The NHL’s point system, 2013-14 edition

While I didn’t grow up a huge hockey fan, one of the passions I’ve recently taken on deals with the NHL’s point system.

It’s one of the least efficient policies in sports.

Teams get two points for a win, and one point for any loss which occurs in overtime or a shootout. What bothers me isn’t the increased percentages of overtime in recent years – yes, NHL games are going to OT more often than ever before (25%), but fans and league officials love shootouts – but that certain teams have found ways to manipulate the system by picking and choosing when they play for OT.

In some research done in the past few years, I found that a handful of franchises have played less aggressively, as judged by recording fewer shots & goals, when playing non-conference opponents as opposed to within conference ones. This stems from former league standards for postseason qualification, which dictated that teams with the highest point totals relative to their own conference would make the postseason. As a result, teams non-conference contests had elevated incentives to play overtime, because ceding a point to the opposition would not impact playoff positioning.

Over the last two full NHL seasons, as an example, non-conference games went to overtime 23 percent more often than conference ones. And Colorado and Pittsburgh were the worst offenders; non-conference games including either of these two franchises were more than twice as likely as conference games to reach OT over this same time period.

In the league’s realignment for 2013-14, two changes are set to increase the effect of the point system on the NHL’s standings. First, teams are slated to play either 28 or 32 non-conference games, higher than the 18 such games played in the former point system. Further, because playoff positioning under realignment is primarily driven by divisional standings, there are higher incentives for teams to play OT games in their within-conference, non-divisional games.

This winter, I’ll be monitoring teams’ performances through (roughly) bi-weekly posts. I hypothesize that teams, as they have done over the past several years, will play their highest percentages of OT games against non-conference opponents, with within-conference non-divisional contests also going to OT more often than divisional ones. Once there have been more games, I’ll also identify which teams are changing their behavior most often.

For now, here are the results from the first 9 days of the NHL season. We’re off and running: 31% of non-conference games have gone to overtime!

Game Type Non-OT OT % OT
Divisional 15 5 0.250
Non-divisional, Conference 8 3 0.273
Non-conference 11 5 0.313

Lastly, I suspect its possible to exploit team behavioral changes in gambling markets. For example, most websites feature “Live Betting” odds which update during the course of games. My advice? Take the under in non-conference games which are tied in the third period. Teams (mostly) don’t try and score in this situation.

Advertisements

4 Comments

  1. “My advice? Take the under in non-conference games which are tied in the third period. Teams (mostly) don’t try and score in this situation.”

    Did you really think sportsbooks offered in-game, third-period total bets? And if they did, what makes you think the market is not already aware of situational aggressiveness trends in hockey?

  2. evo34, sportsbooks definitely offer live-betting for the totals of each game.

    My guess is that sportsbooks are aware that aggressiveness changes late in games, but I’m not sure they are aware that it depends on the types of teams and the season. It’s something I haven’t followed, so I don’t know for sure. As a result, I’d agree that my wording in this post (from last year) was a bit optimistic and strong.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s