On the theories about the Leafs' possession

[mr grieves]

Well, at least one of them. I've heard it said that Corsi numbers don't work with the Leafs because they're not just turning the puck back over with bad shots, waiting for better quality shots -- and hence the high shooting percentage. Corsi can't catch stuff like that, and so there's no proxy for measuring a team's having possession. Someone will just have to sit down and watch the games with a stopwatch.

Well, someone at PPP did, for half the 2013 season, and his numbers suggested he didn't need to watch much further. Read on!

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

Here's the take away with respect to the Leafs play last season:

I think it's safe to say at this point that Carlyle & Co. haven't found some new, ingenious high-quality-shots-only system [...] The Leafs' success last year was akin to dropping into the fetal position in a fight, and then when the bully walks away, kicking him in the balls with his back turned.

Click to read more...

 

[Purity*]

Our shooting percentage was high because we have an offense that scores A LOT off the rush as opposed to a long-possession set-up and score plays.. We had inexperienced and generally bad defensemen with Kostka and Holzer who couldn't clear the zone to save their life, couple that with how our defense was already weak in that regard and there you have it..

 

[TheProspector]

I can solve this one for you: the Leafs are a rush team.

</thread>

 

[Ace88*]

One of the times that advanced stats are useless.

 

[leafstilldeath*]

Our shooting percentage was high because we have an offense that scores A LOT off the rush as opposed to a long-possession set-up and score plays.. We had inexperienced and generally bad defensemen with Kostka and Holzer who couldn't clear the zone to save their life, couple that with how our defense was already weak in that regard and there you have it..

I can solve this one for you: the Leafs are a rush team.

</thread>

What these guys said. Also, Corsi is USELESS by itself anyway.

 

[Paris in Flames]

Remember when we used to outshoot teams significantly and finished bottom of the league?

Yeah. I much prefer this.

 

[My Sweet Shadow]

What these guys said. Also, Corsi is USELESS by itself anyway.

It's like +/-; helpful when looking at the big picture with other statistics, but pretty much useless on its own.

 

[zeke]

The Leafs last year didn't score any more than they should have been expected to score based on career track records.

Anyone who thinks that all the Leafs' skill players will suddenly forget how to score this year is letting stats - which they don't even really understand - blind them.

 

[Erndog]

Basically every single "new shiny statistic" has us being terrible and unlikely to repeat any of the success we did have last year (in terms of win/losses, playoffs, etc).

I'm quite interested to see how this plays out to be honest.

 

[Grant]

One thing I noticed towards the end of the season as well was our shots largely followed the score. We would try and protect leads a lot more than other teams would. Plan on looking at it more indepth this year on a game by game basis, was too much to look at all at once when I noticed

Our record last year when we outshot the opponent was 5-7-0
Our record last year when we were outshot was 21-10-5

And my guess for why we were outshot so much more often is because we heavily protected leads later in the game. And we often were protecting leads because we outscored opponents 46-28 in the first period and 50-49 in the 2nd. Result is that we were often defending leads in the 3rd period getting few shots while the other team was throwing everything at the net causing us to be outshot.

Will be interesting to see this regular season if the same kind of thing happens.

 

[Mimico]

I love how weak the correlation is between Corsi and actual team success. I understand that it needs to be used in context but so does +/- and we all know how much that tells us about a player.

 

[Rogie]

I thought that Chicago, LA, Boston, who all had really good Corsi For Percentages, having won the Stanley Cups over the last 3 years means there is a good correlation; a lot of lower ranked Corsi teams, using Corsi For Percentage, finished out of the playoffs and many of them near the bottom of the standings also supporting a strong correlation between Corsi and success.

 

[mr grieves]

So... the majority of you didn't read the article.

Corsi, Fenwick, fancy stats are pretty good measures of and proxy stats for actual time spent with the puck in the opposition's zone.

 

[Leafidelity]

So... the majority of you didn't read the article.

Corsi, Fenwick, fancy stats are pretty good measures of and proxy stats for actual time spent with the puck in the opposition's zone.

The majority of people don't bother with PPP because of how rigid their opinions are, yes.

 

[lLeafer]

So I guess this explains why the leafs were terrible last year and finished last place. Thank god these stats can let me know what really happened.

 

[Rogie]

One thing I noticed towards the end of the season as well was our shots largely followed the score. We would try and protect leads a lot more than other teams would. Plan on looking at it more indepth this year on a game by game basis, was too much to look at all at once when I noticed

Our record last year when we outshot the opponent was 5-7-0
Our record last year when we were outshot was 21-10-5

And my guess for why we were outshot so much more often is because we heavily protected leads later in the game. And we often were protecting leads because we outscored opponents 46-28 in the first period and 50-49 in the 2nd. Result is that we were often defending leads in the 3rd period getting few shots while the other team was throwing everything at the net causing us to be outshot.

