Jul 202011
 

So I woke up this morning and started reading the usual morning news and blogs and one of the first ones I happened to read was an article at Maple Leafs Hot Stove on the Luke Schenn contract negotiations.  One comment early in the post really caught my attention.

While he may not be an offensive machine (I believe there is still some upside there), at 23, he has blossomed into one of the leagues best shutdown defensemen.

First off, Schenn is 21 not 23 but what really caught my attention is the assertion that he is one of the leagues best shutdown defensemen.  This isn’t an uncommon sentiment regarding Schenn, especially among Leaf fans.  There are a lot of people who believe Schenn has developed into a superior shut down defenseman, or at the very minimum is on track to becoming one.  The thing is, the stats don’t back that up.

2010-11 GA20 TMGA20 OppGF20 ExpGA20 GA20-ExpGA20
Tomas Kaberle 0.638 0.843 0.756 0.800 -0.162
Keith Aulie 0.745 0.845 0.780 0.813 -0.068
Dion Phaneuf 0.783 0.846 0.786 0.816 -0.033
Carl Gunnarsson 0.838 0.831 0.741 0.786 0.052
Luke Schenn 0.838 0.785 0.753 0.769 0.069
Francois Beauchemin 0.884 0.792 0.759 0.776 0.109
Mike Komisarek 0.994 0.782 0.740 0.761 0.233

The above table shows the most used defensemen for the Leafs last season.  Numbers are for 5v5 ice time.  GA20 is the players on ice goals against per 20 minutes.  TMGA20 is a defensive quality of teammates measure in goals against per 20 minutes.  OppGF20 is an offenwive quality of opponent measure in goals for per 20 minutes.  ExpGA20 is an expected goals against average based on quality of teammates and quality of opponents and is simply an average of TMGA20 and OppGF20.  The final column is GA20-ExpGA20 which tells us whether fewer or more goals were scored against than expected when the player was on the ice so negative values are better than positive values.  Note:  Kaberle and Beauchemin’s stats include their time with Boston and Anaheim respectively.

What it all means is Schenn was far from a good shutdown defenseman last year.  His defensive numbers are actually quite poor.  He didn’t play against especially tough opposition (especially compared to Phaneuf and Aulie) and had a very poor GA20.  Overall one could say he was a weak defensive defenseman.  Based on the numbers above, only Beauchemin and Komisarek were worse.  So how does he stack up against the rest of the leagues defensemen?  Lets take a look.

Season(s) GA20 Rank OppGF20 Rank HARD+ Rank CorHARD+ Rank
2010-11 0.838 122/163 0.753 94/163 0.888 121/163 0.922 142/163
2009-10 0.930 145/169 0.744 152/169 0.846 140/169 0.929 142/169
2008-09 0.971 152/159 0.774 44/159 0.818 143/159 0.870 156/159
2009-11 (2yr) 0.876 138/157 0.753 128/157 0.880 138/157 0.922 146/157
2008-11 (3yr) 0.907 161/169 0.762 123/169 0.868 154/169 0.930 160/169

The above table shows how Schenn compares to the rest of the defensemen in the league at 5v5 even strength ice time.  GA20 and OppGF20 are the same as above.  HARD+ is a composite defensive ranking that takes into account the players GF20 as well as defensive quality of teammates (TMGA20) and offensive quality of opponents (OppGA20).  CorHARD+ is similar to HARD+ but instead of using goal data to calculate it uses corsi data.  Personally I think this is somewhat meaningless but this is for those out there who put stock in corsi based stats.

As you can see, his 5v5 even strength defensive numbers range from bad to horrible, his quality of opponent is mediocre at best, his HARD+ rankings are quite poor, and his CorHARD+ rankings are even worse.  IT is really tough to find a compelling argument that Schenn is even an average defensive defenseman let alone one of the best shut down defensemen in the NHL.

