Pull the Goalies

I have been thinking about this particular problem for years. Why is it a good strategy for a hockey team to pull their goalie in the last minute of a hockey game when they are trailing by one goal, but not a good strategy at any other time?

I’ll tell you right off the bat that I have a strong suspicion that almost every one uses this strategy because it’s always been used, and everyone else does. There is not a single coach, manager, or player who has any credible proof that it’s a good strategy. All of the evidence is anecdotal and religious in nature.

I believe that it is actually a bad strategy and that a team has a better chance of tying the game by keeping their goalie in the net. Sound strange?

The assumption at work is that by removing the goal-tender and adding a skater, you thereby increase your chances of scoring to an extent that more than offsets the rather obvious disadvantage of not having a goalie and thereby increasing the other team’s chances of scoring on you.

It is immediately obvious that this strategy is flawed in terms of logic. If removing your goalie and adding a skater gave you a real advantage, teams would do it all the time. Obviously nobody does. So why do teams think that doing it in the last minute of a game is different?

In defense of the strategy, people will argue that you only pull your goalie when you have possession of the puck and you are headed for the other team’s end. They argue that the attacking team will summon remarkable strength and courage in the face of adversity that will somehow bend the rules of logic and result in an advantage that no coach, in the history of the NHL, has been able to demonstrate under any other circumstance.

Having possession of the puck improves your chances for a few seconds, but it doesn’t really address the issue. Teams obtain possession of the puck in their own end dozens of times during a game. If that strategy works in the last minute, there’s no reason why it wouldn’t work in the first minute. Why not pull your goalie every time you have possession of the puck and start a rush out of your own end? Because the other team might get the puck back and score on you? How is this different in the last minute of a hockey game?

I suppose you could argue that 60 seconds is not long enough for the other team to get a good shot at your net. If that was true, we would rarely see an empty net goal. But we see them all the time.

There is another weird consideration. The defending team will quite often change it’s style of play as well– though not as much as they used to. The defending team will suddenly retreat behind their own blue line and form a box. When they get possession, instead of rushing down the ice and trying to score– precisely what the attacking team does not want them to do– they skate up to centre and then take a pot shot at the net. Often, they take the opportunity to change lines, on the weird assumption that the other team is going to get the puck back almost immediately and renew the attack.

The shot from centre doesn’t go in very often because usually one or more members of the attacking team are able to get back on time to block it, and the defending team doesn’t try to create a sustained counter-attack. But it does go in often enough to really finish off the team that is trailing.

We will never know the truth until some coach somewhere decides to go a season or two without ever pulling the goalie. But that would require genuine leadership. It would require genuinely independent thought. It would require someone unafraid of heresy.


The NHL does not even know when the first goalie was pulled. The New York Rangers, coached by Frank Boucher, are credited with inventing the move, in 1939 or 1940 or 1941 (like I said, the NHL doesn’t know).

I wrote here that nobody tries this strategy at any time other than the end of the game, but there was a game, years ago, in which the strategy was employed throughout. The circumstances were thus: two teams were going to be tied in points on the last day of the regular season. The team with the most goals (not the best differential, I note) would be the one to advance to playoffs. So when one team realized it was going to lose this critical game, they began to pull their goalie at every opportunity, in order to try to score as many goals as possible. Think about that. In other words, they acted as if pulling the goalie really was a good strategy. But if I remember correctly, they gave up as many or more goals than they scored.

The result, of course, was a chaotic, bizarre game that called the very integrity of the sport into question. The NHL changed the rules next year to ensure that this circus would not happen again.

[Added 2008-11]

It took me a while, but I finally found an online reference to that game referred to above. It was New York in 1970, in the 1969-70 season. Eddie Giacomin was in — or out of goal. The reference I found didn’t mention the other team involved or the final score.

I came across a report about a study last year that claimed to objectively prove that the strategy of pulling your goalie was, in fact, demonstrably effective. I won’t be convinced until I have a look at the study.

[Added 2022-05]

I did find that study and analyzed it.  The study shows only a tiny, marginal benefit to pulling the goalie.  But the study has one very significant flaw:  not one single NHL team will not pull their goalie in the final minute of a close game so it cannot compare the results with teams that do.  However, the in KHL (Russian NHL) pulling the goalie is quite rare: the prevailing belief there is that it is not a net benefit.

 

Leave a Reply

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