|
Advanced Stats: Some Numbers Do Lie |
|
|
|
|
|
First off, I know my predecessor, Travis, was an advanced stats and analytics guru. Up to this point, in almost a month on this site, I don't think I have mentioned the words Corsi or Fenwick once, nor do I pretend to be an expert on the subject.
Quite frankly, I have been hesitant to delve into that world, because I tend to rely on my observations more than what the stats say. You can look at the stats after a game and wonder if it was for the same game you were watching. Or, by looking at the stats without having watched the game you could get an entirely different perception of how the game was actually played.
The fact is, there is someone different in every arena keeping track of the stats, and their perception is what you are relying on by looking at certain statistical categories, and that cannot possibly be consistent from one rink to the next. Giveaways, takeaways, hits, and most real time stats are subjective.
That being said, there is something to the advanced stats craze that really exploded over the past summer. If you focus solely on those, however, you aren't getting the full picture. For instance, if Paul MacLean was setting his lineup based solely on advanced stats, Bobby Ryan would never see the ice. Based on his Corsi ratings so far this season, (31.48%, -17 Rel Corsi 5 on 5), Ryan is one of the least effective players in the league, and by far the worst on the team. You would have a hard time arguing that Ryan has come even close to meeting expectations, for a variety of reasons. In that case, you could use Corsi ratings to support your argument that he hasn't been good, and that is an understatement.
On the flipside, let me ask you a question: Which Ottawa Senators player has been the most effective at 5 on 5 early in this season?
How far down your list is Milan Michalek? If you are anything like me, it probably wasn't in the top 5 or even top 10.
But, the evidence is there, and the fact is that Michalek's Corsi and relative Corsi ratings are both significantly higher than his teammates, and in fact in terms of 5 on 5 play, only Michalek and Mark Stone are on the positive side of the ledger for the Senators.
And, for what it is worth, the most effective defensemen this season in terms of relative corsi have been Jared Cowen, Mark Borowiecki and Eric Gryba. Cowen, who has the lowest Corsi % off-ice rating, has been so bad that he has been a scratch for 2 straight games, but the corsi numbers do not support that theory. We have seen it with our own eyes, but the stats don't support it.
So the advanced analytics do have their place in helping to determine the effect, but unlike baseball where "Moneyball" took the world by storm for a time (how many World Series' did the moneyball A's win again?), hockey is a fluid game where from shift to shift the unpredictable happens, momentum is a great factor and although analytics can provide nice charts, they aren't the be all, end all factor in determining a team's success.
The number crunchers predicted the downfall of the Maple Leafs last season, based primarily on their low possession numbers and negative Corsi ratings. But if Jonathan Bernier hadn't missed the last stretch of games, they would not have missed the playoffs.
Edmonton has a positive corsi rating so far this season, but with AHL calibre goaltending that doesn't mean much. Ottawa, San Jose and LA are each in the bottom 7 on terms of Corsi, but all three are in the top 7 of league standings so far.
I understand that 5 games is a relatively small sample size, and that in the end the trend usually goes toward the middle. I know some of you read Travis for his advanced stats analysis, which was really second to none. You won't get that from me on a daily basis, but every so often I will take a look at the numbers and try to get a feel for trends that are developing as the season goes.
Call me old-school, but I prefer to be able to see what is going on on the ice and forming my opinion from that. Even then, depending on your perceptions, biases and expectations you won't see the same thing I do, and that is fine, because it creates discussion and debate, which is what we are all here for. We all have our own things that we look for in the grand scheme of things, just like different coaches and general managers do.
All I know is, the Senators have relied on and gotten good goaltending so far this season, and that has put them in a good position early on to reach the goals they have set for later. If that goaltending continues, they will be fine, but if it starts to struggle, so will the Senators, whether they are positive or negative in terms of Corsi ratings.
Ottawa Senators Blogger • 
Join the Discussion:
»