His 4v5 PK numbers are even worse.  Of the 87 defensemen with 400 4v5 PK minutes over the past 3 seasons, Schenn has the highest (by a decent margin too) goals against per 20 minutes on the PK.  His 4v5 PK HARD+ rating is also the worst at 0.652 which is pretty atrocious.  His PK corsi numbers aren’t quite as bad, but are still below average (67th of 87 in corsi against per 20 minutes).

All in all it is really difficult to suggest that Schenn is even an average defensive defenseman.  He is certainly not among the leagues best shut down defenseman.

Now, with all that said, I am all for signing Schenn to a new 3 or 4 year contract in the $3-3.5M range.  He is a physical defenseman who blocks shots (as noted by the Maple Leaf Hot Stove article) and gives a good effort all the time.  It could be that he is just a little over zealous with the hits and blocked shots that he gets out of position a bit too often and just needs to learn when to take chances going for the hit or blocked shot.

Part of the problem is also that he was rushed to the NHL and put in a high pressure situation with a lot of expectations placed on him so we was never really given an opportunity to really learn the craft.  I really hope that with the addition of Liles and Franson and the further development of Aulie and Gunnarsson that some of the pressure comes off Schenn and he can take a bit of a step back and really learn the details of playing defense (maybe some new assistant coaches will help too).  Some good goaltending and team success will take some pressure off him as well.

Finally, as fans, I really hope we stop putting pressure on him to be one of the best shutdown defensemen today and one of the faces of the franchise and we begin to view him for what he really is: a young 21 year old defenseman with lots of ability and drive but with lots still to learn.  There really haven’t ever been many elite shut down defensemen at 21 years of age so we should stop expecting him to be that now and instead hope he can be that a year or three from now.

(Final note:  Strangely enough, Schenn’s on ice offensive numbers are actually reasonably good.  His 3 year on ice GF20 is 0.833 which ranks 44th of 169 defensemen and his HARO+ is 1.030 which is good for 42nd.  His 2 year numbers are even better.)

Jul 192011
 

An interesting statistical debate sprung up today started by Tom Benjamin who wrote about his skepticism of the Corsi statistic.  In it Tom comments on the fact that Ryan Kesler and Ryan Clowe ranked so highly in corsi in response to Greg Ballentine’s posts at The Puck Stops Here.

Greg’s examples, it seems to me, make a good case against the Corsi statistic. First, both the Kesler and Clowe stories tell us how much influence context has – neither Kesler nor Clowe could have done it playing on a different team or even playing in a different role on the same team. In other words, this is not really an individual statistic.

Of course, this got some in the Corsi crowd up in arms and states that Corsi can’t be used on its own without considering its context.  Gabe Desjardin’s comments on Tom’s post with the following:

Corsi, like any other statistic, needs to be understood in the context of other factors. Sneering at it because, like any other simple statistic, it doesn’t provide a unified measure of a player’s complete value doesn’t contribute anything to the larger discussion of hockey analysis.

Ok, so I am glad we have that cleared up.  Corsi is just a stat without meaning unless you consider the context.  Oh good, now it is on par with nearly every other stat in hockey.  Unfortunately people actually use corsi to actually draw conclusions about players.

Here is the thing that really irks me about some in the Corsi crowd.  They just assume that shot quality doesn’t exist.  An anonymous commenter using the name ‘Name’ writes the following:

Even the basis of corsi, that shot quality always evens out, so we just have to measure shot quantity, is inherently flawed. The strategies and styles some teams play lead to giving up a greater number of shots, but reducing quality ones, whereas some teams strive to block or prevent every single shot, no matter where it comes from. Therefore some players, just by playing on a certain time, will inherently be on the ice for more shots against. It doesn’t mean they’re giving up more quality scoring chances, or making lots of defensive mistakes, or failing to control play offensively.

To this comment Gabe comes back with his favourite response to any challenge put his way:

Name these teams and players. Thanks.

Now that is a fair response, unfortunately he ignores anyone who actually names these players.  The reality is shot quality doesn’t even out.  Some players drive shot quality and some players suppress it.  Some players have a significantly different +/- than corsi +/- year in and year out.  I gave an example the other day in Brendan Morrison.  Here are some other names for Gabe to consider.

