On Parity

When the Devils and the Flyers ended regulation at a 3-3 tie on Sunday afternoon, it was the 18th game of the 2012 Stanley Cup playoffs to go to overtime. That’s a lot of OT: 35% of the games, including five elimination matches and two game sevens. The Chicago-Phoenix series went to overtime five of its six games, every single battle except the last. If it had not been for the Penguins and Flyers putting on their retro 80s tribute to the lost arts of pointless hate and shitty goaltending, the entire first round in the East would have played out in tightly-contested, evenly-matched hockey. It’s been a very close postseason.

It’s possible that this means nothing. It might just be a coincidence; hockey is full of those. Sometimes players shoot 35% for a week or two for no apparent reason. Sometimes a team goes on an inexplicable slump and wins only 35% of their games in a month. Maybe 35% of the playoffs going to OT is just another one of these weird runs of hockey luck. It might be so.

But, although coincidence is a possibility, overtimes and seven-game-series are also indicators of parity, and parity is an issue that desperately needs further consideration. Because parity, kittens, is the future, and we best figure out soon if it is the future we want.

First, remember this: hockey is naturally a parity-prone sport. It is a low-scoring game; goals are rare and their occurrence is highly influenced by non-repeating situational factors. The points don’t come too often and when they do, the proximate cause is often a rare opportunity that arose spontaneously from a blender-whirl of offensive effort, defensive lapses, and butterfly-effect bounces. So although there are definitely ways that skill and strategy can influence the likelihood of scoring, there is also a large percentage of randomness involved in the business of getting a puck past the goalie. Since often it is only one of these rare goals that makes the difference between a win and a loss, there is therefore a large percentage of randomness involved in winning a hockey game. The sport’s strong strategic tendency to evolve towards defensive systems, which both reduce scoring and improve the ability of less gifted players to restrain their star counterparts, also vastly increases the inherent parity of the game.

Moreover, several historical trajectories in the NHL specifically have pushed its brand of the game more and more towards parity. Early hockey involved relatively few players bound to relatively few franchises for long periods of time. The best player on a team could easily play 60 minutes a game, every game, for one team, for his whole career. Under such a regime, team-building consisted of finding seven good bodies and signing them to neverending contracts. Constructing and maintaining a dominant team, then, required some deviousness and perhaps a certain lack of moral scruple, but it was relatively easy and frequently done.  The League could easily be unbalanced, with a few elite franchises enjoying long runs of Cup victories and others never so much as touching the thing once.

But over time, rosters grew. The notion of lines and shifts emerged, and more and more lines were added until every team was twelve forwards deep. The best player now divides his icetime with three less-talented counterparts in the same position. He plays at the very utmost 1/3 of a game. The League grew as well, and as more teams were added, talent became further diluted. Diluted talent favors, again, the growth of defensive hockey, as more and more teams are forced to figure out how to compete effectively with less offensively gifted players. Again, defensive systems mean more parity of outcomes.

Then the modern entry draft was introduced, regulating the entry of players into the League. Rather than being able to get ahead by building their own minor and junior systems, hunting down talented 12-year-olds in the desolate northern mining towns, every team now gets access to players at the same time according to the same system- a system which is deliberately constructed to create parity by giving the best prospects to the least successful teams. And players, once drafted, now have the collective bargaining rights to move teams of their own free will at exactly the time they reach the apex of their skill.

The ways by which the early teams built their dynasties are all closed down now. The great managers of old countered the natural parity of the game by hoarding cheap talent that the modern manager must set free or else buy dearly. They controlled sophisticated talent-feeder systems that the modern manager is legally prohibited from using. And they could build what would be, by today’s standards, virtual All-Star teams, whereas now a GM must give money and ice time to players so weak they can only be trusted to play six minutes a night against equally terrible opposition.  And then, to top it all off, after the lockout the NHL initiates still more deliberate parity-inducing measures: the salary cap and floor, which force every team to spend approximately the same amount on rosters, limiting the influence of local economic capacity on team talent; and the three-point shootout game, which introduces a raft of extra bonus points into the season, bonus points which are given exclusively to losing teams.

Executive summary: the history of hockey is that of a naturally high-parity game evolving inexorably towards even more parity. It is a flat world that only grows flatter, the way Iowa segues into Nebraska. It is an endless evening-out.

Some of that is inevitable. Some of it is what we chose. The question is: do we want to keep choosing this? Do we want parity?

