Ranking Goalies By Wins


With Martin Brodeur setting the all-time record for wins, the question of whether he is one of the greats or the greatest goaltender of all time has come to the fore of late. I’m not going to answer that question because frankly I don’t think we have a decent statistical measure to compare goaltenders of different generations, and I haven’t spent the last eighty years watching all of the contenders.

Tom Awad of Puck Prospectus is using a complex statistical measure called GVT to rank goaltenders down through history, but leaving aside whether or not his measure is especially effective, I wanted to note one paragraph that he wrote in his article:

When analyzing a goaltender using GVT, the primary contribution of the goaltender is to block shots; wins, shutouts and similar statistics receive no weighting, nor should they. To those who insist on ranking goaltenders by Wins, Shutouts or Stanley Cups, I answer this: Hasek, 389 wins, 1 Stanley Cup (as a #1 goalie); Osgood, 386 wins, 2 Stanley Cups. Nothing against Osgood, but are these really equivalent goaltenders?

There’s a tendency in sports to judge players by championships; Dominik Hasek, for all of his achievements, was often criticized prior to 2002 because people said, “sure, he’s a great goaltender – but he’s never proven he can win it all.”

It’s a garbage argument.

On a 23-man roster, how much impact does one player, even the starting goaltender, have on the results of his team? Olli Jokinen, for example, has yet to appear in a playoff game – is that because he’s a loser or because he’s spent his career playing for the Islanders and Panthers? Roberto Luongo has been a full-time NHL goaltender for eight seasons and during that span he’s never had a save percentage lower than .914. Despite that, he has appeared in the playoffs only once – is it because he lacks the fundamental character to win, or because he has played for lousy teams? I think the answer is obvious.

So congratulations to Martin Brodeur for his 552 wins; it’s an incredible mark and in part a result of being an exceptional goaltender. At the same time, it’s more a result of playing for a consistently good New Jersey team for the bulk of his career; there’s no way he would have reached that mark if he’d spent his career with the Islanders or Panthers.

  • RCN

    But what exactly is GVT? I went to the link and i still couldn't find his explanation?

    It has always seemed to me that goalie performance was the hardest thing to measure in hockey. I could've sworn that one day I saw some stat on the blogosphere that looked at goalie save pct vs quality of shots, where it actually broke down goalies by making an attempt at compensating for the difficulty of shots they faced. Have you seen that one, Jonathan?

    also, in regards to Brodeur playing on the devils all this time. isn't there another stat i read somewhere that looks at how goalies compare with the other goalies on their team? I mean, nothing against brodeur, but clemmenson looked liked an allstar too, and yet he had never really accomplished all that much in his previous nondevil incarnations.

    one more thing, to add a little spice, i have also felt Fuhr was overrated. once again, would love to share with you the blogosphere site where i finally found some statistical back-up for this claim, but alas, i have lost it.

    in summary, my most confident claim is simply that it seems too hard to really compare goalies, and therefore, until we can any claims of "who is better than who?" must be happily and humbly acknowledged as very flawed.

    btw, i'm loving your blog jonathan–keep up the good work!

  • RCN

    Jonathan Willis wrote:

    @ Robin Brownlee:
    But wouldn’t you agree that wins are more dependent on team than statistics like save percentage?

    Not at all – in fact, Brodeur's save percentage is likely hurt by the fact that NJ gives up fewer shots on average. It's much more impressive to have a .910 on a team that allows 24 shots per game, than a .915 on a team that allows 30. Like Robin said, all tending stats are team influenced, not just the wins.

  • RCN

    @ Jason Gregor:
    How many wins did Roy get because of his ridiculously large equipment?

    I like your show and think you're a hell of a writer, but to start rumors regarding his package, in relation to the ref's love of gay porn in the early 90's…. well that's just inspiring to all of us with our own ridiculously large equipment. Kudos to you sir.