Will be interesting to see this regular season if the same kind of thing happens.

Interesting stuff and it certainly has some bearing, but, isn't it all relative as most teams including Chi, Boston, and La, there CF% all drop quite a bit when they are ahead as well.

The Leafs CF% for 5 on 5 close is 43.6% and 29th overall, versus their 44.1% CF% for 5 on 5 which ranks last, not very much difference.

 

[Badger Mayhew*]

There are a lot of misconceptions when it comes to Corsi.

It absolutely does do a good job of measuring possession. I think the correlation is somewhere around .9

However, Corsi% only has around a .55 correlation to Win%. This is not strong. Corsi is just one piece of the puzzle when predicting team performance.

Also, five and a half years of data is still a relatively small sample size.

 

[zeke]

However, Corsi% only has around a .55 correlation to Win%. This is not strong. ...

Also, five and a half years of data is still a relatively small sample size.

yep and yep.

 

[Cor]

Or possession is ****ed, and people need to watch the game instead of trying to break it down.

There is no formula to building a winning team. Hockey is science in many ways. Stats and Building teams? Not one of them

 

[Da Murf]

I've spent a fair amount of time discussing the Stats-Guys and the corsi relationship and how the Leafs could be looking at stats differently.

1. Corsi is a great measure of how hard the goalie has to work. It aligns quite well with possession and time on attack. However one of the things I've heard alluded to by those who are former coaches ( like Noodles) is that those numbers are inflated by or deflated by the number of rebounds much like save %. So while Reimer had to work really hard to hold the fort last year, he did the job hence the ability to allow the Leafs to play that strategy.

2. What if you looked at each time a puck was shot as a turnover ( or at least the ones where you didn't recover the rebound or win the faceoff). The Leafs don't shot the puck often (aside from Kessel)as they appear to take that approach. So when the do shoot they attempt to maximize shot quality to reduce 'turnovers'.

Just my nickels worth but next game you watch a leafs game closely look at how many times they end a possession with a weak shot to get a change vs the amount of times you see a flip from the wall or point that results in a D zone start.

 

[Blufreezy]

Can't people just watch the ****ing game anymore?

Drop the damn notepads jotting down numbers during breaks in play and start screaming at your TV and banging your knees with your fists like a normal person

 

[bruyns]

OP if you were expecting posters to actually read the article and engage in a meaningful debate you will probably be let down.

People will just claim stats are wrong and meaningless and the Leafs are amazing and will be a playoff team and the numbers don't matter.

I too hope for Leafs' success, but come on people have a shred of objectivity and just consider that maybe the Leafs exceeded expectations due to great goaltending and a high shooting % that may or may not be sustainable. If we are a great team again then WOOHOO!! go Leafs! but if regression happens maybe there is merit in examining articles like the OPs and not getting too worked up and excited over this team.

It's much easier to have modest expectations and be pleasantly surprised by success than to put the team on a pedestal believing they can do no wrong and all the objective stats that suggest the Leafs played over their heads are garbage and have no basis in reality. You can disagree with the article or the value of advanced stats, but if you don't put in the time to read and understand things I don't believe you have the right to dismiss it as incorrect.

Be open minded and cheer for the team you love! Completely ignoring presented facts just makes posters come across as stubborn, ignorant, close minded and incapable of critical thinking.

 

[BoredBrandonPridham]

This report is bogus. Kicking your attacker in the balls with his back turned works every time.

 

[diceman934]

One of the times that advanced stats are useless.

Correct that....most of the time.

Stats can never measure the play in hockey as it can never measure systems that teams play, quality of shots, score opportunities.

 

[mr grieves]

One thing I noticed towards the end of the season as well was our shots largely followed the score. We would try and protect leads a lot more than other teams would. Plan on looking at it more indepth this year on a game by game basis, was too much to look at all at once when I noticed

Our record last year when we outshot the opponent was 5-7-0
Our record last year when we were outshot was 21-10-5

And my guess for why we were outshot so much more often is because we heavily protected leads later in the game. And we often were protecting leads because we outscored opponents 46-28 in the first period and 50-49 in the 2nd. Result is that we were often defending leads in the 3rd period getting few shots while the other team was throwing everything at the net causing us to be outshot.

Will be interesting to see this regular season if the same kind of thing happens.

That seems to be true. Someone did an analysis of this, looking specifically at Grabo's, Phaneuf's, and Kessel's numbers. Across the board, when the Leafs had a lead -- which they did last season a lot, and early -- they took far fewer and gave up a lot more shots than in the past.

I guess the question is this a good thing? We saw the team do that for the last 5 minutes of game 7, and, to me, that was, among other things, a spectacular flaming out of this theory of how to win hockey games.