Some guys who can drive shooting percentage: Sidney Crosby, Marian Gaborik, Nathan Horton, Bobby Ryan, Martin St. Louis.   Henrik and Daniel Sedin.  Alex Tanguay.  Jason Spezza.

Some guys who can suppress shooting percentage:  Marco Sturm, Travis Moen, Tyler Kennedy, Taylor Pyatt, Shawn Thornton, Chris Drury, Jeff Carter. Torrey Mitchell. Kamil Kreps.

No one in the second group had an opposition shooting percentage above 7.6% in any single season over the past 4 years.  Only a handful of times over the past 4 years has any of the players in the first list had an on ice shooting percentage below 9% and only Bobby Ryan’s 23 game 2007-08 season was below 7.6%.

Now, you’ll probably notice that the first group are all first line players who are expected to produce offense while the second group are mostly third line players asked to shut down the opposition.  It’s difficult to suggest it was just luck that this is how things panned out.  No, some combination of talent and style of play will affect your on ice shooting and opposition shooting percentages.  And again it needs to be stated that shooting percentage is much more highly correlated with scoring goals than corsi.

Jeff Carter is an especially interesting case in that you can in no way argue that he has played in front of especially good goaltending that would drive down his shooting percentage against and yet he has a really slow shooting percentage and yet has one of the highest corsi against of any forward over the past 4 seasons (20.1 corsi events against per 20 minutes over the past 4 seasons ranks 300 of 310) but his goals against (0.753 per 20 minutes)  is decidedly average or even slightly better than average (ranks 139 of 310).

So Gabe, those are some players for you to consider.  I look forward to your response.

Jul 162011
 

Last night after news came out that Brendan Morrison had re-signed with the Calgary Flames, Kent Wilson tweeted the following:

Morrison back in Calgary. Check out his corsi tied rating fellow stats nerds: http://bit.ly/q1ywUj

The link is to the Calgary Flames 5v5 game tied corsi ratings which show Morrison had a 0.452 corsi rating (Corsi For %) which was dead last on the Flames.  The problem with jumping to the conclusion that Morrison is bad is two fold:

1.  Corsi generally speaking isn’t good at evaluating players.

2.  One year of 5v5 game tied data is not enough to evaluate players, even with corsi.

Lets take a look at Brendan Morrison over the past 4 years and I’ll show you exactly what I mean.  First lets look just at 5v5 any game score situations.

Season(s) CorF% GF%
2010-11 0.484 0.562
2009-10 0.514 0.627
2008-09 0.498 0.569
2007-08 0.430 0.500
2007-11 (4yr) 0.491 0.577

In each and every year the goals for percentage is significantly higher than his corsi for percentage.  His corsi ratings make Morrison look mediocre at best but his goal ratings make him appear to be quite good.  This isn’t a fluke.  It is occurring systematically, every single season, over 4 seasons in which Morrison played for 5 different teams (Vancouver, Anaheim, Dallas, Washington, Calgary).

Now what about 5v5 game tied situations.  Morrison’s 4 year game tied corsi for percentage is 0.482, his 4 year game tied goal for percentage is 0.592 (which ranks 28th of  217 among forwards with at least 1000 5v5 game tied minutes over the past 4 seasons).

Personally, I’d rather have good goal ratings than good corsi ratings.  Morrison is a good signing by the Flames.

Jul 132011
 

Yesterday I described my player analysis method and used Brad Richards as an example.  Over the next little while I’ll apply my analysis method to a number of players so if there are any players you are interested in seeing my analysis for let me know.  First up is Tim Connolly.  The Leafs lost out on the Brad Richards sweepstakes so lets take a look at how Tim Connolly stacks up.

Let’s start off with a table of what I consider Tim Connolly’s most pertinent information – his 5v5 HARO+ (offense), HARD+ (defense) and HART+ (overall) ratings over the years.

