A novel points system proposed by a biostatistician:
3 points for an overtime win.
0 points for an overtime loss.
Shootouts would retain the current points system.
In an article recently published in the Journal of Sports Economics, I examined the probabilities of overtime in the NHL?s three most recent point systems (pre-1999, 1999-00 to 2003-04 and 2005-06 to 2011-12) and found several statistically significant differences. First, overtime likelihood rose dramatically after a 1999 rule change guaranteed the overtime loser a point. Second, in the current point system, overtime likelihood has been dramatically higher in March and April, when teams push for the playoffs.
Lastly, teams have been playing overtime games at a significantly higher rate against non-conference opponents, with that effect only noticeable in games played since 2005. In fact, over the past two years, non-conference games went to overtime 23 percent more often than conference ones, with several teams appearing to have recognized the benefits to playing for overtime in these contests.
With next season?s realignment plan, teams are slated to play 28 or 32 non-conference games, substantially higher than the 18 non-conference games played in 2011-12. Further, while playoff eligibility is currently based on conference standings, future standards will have teams competing primarily against their divisional opponents for post-season spots. Overtime incentives currently highest in non-conference games will now extend to all non-divisional contents.
Proponents of the current structure suggest it keeps playoff races tighter, but alas this parity is artificial. As Bruins forward Jay Pandolfo correctly points out: ?It?s difficult to catch teams that are losing but still gaining a point.
http://www.thehockeynews.com/articles/52635-NHL-in-need-of-new-threepoint-system.html