Thursday, November 27, 2008

My Definition of a "Money Goalie"

Patrick Roy's number 33 was recently retired by the Montreal Canadiens, prompting journalists everywhere to find a hundred different ways to say, "Patrick Roy was a clutch goalie". One of the most common cliches in this context, of course, is "money goalie".

I think nearly all talk about goalies who are supposedly "money goalies" or are considered "clutch" is a result of the most basic goalie observation bias - the goalie who is in front looks better because every time he makes a save it looks like he is saving the game. Most games simply come down to whatever team had the better scoring chances, but in any close game you can always use hindsight and selective memory to find a save that looks like it "won" the game, and of course this is what gets focused on by the team's fans and by broadcasters and newspaper writers in the game summary. "He made the key saves at key times in the game," they say, or some other similar cliche.

However, this is not unusual. All goalies, including the bad ones, stop at least 85% of the shots against them, even in the most high-leverage situations like late in the third period. If there are 5 minutes to go in a game and your team is up by a goal, the odds are strongly in your favour even with a terrible goalie like Dan Cloutier in net. The team has to just hold the opposition to a few shots against the rest of the way and it is likely they will win based on probability alone. It is even more likely if they manage to add an insurance goal or two. That's why teams have a combined 213-19-32 record this season when leading after two periods. And then of course Cloutier gets named the game's first star for "making the big save", the save that "preserved the win"...

Of course you can also find a number of saves by the losing goalie that kept his team in it, but they aren't as likely to make the highlight reels or create a reputation for the goalie of being the guy who "makes the key saves to keep his team only a single goal down".

Playing goal is a percentages game. The difference between a Hall of Famer and an OK goalie is that the great netminder stops an extra 2-3% of shots, or about 1 out of every 50. It is easy to say that someone like Grant Fuhr stopped the vast majority of the shots against him while his team was up by a goal in the third period, but all goalies stop the vast majority of shots against them in all situations. The investigative question becomes whether Fuhr was actually playing better or whether he was just in that game situation so often that a few timely saves linger in the memory banks long after hundreds of similar scoring chances have been forgotten.

Name anyone often commonly referred to as a "money goalie", and I bet you they played on a good team. In my view, the term "money goalie" is a term that gets applied almost exclusively to goalies who play on very good teams and therefore spend a lot of time playing with the lead. Obviously goalies have some impact on the game, so a great goalie will make it more likely that his team is in the lead, but the rest of the teammates combined have a much greater impact than the goaltender alone.

This mythology usually extends further. People tell stories about Billy Smith and Gerry Cheevers and say things like, "All they ever cared about was winning. They would give up meaningless goals in regular season games, but when the chips were down they were unbeatable." Fine, I don't necessarily buy it, but that is at least plausible that somebody would increase their level of focus and effort when it mattered most. However, here is my problem with this line of reasoning: If you accept the premise that a goaltender is not completely responsible for his team's win/loss record, which I believe any reasonable person would, then it is quite likely that a goalie with such a competitive mindset would nevertheless lose a game or a playoff series, despite his best efforts. After all, goalies like Roy and Fuhr still had more early playoff exits than Stanley Cup victories, even with outstanding teammates around them. How can we be sure that goalies like Gilles Meloche, Cesare Maniago, Gary Smith, Dan Bouchard and Mike Liut didn't have the same focus on winning that Billy Smith and Gerry Cheevers allegedly did? Was it merely because they were never dealt the good fortune of playing on teams stacked with Hall of Famers?

I want to know which goalies "only cared about winning" and "came up big in the clutch" yet played on a mediocre team and had little team success. If we cannot identify anyone who meets that description, then that tells me that clutch play is difficult to identify objectively and that many of the goalies on great championship teams are almost certainly getting too much credit for their contributions to that team success.

3 comments:

Scott said...

I think that you've raised some really interesting points here. One way to look at which goalies may have come up big "in the clutch" is to look strictly at playoff overtime situations. Does the save percentage of some goalies, win or lose, increase significantly in these situations? You would just be comparing their performance in one situation to their own performance in other situations, so it mitigates a lot of the quality arguments. If you could do a shot-quality-neutral save percentage it would work even better. Anyway, that's my suggestion for analysing "clutch play."

ILR said...

Wasn't there a playoff overtime save percentage chart for the Big 5 (usual suspects + Joseph, Belfour)? Off the top of my head, it came out about as expected with no major fluctuations.

Oh, here it is: http://brodeurisafraud.blogspot.com/2007/03/playoff-clutch-performance.html

Brodeur is surprisingly fragile in playoff OT and apart from the immortal Roy, Eddie The Eagle also has an unbelievable save rate. Shot-quality neutral this of course isn't.

Scott said...

Thanks ILR. I hadn't read that one, but that's pretty much what I was getting at. It was a great post too. Plus, the comments are pretty funny in a "laughing at them" kind of way.