There are many ways to achieve victory in the NHL. Teams can win in regulation, overtime or a shootout and earn two points in the standings regardless of how victory is secured. Even when teams lose, they sometimes win—to an extent—by getting one point for falling in overtime or a shootout.
Indeed, how a team loses is more significant than how it wins, especially when points are at a premium in the race for the playoffs. This hedge-your-bets style of play can be traced to the 1999--2000 season, when the NHL began awarding one point for an overtime loss. The five-minute extra period had been part of the NHL since 1983, but until '99 teams that lost in OT got nothing. Now, they were guaranteed at least a point by going beyond regulation. This problem was magnified by the rule change after the '04--05 lockout that mandated shootouts to eliminate ties. By playing for draws in regulation, teams had a fair chance of earning two points for an extra-time victory. We want the two, but we'll take the one, has been an NHL mantra ever since. It's a flawed system that incentivizes teams to play it safe, robbing the sport of dramatic finishes in the third period while creating artificial parity.
In a study soon to be published in the Journal of Sports Economics, an examination of overtime probability from the 2005--06 season through the end of '11--12 shows that, in games decided by two goals or fewer, there was a 47.8% greater chance of April contests going to OT than those played in October.
The solution is simple: Award three points for a preshootout win and none for an overtime loss. The extra point would encourage teams to attack more instead of minimizing risk—to press for goals with the score tied, instead of just chipping pucks into the offensive zone and hanging back defensively. "[Teams do that] purely to ensure a point in the standings," says Jack Edwards, the Bruins' television play-by-play announcer. "That's a negative in a game in which offense has become more and more difficult to generate." (Scoring is down to 5.32 goals per game, the lowest level since 2003--04.)
Awarding three points for a win—even while still allowing for a 2--1 split between winners and losers in a shootout—would spark more shots and odd-man rushes as time dwindles in the third period and during OT. "Right now," says Islanders forward Josh Bailey, "there aren't a lot of chances being taken."
Three-point victories would also fix another inefficiency that is sure to be magnified when interconference play resumes next season. (No games between teams from the Eastern and Western conferences were scheduled during this year's lockout-shortened campaign.) In the 2010--11 and '11--12 regular seasons 23% more interconference games went to overtime than intraconference games. This was due largely to the fact that teams from different conferences were not competing against one another for playoff spots. In such games both teams benefited by getting at least one point. Next season the league's recently adopted realignment plan has teams in the East slated to play 28 such contests and teams in the West 32, up significantly from the 18 nonconference games teams played in each of the previous two seasons.
Yes, shootouts would still be used in a revamped scoring system—but a shootout win would only be worth two points. While shootouts are preferable to tie games, and while they provide plenty of brilliant highlights, they are ultimately skills competitions, akin to the NBA deciding a showdown between the Bulls and the Heat with a game of H-O-R-S-E. Winning in a shootout shouldn't carry the same weight as winning when teams skate five-on-five or, as they do in OT, four-on-four.
"You need the carrot in front of the donkey, and that carrot ought to be three points for a regulation or OT win," Edwards says. "If you want to reward people with the thrill of victory instead of the gimmick of a shootout, make it a real reward."
The NHL, to its partial credit, has debated the issue a few times since 2004. But the league remains as stubborn as a mule. Time to take a bite out of this carrot.
Michael Lopez is a biostatistics Ph.D. candidate at Brown and author of the study cited here.