The amount of parity in the League affects far more than just one team’s chances of winning the Cup. It has consequences for the structure of the game, for the kinds of strategies and tactics we see played out on the ice. It has consequences for the way teams are built and managed. It has consequences for the life-cycles of franchises. Most of important of all, though, is this: the amount of parity in the NHL defines the causes of winning. And- here’s where it gets personal- if it defines the causes of winning, then it will define the kinds of stories. In choosing for or against parity, we are choosing the hockey narratives of the future.

Parity favors some elements of hockey over others. It does not, for example, favor the work of general management. The more all teams are inevitably equal, the less influence good management has on outcomes. Even now, the best GM in the NHL using all the most sophisticated technologies and advanced metrics available has only a small edge of the competition. The more parity increases, the less far behind the stupid franchises can fall, and an incompetent manager riding a lucky season has more and more of a chance of snatching the Cup away from the shrewd builder of a vastly superior team.

It does not favor dominant skill. Of course, barring the attrition of franchises, the number of good players in a high-parity league will be the same as in a low-parity one, but between the entry draft and the salary cap, the forces of parity make it difficult for talent to collect heavily in any one place. Talent thinly spread across thirty teams means thirty very equivalent teams, meaning the difference between one winning and another will be things other than skill level.

It does not favor the building of dynasties. No team will ever again put together a roster that will stay stable for a decade and win seven Cups in the span. Either the roster will destabilize due to free agency and perfectly reasonable contract demands or the victories simply won’t come at the expected rate. Either way, in a parity NHL, there is no great generation of a team. There are only great years interspersed with poor ones, with an ever-shifting cast of characters.

Parity favors randomness. It favors a wider range of teams winning by narrower margins on the basis of smaller causes. As parity expands and the talent distribution between teams gets narrower and narrower, fewer and fewer games become decisive victories. More games go to overtime; more series go to seven. As teams match more evenly against each other, short-term unsustainable forces have more influence on major outcomes- bouts of flu and unexpected injuries have more power to utterly derail a season, just as bursts of hot goaltending and shooting have more power to take a team to the top of the standings. Skill matters less, bounces matter more. This also, of course, goes for human choice. The more even opponents are, the closer games are, the greater the likelihood of one individual effort or failure having decisive impact.

It also favors extravagance, gambling, rapid change, and uncertainty. With the rewards for sensible, measured management being so slim and the chances of luck-driven outcomes being so high, parity encourages managers to take risks. It encourages short-term thinking without regard for long-term consequences. A GM whose work is subject to a lot of luck is in an eternally precarious position, his job might end rather suddenly mid-season regardless of the intelligence of his moves. The incentive for him, in Parityland, is often to do the popular thing rather than the right thing, for the difference in on-ice outcomes between the two is uncertain, whereas the difference in fan and media reaction is clear.

This isn’t necessarily all bad. Personally, I think I like the shape of a parity League, but then again, I like unpredictability.  I like broad talent distribution and the hopeful thought that absolutely any team could win the Cup in the next five years.  I like overtime, and underdog victories, and game 7s. I like stories about heroic individual efforts, buzzer-beating goals past momentarily upended goalies with everything on the line. For mine own part, I can live quite happily in a world without dynasties, where GMs are forced to fight bitter wars over the tiniest scraps of potential edge and often go mad in the process.

But a flat NHL requires adapting the way we think and talk about hockey. The more parity in the game, the more teams like Vancouver and San Jose will appear- teams who play the game well, insofar as it can be played, and are regular contenders because of it, but who simply never get to the Cup. An old-fashioned hockey mindset would hold that these teams must be somehow fatally flawed, and once upon a time, that might have been true. But parity means that plenty of good teams will go through that experience, of being able to make the postseason consistently only to find themselves often bounced out by, well, a bounce. In Parityland, the winner doesn’t necessarily have a magic formula and the loser didn’t necessarily do anything wrong. Winning doesn’t necessarily mean being inherently better at hockey anymore. Rather, it becomes a tautology: the winners won because they won, because somebody had to win, and they happened to be ahead when time ran out. Parity means more chances for everyone. It also means more chance in the game.

We are moving towards a singularity for which we are not prepared. If we continue pursuing parity as the ideal of the NHL, we will eventually come to a point when skill is distributed with such perfect balance and equity that there is nothing to determine outcomes but utter chance. That point hasn’t come yet, but it will, if we keep pushing parity of every level, parity of inputs and outputs alike. Hockey people are not prepared for this. We are still uncomfortable with talking about luck and randomness in the game. We still balk at acknowledging the extent to which the results in front of us are driven by chance, preferring to fall back on the more comfortable assumption that every team that wins does so for good reason. We think about parity hockey in dynasty terms, still. Which suggests that, when the singularity comes and all results are based on luck, we won’t even know it.

