In light of last night’s game in Vancouver, where Dallas Eakins started Ilya Bryzgalov despite Ben Scrivens having given the Oilers a rare (for them, not him) decent goaltending performance the night before in Edmonton, I’m writing about playing goalies in back to back games over at Sportsnet this week.
One thing that I doubt will make the cut for that piece that I thought might be of interest to the more analytically inclined crowd here: starting goalies in back-to-back games following a shutout. Eakins indicated in his press conference after Sunday’s game that this might be an exception to the idea that you shouldn’t start goalies in back to back games.
I’d consider it except the stats don’t line up for you very well going back to back with a goalie. It’s one thing you usually push a guy into if he had a shutout or something, I think you have to go with him. We’ll talk about it a little more but my early vote will be to go with Bryz.
The shutout business caught my attention. I’ve put together a database with every appearance from goalies in back to back games between 2000-01 and 2012-13. That gave me 2800 sets of games. Overall, the goalies had a .9147 save percentage on 74,631 shots in G1 and a .9068 save percentage on 76,250 shots in G2.
What about guys coming off appearances in which they didn’t allow any goals though? The hot hand? Well, they had a 1.000 save percentage on 6,651 shots in G1 and a .9089 save percentage on 8,354 shots in G2. If you back these shots out of the overall data for G2, it means that guys who are coming off a shutout appearance have a collective save percentage of .9089 in G2 and guys who are coming off an appearance in which they allowed a goal posted a collective .9066 in G2.
Does that justify Eakins position, that two point edge? I’m not sure that it does. The edge is probably not really as big as it appears. Who’s more likely to get a shutout in G1? Henrik Lundqvist or Cam Ward? Obviously, it’s the guy who’s better at stopping the puck; Lundqvist has eight shutouts in the first game of a B2B; Ward has three. Who would you bet on to be better in G2? Lundqvist or Ward? The answer’s pretty obvious.
I’d guess that the issues that make back-to-back games a good time to give a game to each of your goalies – fatigue and lack of recovery time – exist regardless of whether or not a guy got a shutout in G1. Eakins position may still be defensible but if it is, it’s unlikely to be because you can expect a particularly good performance from the guy in G2. A defence would have to be premised on the team and goalie expecting a great game to be rewarded with the next start. If you haven’t conditioned your players out of that expectation, you’re going to have to deal with fallout if you don’t meet it. Other than that, the expected performance drop when a goalie plays in B2B games seems to me to make it an obvious time to give both guys a game.Email Tyler Dellow at firstname.lastname@example.org