Hockeymetrics, a prelude...

Baseball is now a full generation into what is now known as Sabermetrics. Sabermetrics has received a series of promotional bumps over the years, starting with the Bill James Abstracts of the 1980's; the publishing of Moneyball, about the Oakland A's under GM Billy Beane; and most recently, the World Series title won by the Red Sox, whose GM, Theo Epstein. Epstein ended the Curse of the Bambino as the youngest GM in history and as someone who never played the game beyond high school. Epstein is, no surprise, a student of sabermetrics. Several years ago I co-authored a column on SportsIllustrated.com about hockey stats. More specifically, I explored doing things with hockey stats that no one had done before. With the column, I wanted to do for hockey what James and other had done for baseball. Surprisingly, several sabermetric methods translate very well from baseball to hockey. I say surprisingly because baseball is very much an analog sport, whereas hockey is digital. To further explain, baseball is a series of distinct events, each one initialized by the pitcher hurling the ball to the catcher. From there, a myriad of outcomes and resultant sub-outcomes is possible, but at the end of the play we return to the pitcher, on the mound, ready to toss that next pitch. Hockey, in contrast, is a constant stream of seemingly randomized and chaotic events and outcomes. I wrote the column for several months, but was disappointed by the lack of response. Perhaps it was because I was writing for SI.com and not THN, and I wasn't connecting with readers. The one thing that did come out of it was I was offered a book deal with Brassey's Publishing (now Potomac Books) to compile a hockey version of the Baseball Prospectus. Unfortunately, the money offered barely covered the cost to acquire the NHL's RTSS database, and when split out to accommodate the half-dozen writers I assembled, just wasn't worth the effort, so I junked the project. In the years since, I've allowed my methods to age. The statistical science has a name, Hockeymetrics, unoriginal as it is, and there are perhaps a dozen or so people around North America that dabble in it. I've kicked around some of my ideas with some coaches I know, and most have been interested in my concepts. I've also been told that former Maine Black Bears coach Shawn Walsh was a bit of a stat-geek, so maybe there is room in this game for some of my methods, if not my madness. I figure now is as good of a time as any to start discussing these things. Anyone who wants to get a head start can review my columns from an index which can be found here. As always, comments are welcome. For those of you whose brains turn to mush during these types of discussions, rest assured I will keep up with my usual blogging. Think of the stats discussion as a bonus, not a new direction.