[What precedes is, obviously, more a speculative outline than a fully-developed argument.  The fully-developed argument would be tens of thousands of words long and probably require a binding.  There is a lot left unsaid which needs saying, and a lot that can and should be debated.  So, if you had the patience to make it to the end and you feel like arguing, please argue with me.  I want to talk about this one.]

Comments (19)

  1. This ties in with my hatred for the current draft system and with a piece Tyler Dellow wrote recently about the rise of parity. I’d prefer a league where skill (both in the players and in the management) plays a larger role than luck. Rewarding teams that failed (and make no mistake, very few teams that get high draft picks originally intended to do so when the season began) rewards incompetence and furthers parity.

    I’m a Leafs fan so I’ll give an illustration from their perspective. I think the post-lockout Sundin teams, which narrowly missed the playoffs every year, deserved a high pick a lot more than the current abysmal incarnation does. All the current system does is reward incompetence and thereby help entrench incompetent management.

    • I don’t see it as rewarding incompetence. It is in the leagues and PA’s best interest to have as many ‘healthy’ teams as possible. A high draft pick gives season ticket holders a reason to renew when they otherwise wouldn’t. It provides a hope that the current suckitude of your team is not a permanent situation. Also, teams that suck will naturally have a harder time signing UFAs so you need to give them a way to get competitive again.

  2. In my opinion, there will always be haves and have-nots. A lot of that will be based on the GM’s shrewdness, or lack there of.

    Here is my argument in favor of parity: (Practically) no one enjoys watching a team lose year after year. Example: Atlanta.

    If we are serious about wanting franchises in “non-traditional” hockey markets, they need to be a part of playoff hockey. Otherwise, hope lags, followed rapidly by attendance and interest in the team. Once that happens, queue the inevitable move to the frozen north, into markets that are maybe not as large, but will support a team through rough times.

    I am not saying let’s be sure to purposely spread talent equally through the league. Let’s just make sure that teams are able to develop and hold on to a couple high talent players, so fan bases don’t become utterly hopeless.

    • Good point, although parity also creates problems for fan interest down the road. Most fans want to believe (I think) that their team is building towards some durable success, and parity makes that hard. It creates, at best, a lot of bubble teams and also-rans, while at the same time not entirely ensuring against unmitigated failure. Parity makes it harder to win consistently, but could be it doesn’t make it any harder to lose consistently.

      • The problem with durable success for some teams is that it’s inevitably creates durable failure for other teams. Sure, there are lovable losers like the New York Rangers (1940!), but then there are the losers that get themselves into financial trouble. Not so good for the league on the whole, if they have to either contract or find a way to support these teams who can’t get attendance figures or TV revenues.

        I’m sounding increasingly like a hockey communist, aren’t I?

  3. Do you think, Ellen, what you’re talking about can be situated in the conversation about hockey and the intersection between religion and science? When fans want to see the “Good Guys” win, when they talk about “the right thing to do”, or when people mention grand narratives – are they looking for a spiritual hockey experience? Is parity the emergence and the beginning of an awareness in hockey culture that there are no Hockey Gods?

    I think the more fans watch hockey, read blogs, inhabit Twitter – the more they give microscopic attention to the sport – the more likely they are to interpret hockey scientifically. The same might go for Owners and GMs. When I was younger and less inclined to know the smaller yet significant details of the sport, I believed in hockey heroes and that Wendel Clark was doing a public service for the Toronto Maple Leafs by leading the team I was devoted to deep in the playoffs. I suppose I had Hockey Faith and when things were bad for the Leafs, I felt bad. As if there was a lesson for all fans to learn from the mistakes that the team had made – a story being told.

    I still like submitting to this feeling from time to time but am now conflicted, knowing all that I know because I follow everything to a finer point. Like looking through a microscope. But for those who follow less intensely, perhaps they do not even notice parity because they subscribe to one team only, hold up favourite players as Saints, and look to fit storylines into greater narratives that they simply want to believe in.

    • This is a fantastic comment and it does a great job at explaining why there’s such a strong resistance to advanced stats in the traditional hockey community–as Neil Young once sang, “when the aimless blade of science slashed the pearly gates…” I’m not sure, however, that it works as an argument in the case of parity. I think what the article is espousing is more that, stats or no, in hockey’s past those great Saints actually did exist. The Canadiens won all those cups because they just had better players, or, closer to my own heart, the 80s Oilers were able to win their cups because they really were ridiculously better than the teams they were playing (I think in one cup run they managed to lose maybe one or possibly two games in the entire playoffs). Randomness wouldn’t be that much of a factor when you had Wayne Gretzky going up against Richard Brodeur.

