Is it Fit to be Tied?
Copyright Iain Fyffe, 2002

Please visit Puckerings for more hockey stuff by me

Why is the NHL so concerned about ties?  With so many things about the game to be concerned about, why the recent emphasis on rule changes intended to reduce ties?  First, 1999/2000, we started having a point for an overtime loss.  Fine, make the NHL more like the minor leagues.  That wasn't enough, though.  Last year we started having four-on-four play in overtime!  What the hell is that all about?

But before I complain too much about these rule changes, let's see if they actually accomplished their task.

We'll look at the data from the past three years. The following table shows total league games (GP), games that ended in ties (T), and the percentage of games ending in ties (T%):

Awarding a point for an overtime loss reduced T% by 1.9%; it seems to have worked.  Four-on-four in overtime reduced T% by only 0.3%; this is a minuscule number for such a radical rule change.  It seems that four-on-four is a flop.  The 0.3% change amounts to about four ties in a season, and that's not much.

Let's examine the results on more detail.  The following table shows games that went into overtime (OT), the percentage of all games that went into overtime (OT%), and the percentage of overtime games that ended in ties (T%): 

Now the effects are clearer.  Awarding a point for an overtime loss had two results: (1) overtime games were decided much more frequently, and (2) more games went into overtime.  Both of these make sense.  More overtime games are decided because a loss is as good as a tie; both are worth a point.  But this fact also gives teams incentive to play for overtime.  If they can get into overtime, a point is assured.  So teams have incentive to play boring defensive hockey near the end of a tied game if all they desire is a point.

Again, playing four-on-four in overtime had absolutely no observable effect.  0.3% less games went into overtime, and 0.7% more overtime games were decided.  Again, for such a drastic rule change, these results are pathetic.  Four-on-four is definitely a flop.

Time for some opining.  Either ties are acceptable, or they're not.  The way it is now in the NHL, sometimes ties are okay, sometimes they're not.  What is the point of having a 5-minute overtime period?  If ties are not acceptable, you should play until somebody wins.  If ties are acceptable, then playing an extra 1/12 of a game (or less) is ridiculous.  The NHL is schizophrenic in its attitude toward ties.  In my mind, there's nothing wrong with a tie.  It's a perfectly legitimate outcome of a hockey game.

But if you really want to reduce the number of ties, play 10-minute overtime periods, or just keep playing until somebody scores.  This would also prevent playoff hockey from being a different game than the regular season.  Forget about playing a different game (that is, four-on-four or a shootout, they're not hockey) in overtime.

But there's one question left: do fans really care about the number of ties?  Presumably, that's the NHL's prime reason for the rule changes; they believe that if ties are decreased, fans will find the game more exciting, and attendance will increase.  Ties have in fact been decreased; now let's look at attendance.  Note that the following figures include only the teams that played in all three years (that is, Minnesota, Columbus and Atlanta are excluded):

So, in the year in which ties dropped significantly, attendance did not change.  And the following year attendance "jumped" by 0.8%.  Over the two years of tie-reducing rules, attendance has barely moved.  So what was the point of toying with the game so?  It seems fans are not concerned with the number of ties in the NHL.  Perhaps they view ties as a legitimate outcome of a hockey game?  What a concept.