Every person who follows the Edmonton Oilers closely knows that Devan Dubnyk is the best goaltender on the club with the rather unfortunate exception of everyone who actually works for the Edmonton Oilers. Bruce showed that pretty clearly earlier today by using a wide variety of comparisons that all show Dubnyk to be the better man. The really confusing thing about the team's unwillingness to change is that the team actually played much better with Dubnyk between the pipes, and even though I don't think that team performance isn't often a good reason to change goalies, NHL coaches often seem to disagree. Not this time. So what's going on here? Is there any reason for the team to prefer the old Russian to the younger puckstopper with the better season?
No. But that doesn't mean Dubnyk is any good. The rebuilding Oilers should definitely be starting Dubnyk in 2011-12 to help them find out if he is, and in fact, they should buy Nikolai Khabibulin out of his contract and replace him with another youngish goaltender (like Josh Harding or Viktor Fasth) to double their chances. Instead, they'll be sticking with Khabibulin, and while that's a pretty dumb thing to do, the difference in expected performance between the two netminders isn't nearly as wide as the chasm that existed between their numbers in 2010-11.
In large part, that's because we don't have much data for Dubnyk. Sure, he was excellent in 2010-11, but he was awful in 2009-10. His AHL performance that year was very good, but the two seasons before that were quite poor. It's a mixed bag. Even comparing him to Khabibulin doesn't make him look all that good. If we look at EV Sv% over the last four seasons, Khabibulin has posted a .917 save percentage on 3,850 shots, compared to Dubnyk's .914 on 1,316. Of course, we all know that Khabibulin is just plain terrible on the PK - over the last four years, Khabibulin's PK Sv% is well below-average at .852 (which is still a lot better than the .812 he posted in 2010-11!) - but Dubnyk's career record isn't that much better (.870). If we use this larger bunch of data (last four years), and assume that 80% of all shots will come at EV with the other 20% coming on the PK (we'll ignore PP shots against), then Khabibulin would be expected to put up a .904 save percentage next year, compared to .905 for Dubnyk. Not much to choose from there.
Of course, Dubnyk is a young netminder, and his recent performance is his best, so there's some evidence that he's improving, while Khabibulin is an oldster, and his recent performance is his worst, so there's some evidence that he's already fallen off the Cujo Cliff. So yeah. The Oilers should be starting Dubnyk, but it's not like the guy guaranteed to be savior. He might not even be average.
Projection: Dubnyk ends up starting just over half of the games and ends the year with an overall save percentage between .905 and .910.