      Another thing to consider, in talking of parity and the age of advanced stats: how much does it matter to look into all these stats, your corsi relative/fenwick/scoring chances etc if they point to the fact that one team should win a series, scientifically speaking, but randomness dictates that the other team wins anyway? What explains that better: luck or a nice narrative?

      • Advanced stats can’t tell you who should win a series. What they can tell you is the likeliest result over time. They can’t tell you who is going to win a game, but they can tell you which player is better in a given situation given a large number of games. A seven game series isn’t ever a large enough sample size.

        If you have flip 100 times, you’ll probably get results that are reasonably close to 50 heads and 50 tails, but if you look at random sections of seven flips, you’ll find some that are weighted heavily towards one or the other. Possibly even seven in a row.

        Where advanced stats are useful is in evaluating a team over time, identifying players who are better or worse in particular situations and suggesting possible improvements by substituting players who are better at a particular role. But prediction is never going to be accurate unless one team is enough better than the other that it would overwhelmingly win a much longer series than seven games.

    • There’s an element of simple faith in fan culture, sure, but I think for most people it breaks into an at least quasi-analytical mindset by adulthood. Most non-specialist fans I know- people who aren’t watching/reading about hockey ten hours a day- still try to think about the logical causes behind winning. They may be less rigorous, but they still think about the game in terms of structures and systems. They’re just very old school structures and systems.

      Thing is, I think Parityland actually creates more of the kind of narratives fans love. More individual heroics, more importance placed on short-term phenomena, more underdog upsets. The problem is that fans want these kind of stories but they also want to be able to attribute durable causes to them, and you can’t have it both ways. The Canucks can’t both take the Bruins to seven games in the SCF AND be a horribly flawed team that’s never going to win. The Flyers can’t both be clearly better than the Devils AND be winning on OT goals from Danny Briere. Parity forces a trade-off between dramatics and just outcomes, and it’s tough to choose between the two.

  4. I just don’t really think the league is headed for total parity. Even if it was something that could be achieved, i don’t believe that is the goal. Not to mention that people have proven, time and time again, to be infinitely creative within their constraints. Teams, GMs, and individual players will find ways to increase the odds of winning. If that means new strategies, new business models, taking pay cuts, etc… they will do it if it means they are more likely to win.

  5. I don’t think we’re approaching a parity singularity. The NHL is still a league where rosters are decided by humans, humans with a great variety of competency. Injuries still happen, surprises still happen. I would argue that controls like the draft and salary cap are not to create equity, but fairness. Fans are okay with dynasties, as long as they are fairly created dynasties, that all teams have an equal ability to create.

    For example, in the NBA there is a perceived lack of fairness between large warm weather cities and cold, small cities because players will not consider some teams. Fans of the Raptors or Bucks feel they are at a disadvantage that they cannot control.,

    • Two points:

      1) I don’t know that you can argue that the entry draft and the salary cap create ‘fairness’. Fairness, it could be argued, means getting what you earn. The Canadiens built their 1950s dynasty because they took the time to create a broad, well-tended network of farm teams- it was their ingenuity that made them more successful, and that’s not unfair. Similarly, why is it ‘fair’ that the worst teams get the best players in the entry draft system? Why is it ‘fair’ that poorly managed teams get revenue sharing from successful teams? Sports are a competition; the point theoretically should be to reward the people who play the game (both on and off the ice) best. A system that compensates the worst for being bad isn’t making things more fair, it’s just making things more even.

      2) It’s not just parity of inputs, it’s parity of outcomes as well. Three-point shootout games, for example, create parity on the backend by inserting more points into the standings, making it easier for weaker teams to stay in competition for the playoffs longer. Why? That’s not levelling the playing field, that’s levelling results.

  6. Not being a pious man by any stretch and fond of riding tandem with the random, your elucidation of a state of parity and the significance of chance & chaos in determining success or failure called to mind an apt verse of succinct existential wisdom from Ecclesiastes in the KingJames Bible ;

    I returned, and saw under the sun, that the race is not to the swift, nor the battle to the strong, neither yet bread to the wise, nor yet riches to men of understanding, nor yet favor to men of skill; but time and chance happeneth to them all.

