It seems that the Philadelphia media is finally getting with the times regarding new advances in hockey analysis, but their new found reliance on advanced metrics in mainstream writing leaves much to be desired.
We don’t need to go into a whole history of mainstream writers’ feelings towards advanced statistics, but the long and short of it is that there has always been a palpable resistance from the press box types to utilize them in their writing.
With this past summer ushering statistics based writers into NHL positions — including our own Eric T. — it seems that the beat writers in Philadelphia are finally catching on to the usefulness of advanced stats to inform their readers.
Well, David Isaac of the Courier-Post seemed (sort of) up to the task, posting a piece highlighting the Flyers’ focus on puck possession. The main takeaway for Isaac? Well, while sustained puck possession may not be realized by the entire team, the top line is excelling in this category. Take this snippet, for example:
The relative corsi for percentage statistic measures puck possession in the amount of shots taken while a player is on the ice against the shots taken while he isn’t. Giroux, Voracek and Simmonds are all on career highs in that category at 11.0 percent, 9.3 percent and 8.6 percent respectively.
Now, there are just a few issues with Isaac’s conclusion here, which we’ll address point by point. First off, four of us at Broad Street Hockey were unable to replicate the statistics Isaac presented here. We scoured the bevy of newly released stats sites, searching among all situations (even strength, powerplay,etc.), and simply could not replicate the numbers.**
But more importantly, the quote above from Isaac displays a fundamental misunderstanding of possession statistics, and more importantly, relative corsi. Sure, Giroux, Voracek, and Simmonds may be on their way to career highs in this stat, but it does not necessarily have anything to do with their ability to possess the puck more frequently than in previous seasons. By the very nature of the statistic, it’s entirely possible that the supporting cast of these three players got significantly worse, providing a bit of a boost to their own metrics. Isaac never explores this possibility.
Hell, there are guys on the Buffalo Sabres with positive relative corsi ratings, but we wouldn’t necessarily call them possession players. Right?
The point here is that while it is welcome to see a guy like Isaac try to use stats that he and his mainstream brethren have been (up to now) so opposed to in the past, there’s just a bit of necessary homework that has to be done to make it comprehensible.
It’s a move in the right direction, for sure, but let’s hope Isaac and his more antediluvian colleagues get a little more education on the matter before going forward.
**Update: It appears Isaac got his numbers from Hockey-Reference.com. I apologize for the error.
