Tonight I had a fairly circular twitter chat about the relevance of statistics to a team’s performance. Of course we were talking past each other, since he is a math fan and the church of math would probably burn me at the stake if it had the emotional wherewithal.
Anyway, it made me think more about the ways we assess a given game or streak or season or player. My problem with statistical analyses isn’t that they are innacurate but that they are misleading in their predictions.
Math shows how things equalize, stats give us averages. I call this flatline logic because it seeks a straight line, like a flatline on a heart monitor. It is a kind of death to me, but not a little one.
Life would be boring if it were truly predictable in large segments. But it is comforting that you can know roughly when the seasons will turn, or if a storm is brewing. Stats can give you a general idea when to expect these things, but observation, as by doplar radar, will be much more precise than an almanac.
The same is true of people and, since they are people, hockey players. Simple observation, even as little as we see on tv, can tell us if they will play a good game or not. There is no stat tracker for body language, and yet any hockey fan can tell when a team has given up. We can see it in their stride, we see it in their hesitation. We may see it at the start of the third, only down by one goal. Stats will say one thing about the odds of tying the game, but how we see them playing will tell us more.
One player’s signals won’t tell you how the game will turn out, but it is a better indicator of that player’s impending performance than his statistical mean is. It will tell you if his performance is likely to be above or below that flat line. Do the same for 19 more guys and you can get a pretty good picture of what the team is about to do in this game.
That measurement is all about the living, the now, the mental, the emotional. It is not about seeking the middle, it is always struggling to rise.
How they play the game tonight will also tell you more about how they will play the next game than the stats for the last 20 games. Will it tell you how they’ll play the next 20? No, but neither will stats.
Take save percentages and shot quality. Apparently, after 500 shots, that equalizes or something. That’s nice, for fans of the flatline. But even Michael Leighton has never faced 500 shots in one game in any league, so I deem this statistic irrelevant. It’s about wins, right? So unless facing 500 shots can determine the outcome of this game, it doesn’t matter.
But it’s more than that. I don’t really need to predict the outcome of a game. So I don’t care how good or bad stats are at that.
I guess what pushes my buttons about flatline logic is that it strives to diminish the individual, the moment, it snuffs out what little free will I believe we have. It tells us simply that all is predetermined, no matter how we try or work, we will never achieve or be more than we are. It champions the idea that most of what we do will always trump the best we ever do.
This is heretical to my world view. It is Logan’s Run, without Logan running. Statisticians claim to be above emotional influence. This is intriguing to me, because of how utterly depressing I would find it to believe mean statistics were more relevant than present observation of the human variety. Run, Logan, run.