (miss the net or being blocked)

Shots taken by defenders out of all shots:

Avs: 26%
Kings: 29%
Blackhawks:28%
Sharks: 30%
Blues: 32%

Furthermore, some of Avs defenders don't shoot much from the point compared to other defenders. The shot charts of Nick Holden and Tyson Barrie look more like the shot charts of forwards since they jump up in the play all the time. These numbers also are in line with Duchene's comment that Avs want to be a team that hangs on to the puck instead of a team that gains the blue line and throws the puck on net.

If Avs wanted to inflate their shot attempt stats, they'd throw the puck on net from the points like other teams do.

I think the advanced stats obviously carry weight as they've been shown to be more predictive of playoff success than the standings are. It's interesting though that a team like Colorado finished as high as they did and NJ didn't make the playoffs.

To me it's a combination of two things. Goaltending (obvious variable) and scorers who can finish their chances. Colorado has several players who can finish and put the puck in the net. The system may not be good and the team may not play well overall but when you've got elite scoring talent and great goaltending you can go further than you otherwise would've been able to.

I suspect the idea that Colorado is taking time setting up better scoring opps is probably nonsense. I think they are mostly getting outplayed. But they have the goaltending and elite scorers to overcome it.

 

[nobuddy]

The data presented in the video is worthless.

All it says is how often the Avs have the puck in the offensive zone. It doesn't express it as a ratio. For it to be a meaningful counter to criticism of their CF% and FF%, it would need to compare their zone time to their opponents' zone time.

The Avs were the fastest team in the league last year. Logically, this would lead to less time spent in the neutral zone.

I also like how this is presented considering the 4 teams ahead of Colorado were LA, Chicago, Boston, and St. Louis. All of which were top 6 FF% teams.

 

[Mubiki]

I think the advanced stats obviously carry weight as they've been shown to be more predictive of playoff success than the standings are. It's interesting though that a team like Colorado finished as high as they did and NJ didn't make the playoffs.

To me it's a combination of two things. Goaltending (obvious variable) and scorers who can finish their chances. Colorado has several players who can finish and put the puck in the net. The system may not be good and the team may not play well overall but when you've got elite scoring talent and great goaltending you can go further than you otherwise would've been able to.

I suspect the idea that Colorado is taking time setting up better scoring opps is probably nonsense. I think they are mostly getting outplayed. But they have the goaltending and elite scorers to overcome it.

This. A high PDO can overcome getting outplayed, or vice versa. There are always a select few teams that are considered "corsi breakers" each year, but that's to be expected. Generally speaking, possession stats are an incredibly accurate way to predict success.

This argument always reminds me of a similar situation in football. Once every decade a team with a crap qb wins a super bowl, and tons of people convince themselves you can get by with mediocrity at the position. People hyper focus on the outlier and ignore the reality staring them in the face.

 

[OCSportsfan]

I am a big fan of statistics, and would love to have a catch all stat to say this team is better than that team, but not sure Corsi/Fenwick is the answer. Do they have a correlation with possession? Yes for some teams. Do they tell the whole story with ever team? No. and there lies the problem. Different styles of play generate different numbers of shots.

I think where much of the argument comes from, is those on each side are writing their comments as though it is 100% correct or 100% not correct.

I would say from what I have read and watched, Corsi etc. are a good predictor of future success around 60% of the time.

Basically what I mean is that barring all other factors, the team with the better Corsi will win 60% of the time on a neutral surface.

Now PDO is complete crap and shooting percentage and save percentage can be sustainable in my opinion, year over year. Much depends on the Goalie (generally goalies will be consistent year to year)and the type of system you run. Look at the Ducks, they will always be one of the higher PDO since their forwards take most of their shots and from high percentage areas.