Not too long ago – before the salary cap -- teams were unconstrained in the amount of money they could spend to attract hockey’s top talents. Not surprisingly, the difference between what large market teams spent and small market teams spent was large, and it showed on the ice. (Although spending more was certainly no guarantee of improvement. I’m looking at you, 2003-04 New York Rangers.)
Small market teams were feeling squeezed and so, ten years ago, hockey fans lost a season’s worth of hockey as the players resisted the owners’ attempts to implement a salary cap. As we all know, the owners were ultimately successful, and the league is in much better financial health today, as evidenced by the recent Canadian broadcast deal and tremendous growth in team valuations (as estimated by Forbes).
The idea that competitive balance is critical to the financial health of a sports league has been around for years. In 1934, the NFL implemented the first reverse order draft in professional sports. Tim Mara, then the owner of the New York Giants, famously said “People come to see competition…We could give [it to] them only if the teams had some sort of equality.”
And the salary cap has, in fact, been quite successful in that regard. Today, there’s more competitive balance than at any time since the years before the league consisted of just 6 teams (current tanking for Connor McDavid/Jack Eichel aside).
The current state of parity is relatively easy to see. Sports economists have studied the importance of competitive balance quite extensively, and there is a well-established measure for it.
The basis for this measure is the spread of winning percentages. When teams are relatively equal, we should see less spread in the standings than when they are relatively unequal. To be clear – perfect parity does not mean all teams will end up with a record of exactly .500 – some teams will win more and others will win less just because of luck.
The standard measure of parity assigns a score of 1 when the spread in the standings is the same as what you would expect to see if all teams were equal, and that number increases as the spread increases (signifying greater inequality). Specifically, a score of 2 means that the spread is twice as much as would be expected if every game were essentially determined by a coin flip. (For those interested in a more technical description of how to measure parity, or just more on this topic generally, go to www.depthockeyanalytics.com, where an accompanying post will delve into the gory details.)
Using this measure, we can see that the only era with greater parity in NHL history is the 1931-42 period, when the league featured the first Ottawa Senators franchise, the Montreal Maroons, and the New York Americans, as well as the “Original Six” franchises.
In comparison to the other major sports leagues (NFL, NBA, and MLB), the NHL has at least as much parity as any other league these days. The NHL has employed a hard salary cap since 2005 and the NFL since 1994. Not surprisingly, those two have exhibited almost identical levels of parity since 2005.
|Era||NHL 1918-1930||NHL 1931-1941||NHL 1942-1966||NHL 1967-1993||NHL 1994-2003||NHL 2005-2013||MLB 2005-2014||NBA 2005-2013||NFL 2005-2013|
|Average Seasonal Parity||1.71||1.47||1.82||2.16||1.74||1.57||1.70||2.77||1.56|
Given that the salary cap constrains teams in their spending, how do they gain a competitive advantage? These days, if a team wants to improve, it can’t spend more so it has to spend smarter, and the use of analytics is proving to be a very helpful tool in that regard.
Using analytics together with traditional methods of information gathering, such as scouting, can help teams find useful players that are undervalued by the market, improve draft selections, improve player usage, and more. And it doesn’t take much to help turn around a team’s fortunes. Since the lockout, 23 teams have missed the playoffs by 3 points or less – almost 3 a year. Even small changes in the number of games a team wins can have big effects on positioning in the standings when teams are clumped together so closely.