I was perusing the standings last night, for no apparent reason, and noticed a couple of anomalies that brought an old source of irritation back to the surface -- the three-point games in the NHL. Specifically, I was noting the fact that the Blue Jackets have the same number of wins as New Jersey and Philadelphia . . . yet trail the Devils by six points and Flyers by nine points in the standings. Despite a horrific year of injuries and related turmoil in Columbus, the Blue Jackets have just one fewer win than Florida -- which holds the ninth position in the standings, on the doorstep of the playoffs -- yet lags the Panthers by a dozen points in the standings.
The answer, of course, is the dreaded "three-point game" -- that numerical curiosity that can be invoked by silent mutual assent of two teams -- or not -- as their particular situations dictate. It's existence has a disproportionate impact upon teams that have slow starts in October & November, as the hill to climb steepens when victories frequently gain only one point in the standings, rather than two. As of this morning, there have been 1,002 NHL games played this season, which theoretically should result in 2,004 points being awarded. However, thanks to the idiosyncrasies introduced by the three--point game, 2,255 points have been awarded, a roughly 10% premium. However, rather than vanquish the three-point game, we should embrace it, make it the standard, and return the game to numerical rationality.
The three-point phenomenon arose largely because of a media-driven belief that ties are an abomination, and that overtime and a skill contest provide not only a winner, but extra advertising and revenue opportunities for the game as a whole. Even leaving the perhaps overly-cynical view of motivation aside, the system at once recognized the fact that extra effort deserves some additional recognition, which can be viewed either as giving the losing team the "extra" OTL point, or giving the winning team the "extra" point over what they would have had under the system that permitted ties. Not necessarily a bad idea, in the abstract, but the execution was horribly flawed. Leaving the manner of tie resolution (i.e. overtime and shootout formats) to a separate discussion, the system created a mathematical oddity, where now some games were worth more than others. More troublesome was the fact that there was no objective way to predict or control when those "extra" point games would arise, and that the combatants themselves could effectively control the question via their play (or lack thereof) in the waning minutes of games that would otherwise have ended in a tie.
I'm a bit of a purist when it comes to things like statistical comparisons, as without some objectively rational and repeatable standard, the basis for comparison across teams, years and decades becomes impossible. I have bristled at the seemingly endless wailing that the "steroid era" has disrupted the sanctity of the home run record. While I love baseball, it's difficult to get too worked up about a record where teams have changed stadiums, fence distances and fence heights, the rules have altered mound heights, and the ball has changed in composition. Yes, steroids enable athletes to recover faster, just like the handfuls of amphetamines gobbled by players for decades to overcome everything from hangovers to the common cold. Had Willie Mays played in Milwaukee County Stadium and Fulton County Stadium, rather than the Polo Grounds and Candlestick Park, he would likely have had 800 home runs. However, I digress. The point is that there has to be some consistency in the arithmetic applied to the game. This is especially true where, as here, the numerical variations affect not statistical comparisons, but the very essence of the standings and team success. A game is a game, and it needs to be worth the same number of points, no matter when, where or how it is played.
The rule book provides that a game of hockey is sixty minutes long, with five skaters and one goaltender for each side. Two points go to the victor, zero points to the loser. In the days of the tie, those two points were simply split between the two teams, rewarding the teams commensurate with their effort, yet preserving the mathematical integrity that every game was worth two points. When we decided that ties were no longer acceptable however, we apparently determined that the math was no longer important, and suddenly created the "extra" point -- and it matters not which team you view that point as going to. From a pure logic standpoint, this makes no sense. A team that is able to overcome its opponent in the stipulated sixty minutes would rightfully be viewed as having outperformed (in success terms) a team that needs more time to accomplish the same feat. Similarly, a team that is able to stay stride for stride with its opponent through regulation and into overtime or a shootout has performed more ably than a team that succumbed in three periods. While the latter is accounted for by the OTL, there is no corresponding penalty for the team that wins in extra time, when compared to the team that prevails in regulation. Of equal importance is the fact that teams can -- and do -- manipulate the occurrence of three point games. How many times have we seen the pace of a game suddenly come to a crawl as the clock winds under two minutes to play, as the defensemen hold the puck behind their own net for extended periods, waiting out the clock to guarantee the two points, while the opponent drifts aimlessly in the neutral zone?
The solution is neither difficult nor arcane -- simply adopt the three-point game concept already in place internationally and in the college ranks. My personal preference is three points for a regulation win, two points for an OT or SO win, and 1 point for an OT or shootout loss. Some would argue that the three points should be available for an OT win, with the 2 - 1 split happening only in the shootout. While that would be an improvement over what exists today, I'm not a fan of awarding the same points for something that is not the same game. Whether it is 4-on-4, 3-on-3 or a shootout, it remains some variation of the game of hockey, devised solely to resolve a tie. If overtime was a full period of 5-on-5, I'd have less problem with the concept, but it isn't and won't be, for a lot of obvious reasons.
Ironically, it is the traditionalists who scream the most about the possibility of changing the point allocation system. They argue that the gap created between winning and losing would increase, making it harder to overcome deficits. That, of course, is simply not true. If Team A and Team B are tied, and Team A wins its next game in regulation, while Team B loses in regulation, it is true that the gap between the clubs is now three points instead of two. However, that gap still represents just one regulation victory, so the result necessary to close the gap remains the same. What does change is the possibility that a winning team might get only two points instead of three. The significance of that prospect is that now a team coming down the stretch of a tied game now has a thornier dilemma. Do they play "safe", possibly guaranteeing a single point, but knowing that they will lose a point to any competitors who earn regulation wins? I suspect not -- at least not as frequently as it arises today.
In applying the three point system to this year's standings, some subtle but interesting changes occur. Tampa Bay assumes the top spot in the league, with 127 points (vs. 90 under the current system), followed by the New York Rangers (125 vs. 91) , Montreal (123 vs. 91), Anaheim (120 vs. 91), New York Islanders (120 vs. 90) and Nashville (120 vs. 91). The Pacific would be shuffled, with Vancouver moving to the #2 slot, Los Angeles to #3 and Calgary to the second wild card slot. Chicago, meanwhile, would fall to the first wild card slot, with the Wild claiming third in the Central. However, it is admittedly a different matter to retroactively plug in this system, versus actually playing with it in place. I suspect that the number of OT/SO games goes down under this scenario.
From where I sit, this is a simple, elegant solution to a mathematical and logical abomination that is anti-competitive and inherently subject to manipulation. Sure, the prospect of manipulation remains, but now there is a greater price to be paid, so I think that factor is reduced significantly. It's already in wide use elsewhere in hockey, why not the NHL? What say you?