Season(s) HARO+ HARO+ Rank HARD+ HARD+ Rank HART+ HART+ Rank
2007-11 (4yr) 1.171 18/310 0.985 152/310 1.078 31/152
2008-11 (3yr) 1.242 23/319 0.980 158/319 1.111 36/319
2009-11 (2yr) 1.169 67/319 0.975 170/319 1.072 85/319
2010-11 1.045 156/336 0.856 268/336 0.951 220/336
2009-10 1.289 32/338 1.082 100/338 1.185 34/338
2008-09 1.615 2/335 0.941 187/335 1.278 16/335
2007-08 1.322 38/328 0.974 159/328 1.148 55/328

Generally speaking Connolly’s offensive rankings have been well over 1.00 and ranking very highly among all forwards with at minimum 500 minutes of 5v5 time per season and his defensive rankings have been middle of the pack.

Based on Connolly’s offensive statistics he is legitimately a first line center though he has played against relatively weak defensive competition (232/310 in 4 yr OppGA20) as he has played behind Derek Roy in Buffalo.  Last year he played against somewhat tougher defensive competition than he did in 2008-09 and 2009-10 as Derek Roy was injured for more than half the season and he had his worst offensive (and defensive) season so that should be a bit of a concern for Leaf fans.  Still, one season is too short to draw any conclusions so it could just be an anomaly as well but it is something to watch for next season as he’ll likely be given top line duty in Toronto with Phil Kessel and Joffrey Lupul and play against the oppositions better defensive players.

Of interest to Leaf fans who have suffered through several years of poor PP and PK play is Connolly’s special team numbers.  Over the past 4 seasons Connolly has been played a significant role on Buffalo’s power play and the results have generally been good (his 4 year 5v4 HARO+ rating is 1.169).  Connolly has also played a fair amount (about 100 min/season) on the Buffalo PK unit and his performance has been better than what one would expect from his 5v5 defensive numbers.  His 4-year 4v5 PK HARD+ rating is a more than respectable 1.196 so maybe he can play defense when is he trying to stop the opposition from scoring as opposed to trying to produce offense himself.

Based purely on his performance over the past 4 seasons it seems Connolly is a more than reasonable gamble as one could argue he has legitimate first line offensive capabilities and is at least middle of the pack defensively.  The big question of course with Connolly is his health.  Has has played just 48, 48, 73 and 68 games over the past 4 seasons.  The good news is he hasn’t had a significant concussion in several years and his injuries over the past couple of seasons have been non-serious in nature.  If he can be healthy enough to play 70+ games I think a year from now we could look back and say that Connolly was one of the better free agent signings of the 2011 off season, even with a $4.75M cap hit.

Jul 122011
 

Over the past couple of weeks I have had several comment discussions regarding some of my recent posts on player evaluation and Norris and Hart trophy candidates which centered around which is a better method for evaluating players:  corsi vs goal based evaluation.  A lot of people, maybe the majority of those within the advanced hockey stat community, seem to prefer corsi based analysis while I prefer goal based analysis and I hope to explain why with this post.  I have explained much of this previously but hopefully this post will put it all into one simple easy to understand package.

There are two main objectives for a player when the coach puts him on the ice:  1.  Help his team score a goal.  2.  Help his team stop the opposing team from scoring a goal.  Depending on the situation and the player the coach may prioritize one of those over the other.  For example, a defensive player may be tasked primarily with shutting down an opposing teams offensive players and scoring a goal is really a very minor objective.  Late in a game when a team is down a goal the opposite is true and the primary objective, if not sole objective, is to score a goal.

I think we can all agree on the previous paragraph.  Goals are what matter in hockey so right there we have the #1 reason why goals should be used in player evaluation.  The problem is, goals are a relatively rare event and thus ‘luck’ can have a serious impact on our player analysis results due to the small sample size that goals provide.  This brought on the concept of corsi which is nothing more than shot attempts and is used as a proxy for scoring chances.  The benefit of corsi is that shot attempts occur about 10 times often as goals which gives us a larger sample size to evaluate players.

Continue reading »