By the time I got to hockey, the system was pretty well worked out. Structurally similar to soccer, but with a much more extensive statistical record, it was a fairly simple matter to adapt the same formulas.
For the most part, hockey players don’t rate quite as high as other sports, with only Wayne Gretzky getting over the 30-point mark (I have some hope for Sydney Crosby and Alexander Ovechkin). I attribute this to the huge rosters hockey teams use, coupled with the practice of constantly shuttling players in and out of the game at intervals of less than a minute. The end result is that even star players are on the ice for no more than a third of the game. It’s harder to dominate in those circumstances.
Offense 1 was simply goals scored divided by the league average. Offense 2, based on assists, was a bit trickier. Assists in hockey, as in basketball, have been defined differently through years. Unlike basketball, more than one player can assist the same goal, which means assist totals leaguewide are significantly higher than goals scored. Lastly, also like basketball, certain official scorers are more generous when deciding whether a particular star player has earned an assist. To account for all of this, the value of an assist is linked to the team’s assist-to-goals ratio.
On the other side was the usual question of divvying up credit for team defense. We don’t have individual ice time before the 1998-99 season, but we do have penalty minutes, which are a useful, if imperfect proxy for how involved a player is on defense. The obvious problem is fighting, which leads certain players (Maurice Richard, say) to have a outsized share of the team penalty minutes.
To make up for this, Defense 1 is a control, based on a theoretical maximum ice time by position — 20 minutes a game for forwards, 30 for defensemen. Defense 2 is divides team goals prevented according to a player’s percentage of team penalty minutes. These are averaged as usual, except that Defense 1 also acts a maximum, to keep the pugnacious types under control.
For goalies, it’s entirely different. Defense 1 is goals prevented, and Defense 2 is a bonus based on how much better or worse a goalie was than the other goalies on his team.
|Centers||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
A few players have higher offensive averages than Gretzky, but only with substantially shorter careers, usually in the 1920s and ’30s. Nobody with more than 1,500 games is even close.
Lemieux is a fascinating what-if, because of all his health problems. For the most part, it’s almost impossible to keep up that kind offensive average. Too bad he didn’t get the chance.
For the modern players, 160-game values decline rather dramatically. There’s only handful of real, top-tier stars. I think this is due to the talent pool in hockey being almost entirely limited to Canada and a few Eastern European nations.
|Left Wing||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
The decline is even more dramatic when we come to the left wingers. After Bobby Hull, we have a couple of guys that played in the ’20s and then free-fall. What we see over and over is that even great goal-scorers decline in the second half of their careers. Centers at least have assists, which are much more stable, as a cushion. Wingers have it tougher.
|Right Wing||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
Things are a bit better at right wing; you go legitimately five-deep here. I don’t want to sound like I’m down on hockey. My point is that it is sustained by a relatively small number of big stars and it is more dependent on rank-and-file players to provide the day-to-day excitement.
I toyed with the idea of not including Gordie Howe’s return from retirement in the WHL, but he actually comes out a bit ahead in the end.
I think Jagr is finally retired. I may still have to update him.
|Defensemen||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
Talk about what-ifs … Bobby Orr was really on his way to something amazing.
I often hear that the Golden Age was dominated by defensemen. But by my math, Eddie Shore and King Clancy didn’t have a patch on contemporary forwards. Unless crippling opponents is an admirable skill …
|Goalies||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
Terry Sawchuck is probably the biggest surprise. He was great for his first five years, but then declined after that, mixing in the occasional season at his old level with a lot of mediocrity.
The tiny offensive numbers come from occasional length-of-the-ice assits.
|Soviet Players||Games||Total||Per 160||Sqr Root||Sum||Off||Def|
Communist sports leagues are a bit sketchy, due to state interference, but I thought it was worth it to include these guys. I didn’t count international games because the quality of competition is all over the place. I tossed around the idea to include post-professional games, but then the NHL boycotted the 2018 Olympics and I decided it wasn’t worth it.
If anyone has detailed numbers for Anatoli Firsov, I’d love to see them.
To do list: