Thursday, April 19, 2012

Do professional athletes get tired?

Coach Tortorella clearly feels that they don't. But I wonder if they might. I'm pretty sure that at the end of an NHL game, especially a playoff game, the players are a little exhausted. Exhausted enough to affect performance? Maybe.

As you know, two nights ago, the Rangers fell to the Senators, 3-2 in overtime, to even the series at 2 games apiece. As you probably also heard, the loss was the Rangers' seventh consecutive Playoff OT loss - you have to go back to April 29, 2007 (Game 3 of the Second Round, against Buffalo) for the last time the Rangers won a Playoff game in OT. This is, of course, a very small sample size spread over a number of years, and therefore meaningless. But I do wonder if John Tortorella's Rangers, specifically, might be less geared toward winning an extended game, because the coach tires his players out so much.

In the Tortorella era, the Rangers are 3-2 in Playoff games that end within 60 minutes and 0-4 in playoff games that are tied after 60. Instead of doing my job this morning, I made a spreadsheet to take a look at variation in ice time among the Rangers and their opponents in those four overtime losses. Torts certainly seems to lean much more heavily on certain guys, while barely playing others. Last night, for example, Mike Rupp only got 4:12, John Mitchell 3:59, Stu Bickel 3:33, and Chris Krieder 3:29. That's four guys that added up to only 15:12 of ice time. In a regulation game without any penalties, that leaves everyone else averaging over 20 minutes. As an overtime game waxes on, that certainly might give an advantage to the more rested team.

For each of the four OT playoff losses of Torts's tenure, I found the average ice time (split up by team and position, of course). For each skater, I looked at how far his individual ice time deviated from the average. I then averaged all those deviations, to get a number that kind of represents how "spread out" all the ice time was. For example: if 6 defensemen each played exactly 20 minutes, their average ice time would be 20 minutes, and their average deviation would be 0. If 3 of them each played 30 minutes and 3 of them each played 10 minutes, their average ice time would still be 20 minutes, but their average deviation would be 10 minutes, because each of them was 10 minutes away from the average ice time. See?

So this measurement gives us a mediocre idea of how unbalanced the team's ice time was. The higher the average deviation, the more that some people played too much while others didn't play enough. Here's what I found:

Against Washington last season, Torts didn't actually vary his forwards much less than Bruce Boudreau varied his. In Game 1, Ranger forwards varied by an average of 4:51, and Caps forwards varied by an average of 5:03. In Game 4, Rangers varied by 6:06, and Caps by 5:42. Not a huge difference. The blue line, though, was a different story. In Game 1, Ranger D-men varied by an average of 5:15, while Caps D-men only varied by 2:35 (less than half). In Game 4, the story was the same: Ranger defensemen varied by 6:58, and Caps only varied by 3:24 (again, less than half).

The series against Ottawa so far tells a similar story, although there's a disparity in forwards as well. In Game 2, Ranger forwards (discounting the wrongly booted Brandon Dubinsky) varied by 5:33, while Senator forwards varied by only 3:46. In Game 4, Rangers varied by 5:36, and Senators by only 2:53. Meanwhile, the defensive disparity stayed just as strong: in Game 2, Rangers varied by 7:06, Senators (discounting the rightly booted Matt Carkner) by just 3:11. In Game 4, Rangers varied by 6:48, Senators by 4:18.

So, on average, across these four games, Ranger ice time on defense varied almost twice as much (1.94x) as their opponents'. In the current series, even Ranger forwards varied their ice time 68% more than their counterparts from Ottawa. Could this have led to more fatigue, and therefore less of a chance to win? Well, it certainly didn't help. This may be too small a sample size to be meaningful at all, but hey, that's what happens in the playoffs.


  1. I'm sensing a common thread throughout your last 3 blogs...

  2. Jeremy Roenick! I always figured you were a reader, glad to see some proof! By the way, I LOVED your performance of Funky Cold Medina two years ago - really, A+ stuff.

    But seriously, what's the common thread I'm missing? The last two entries were about the lack of reasonable discipline in the NHL, this one is about math that no one cares about.