Winter Olympics 2014: Ryan Miller should be Team USA's starter

Jonathan Quick will start for Team USA in their opener tomorrow against Slovakia but, beyond that, Dan Bylsma would be wise to make sure it's Miller time in Sochi.

For all the talk about a goaltending controversy brewing for Team Canada, as they decide between 2010 gold medalist Roberto Luongo and rising star Carey Price, the more important crease-related decision at the 2014 Winter Olympics may rest with American head coach Dan Bylsma and the rest of his coaching staff. The first iteration of that decision has been made, with Bylsma announcing earlier today Los Angeles Kings goaltender Jonathan Quick will get the start in tomorrow's Team USA opener against Slovakia with the Buffalo Sabres' Ryan Miller backing him up.

On the surface, the Quick vs. Miller debate might seem like little more than arguing over heads and tails on a coin flip. And in a lot of ways that's true; over a two-week, single-elimination tournament, any NHL-caliber starter is more than capable of getting hot and stealing rounds or even a medal for his country. We saw it four years ago with Miller and, to his credit, Quick managed a similar feat over a longer high-stakes tournament in leading the Kings to the 2012 Stanley Cup (of course, the team in front of him had a whole lot to do with that as well).

And yet, as much as factors outside of Bylsma's control may end up determining Team USA's success in net, he would be foolish not to play the goaltender with the better odds of going on one of those legendary runs. That goaltender is Miller. He's been better this season, with a .923 save percentage while playing for, by far, the worst team in the NHL compared to an injury-hobbled Quick's .911 playing for one of the best. He was better last season, posting a .915 SV% compared to Quick's dreadful .902. And he's been better over the past five seasons. Since 2009, 43 NHL goalies have faced 3000 or more shots; here's a look at the top 25 in SV% over that span:

Goaltender Goals Against Shots Against SV%
Cory Schneider 236 3353 .930
Tuukka Rask 356 4898 .927
Henrik Lundqvist 632 8260 .923
Tim Thomas 447 5342 .923
Tomas Vokoun 444 5208 .921
Ryan Miller 679 7814 .920
Pekka Rinne 555 6929 .920
Jaroslav Halak 442 5368 .918
Jimmy Howard 602 7311 .918
Roberto Luongo 567 6953 .918
Carey Price 640 7782 .918
Sergei Bobrovsky 371 4462 .917
Kari Lehtonen 547 6583 .917
Antti Niemi 586 7040 .917
Craig Anderson 647 7682 .916
Jonas Hiller 597 7127 .916
Cam Ward 585 6939 .916
Evgeni Nabokov 602 7074 .915
Jonathan Quick 602 7074 .915
James Reimer 326 3824 .915
Semyon Varlamov 466 5484 .915
Ilya Bryzgalov 627 7298 .914
Corey Crawford 428 4976 .914
Marc-Andre Fleury 642 7441 .914
Mike Smith 534 6245 .914

The average SV% of all 43 goalies is .914. Quick has barely been north of that while Miller ranks in the top six, with four of the five goaltenders ahead of him having faced anywhere from 2500 to 4700 fewer shots. Apart from Quick's brilliant 2011-12 season, there just hasn't been a prolonged period of time in recent history during which Los Angeles' starter has outplayed Buffalo's. The primary argument for going with Quick seems to be that he has won a Cup, which was unsurprisingly the first thing cited by Bylsma when announcing the decision.

The problem here is that counting rings and wins just isn't a very effective method of goaltender evaluation. If every goalie in the NHL had the exact same team playing in front of him, it might make sense to determine which ones are better based on their win total or trophy case (even then, I'd be skeptical based on how much variance is at play) but that's not the current setup. Goalies have zero control over how much offense their team generates or how many shots their team allows (outside of the minimal impact made by rebound control or puckhandling), both of which are overwhelmingly major factors in determining wins, losses and championships. The sensible approach is to evaluate the one aspect of the game goalies can control: the rate at which they stop the shots they face.

Which isn't to say save percentage is a flawless metric or subjective evaluation shouldn't be taken into consideration. Analysts like Chris Boyle have done intriguing work in determining team effects on shot quality allowed. But it's impossible to argue with a straight face that the Kings have been a worse defensive team than the lowly Sabres, either this season or any of the past five. If anything, the opposite is true and the gap in talent level between Miller and Quick is even larger than save percentage suggests. As for subjective takes, I'm far from an expert on the technical aspects of goaltending but those who are tend to peg Quick as a vastly more aggressive goaltender than Miller positionally, which can be a detriment on the larger ice surface where generating cross-ice chances is usually a focal point of offenses.

Ultimately, a lot of the deifying of Quick stems from two undoubtedly excellent playoff runs. Those stretches tend to stand out in people's minds (including Bylsma's) while Quick's awful 2012-13 regular season, his struggles this year or the fact that he wasn't much more than a mediocre starter prior to an incredible 2011-12 are largely forgotten. The mythology of clutch convinces fans and executives alike that players who have come up big in key situations before are more likely than others to do it again, regardless of how their overall performance stacks up against those other players. Eric T. of NHL Numbers looked at whether some goalies are streakier than others and found no real evidence of it at the NHL level:

There is very little difference between goalies and perfectly consistent robots, and certainly nowhere near enough difference between the goalies to label one of them streaky and another consistent. Goalies should be evaluated based on how much skill they have demonstrated, not how often we remember them going on a hot streak.

There's nothing wrong with starting Quick against Slovakia, as long as it's part of a plan that involves Miller getting the start against Russia on Saturday and taking over once the elimination round begins. The fact that Bylsma was coy in announcing his intentions regarding future games, as opposed to Mike Babcock who has stated he'll go with Luongo against Austria after Price faces Norway, suggests Team USA's blueprints may differ. Quick could certainly have an excellent tournament and win the U.S. its first Olympic gold medal in men's hockey since Lake Placid but, as long as we're dealing with weighted coins, I'd feel more comfortable flipping the one with a better chance of coming up heads. That's Miller.