    Same as it ever was…. through the ages mankind has deployed narrative and superstitions as comforting remedies in dealing with the disconcerting realm of luck and chaos.

    The case of Phoenix and their current success seems to contradict what you hold to be some of the key elements vital for success in the age of parity
    I believe that much of the Coyotes success in these early stages of era of parity is due to the rewards of smart , measured ,sensible management , reinforced by force of circumstance in which the entire day to day focus of a desperate franchise is doing what whatever necessary to ensure it’s precarious survival. The utilitarian narrative of this group of overachievers is rooted in it’s history of struggle evidenced by the absence of panic and fixity of purpose in their game despite often being severely outplayed in every category except goalkeeping. ‘Coyote hockey is the Zen of hockey….

  7. “We are moving towards a singularity for which we are not prepared. If we continue pursuing parity as the ideal of the NHL, we will eventually come to a point when skill is distributed with such perfect balance and equity that there is nothing to determine outcomes but utter chance.”

    Could you sound any more pretentious and overly dramatic?

    • Oh Lordy, yes! I can be at least twice as pretentious and probably ten times more dramatic if I really put my mind to it.

      Did you have a point to make or…?

      • Your whole concluding paragraph sounds like a doomsday prophecy. Hockey has ALWAYS been this way. Especially in the playoffs, games have always hinged on the stray, lucky bounce and the guy who is there to bury it. The 50′s Canadiens played in a league with much fewer teams. I really don’t see why you feel this argument requires “tens of thousands of words.” Considering it would be your prose, it would be like trying to wade through quicksand. Good day.

  8. That’s a really interesting argument. I’d never quite put it together that an increase in parity implies an increased role for randomness. Unanticipated consequences ahoy!

    Although it does also suggest that individual superstars, even on teams that are of average quality, will also have a much greater potential impact. One Crosby or Ovechkin on a tear could take you a long way – a lot more Cinderella runs possible.

    I’m not sure that I entirely agree on the GM front, though. As we’ve seen, bad coaching and bad GMing can still do a lot of damage to a team, and that’s still going to be true. Yes, a “good” team in a parity league that is only 1% better than the average can be tripped up by an unlucky bounce or two. But truly terrible drafting is still going to sink a team. Perhaps what actually happens is a mostly flat NHL, with a bunch of basement teams who have made really bad management choices?

  9. Ellen,

    Great post, but I have a few complaints.

    Don’t call me a kitten…

    Stupid management, like that Wang that runs the Islanders, will overcome parity in spite of all efforts to thrust it upon them. And those who ignore science in favor of truculence may be doomed to set records for missed play-offs.

    I think that an important element of chance may be the personality make up of the team. In a prior lifetime I was a goaltender, wearing a Jason type mask which dates this story. I remember playing for teams who were in the bottom half of the standings and that on paper did not stand a chance of success once the playoffs started. But some combination of guts and heart helped us overcome far more skilled teams, once winning a championship and in another year an OT goal stopped us short of another. Perhaps that is only to be seen at the amateur level because the pros are too well compensated and their competitive fire is just a warm lump of coal. I think Olympic hockey proves that desire and character can produce synergy.

    But if some GM were to figure out the secret of building a team on the qualities of desire, character and the intangible heart – parity might be pushed aside. I can see scientists running around the junior leagues, taking DNA samples and putting players through brain scans, while physcologists probe their physche. But this assumes of course that the GM and every staff member privy to the “secret” are somehow sworn to secrecy and are locked in the team offices when they are not guarded by burly persons with communication devices in their ears. Otherwise parity will once again reign.

  10. Great article. My concern about parity is how it is playing out in the on ice product. Lower and lower scoring games. Goalies are only going to get better, coaching as well. I don’t need a 6-5 game to enjoy it, but I do need a ton more chances. I have a suggestion as to how to solve the problem (admittedly some may not see it as much of a problem). Outlaw the neutral zone trap, and any/all of it’s derivations. Basically force teams to send in 2 forecheckers. No more sitting back in the 1-4 or 1-3-1. Obviously it would be a tough sell, you’d need a ref upstairs to judge and then 1 warning, 2nd time it happens it’s 2 minutes.

    Hockey is a beautiful fluid game but the trap sucks the life out of it. Why is an offensive chance filled game any less “pure” than the defensive snore inducing contests we see most often today?

    Curling reapplied it was a SPECTATOR sport. It instituted the free guard zone and as a result became watchable again, and in fact flourished. To me outlawing the trap is the same thing.

Leave a Reply

Your email address will not be published. Required fields are marked *