With the NHL on pace for a record 25 percent of games going to overtime this season, there’s been much hubbub about finally correcting the sport’s loser point debacle.

For those unfamiliar with the issue: The NHL currently awards two points for a win (whether that be in regulation, overtime, or shootout), 1 point for a loss in overtime or shootout, and zero points for a loss in regulation. This means that for OT or shootout games, there are a total of three points distributed between the two teams compared to just two points for games that end in regulation. This creates a perverse incentive—one teams are clearly responding to—for teams to intentionally play for overtime to get this additional “loser point,” and it can have major standings implications.

Let’s say a team were to secretly agree with its opponents to intentionally force a tie in regulation for all 82 regular season games and then play out OT and a shootout. If that team were to win 50 percent of its games, it would finish the season with 123 points, good for the third-highest point total in the last 30 years.

In 2012, because the Los Angeles Kings had accumulated 15 loser points, they were able to sneak into the West’s final playoff spot and then go all the way to hoisting the Stanley Cup two months later.

The vast majority of hockey fans will agree that the loser point needs to die a swift death. There are largely three suggested alternatives floating around:

• The pre-2005 system of ending games in ties if overtime is scoreless, with each team getting a point


• A “3-2-1-0″ system in which teams get 3 points for a regulation win, 2 for an OT or shoot out win, 1 for a shootout or OT loss, and 0 for a regulation loss


• Continue playing sudden death as long as it takes until a goal

All of these proposals distribute the same total number of points for each game regardless of the outcome, removing the perverse incentive to force overtime. However, the tie and the 3-2-1-0 systems come with one major drawback: They make regular season games even more meaningless and less exciting than they already are.

Let me explain: In most American sports (with the exception of the NHL), games have a binary outcome—a win or a loss. This makes the difference between those two—and only two—outcomes significant. The further you granulize results from binary outcomes to ones with three and four outcomes, though, the less exciting you make games. For example, let’s extrapolate this out and ponder what’s more exciting: A single coin flip between you and a friend to decide who gets $100 (two outcomes), or 100 coin flips with the winer of each taking $1 (100 outcomes)?

In sports, we could extrapolate this granularity out to its extreme and decide the NBA standings by, say, mere points for – points against. This would leave games with about 60 possible outcomes (assuming the most NBA games are decided by 30 points or less), making individual games almost entirely meaningless. There would be almost no difference between a game in which you outscore your opponent 98-97 and one in which you score one fewer basket and underscore your opponent 96-97. In both the coin flip and basketball examples, the winner-take-all situation is far more exciting.

Now, I’m clearly taking this argument to its extremes, but you can see the point I’m making: the fewer possible results of an event, the more exciting it is. That’s why the NHL should stick to a system that merely allocates a winner-take-all outcome rather than one in which a the value of a win can be sliced and diced for distribution. While this may create some unfairness—i.e. a blowout would count just as much as a close victory—this luck would largely dissipate over the course of a long 82-game season. With so many games, the season is already extremely granular and needs no further precision.

Now, to be clear, given the fact that the season is 82 games, and the only reward for higher seeds is a slightly weaker opponent and an extra game played at home if the series goes to seven games, regular season games are already near-meaningless. According to a FiveThirtyEight analysis, the average game in the NBA regular season (which has an identical structure to the NHL’s) shifted a team’s championship odds by only 0.04 percent—many hundredths smaller than the significance of a playoff game. That doesn’t mean we should diminish their significance any further, though.

The only American sport in which it would really make sense to further granulize games would be the 16-game NFL. NFL games do technically have three outcomes now, but ties are extremely rare and only occur once every couple years. To make the standings more accurately reflect team performance, the NFL might consider forgoing overtime and simply having games end in ties after regulation. This would not have much impact on excitement and would also save additional playing time in a sport in which extra time on the field is especially physically grueling.

As FiveThirtyEight’s Nate Silver shows, though, aperpetual sudden death system would only burden teams with an average of two extra minutes of ice time per game—even less if a 3-on-3 format is adopted. This is a tiny price to pay for bringing some excitement to a very unexciting regular season.

Follow Jim Pagels on Twitter at @jimpagels