Thursday, November 12, 2009

The Goalies With the Hardest Jobs

Having seen which goalies had to do the least for their teams to win, let's look at the reverse side of the coin. Which goalies needed to be at their absolute best almost every game to keep their teams ahead on the scoreboard?

1. Ron Low, .920
2. Denis Herron, .915
3. Cesare Maniago, .914
4. Roberto Luongo, .913
5. Marc Denis, .912
6. Gary Smith, .911
7. Doug Favell, .911
8. Greg Millen, .911
9. Kevin Weekes, .910
10. Guy Hebert, .910
11. Bruce Gamble, .909
12. Jeff Hackett, .909
13. Jim Rutherford, .908
14. Gilles Meloche, .908
15. Mike Dunham, .908
16. Tomas Vokoun, .908
17. Mike Palmateer, .907
18. Ron Tugnutt, .907
19. Dwayne Roloson, .906
20. Manny Fernandez, .906

Just a reminder that these numbers are purely based on save percentage, and don't take into account the number of penalties taken or any shot quality effects. Some of these goalies would have therefore in reality had an easier or tougher time of it. Luongo, for example, has faced a high rate of power plays against throughout his career. On the other hand, Roloson and Fernandez wouldn't make the list if special teams were accounted for.

In contrast to the low win threshold group there are no Hall of Famers on this list, although Luongo should be considered to be on a Hall of Fame track by anyone but the most extreme Cup counter. If we extend the list, the first Hall of Famer that we would hit would be Gump Worsley at .904. The only other enshrined goalie with a number above .900 (the league average baseline that all numbers were adjusted to) is Tony Esposito's .901.

One good reason that we don't see many of the top goalies on this list is that good goalies usually end up on the better teams. Worsley would have made the top 10 based on his years in New York and Minnesota only, but he also played a few seasons on some great Montreal teams. As a Sabre Dominik Hasek would have narrowly missed making this list, but playing for the Senators and Red Wings pulled down his overall number.

I was expecting to see Ron Low first on this list since he is famous for having the worst career winning percentage in history. We also see all 3 of the "bad team goalies" named by Ken Dryden in "The Game" show up in the top 7 (Herron, Smith, Favell). If Dryden needed a fourth example he'd probably have picked Meloche, who also makes an appearance.

All of the goalies on this list were good enough to stick around in the NHL long enough to play at least 300 games. Some weren't particularly good, but most were solid goalies in poor situations. I think several of them are probably quite underrated based on their numbers, in particular Maniago, Hebert, Palmateer, and Vokoun. Just a caveat, though, that the observed relationship between team results and goalie results means that more work needs to be done trying to disentangle team effects from goalie performance for the 1960s and 1970s goalies.

18 comments:

Statman said...
This comment has been removed by the author.
Statman said...

Very interestING!

(have you found any discrepancies between the HAG goalies stats & the Tremblay book?)

Anonymous said...

I'M quite surprised Allan Bester didn't make the list.

The Contrarian Goaltender said...

Bester didn't meet my minimum games cutoff, and that's why he didn't make the list. His career adjusted win threshold was .916.

The Contrarian Goaltender said...

Statman: I haven't compared the numbers in detail yet, I'm looking at mostly team level stuff right now so I haven't got into it too deep. I think they match up fairly well though.

The Collector said...

Excuse me but how does 7uongo have one of the hardest jobs in hockey. Since 2006 the Canucks have been one of the premiere defensive teams of the NHL and he is supposed to excell in that environment. The Canucks have had superior defense to the NJ Devils, post lockout, and they completely dominated the game 6 in the semifinals, when Bobby 7ou let in seven goals. The facts just don't add up, here.

The Contrarian Goaltender said...

I guess I can't assume that someone who so cleverly refers to him as "7uongo" would be aware that the first 6 seasons of Roberto Luongo's career were spent in New York and Florida.

Secondly, the Canucks have been right at league average in terms of win threshold with Luongo on their team. That should be fairly obvious to see if I give the Canucks' ranking in goals for and shots against in the last three seasons:

2006-07: 21st GF, 14th SA
2007-08: 23rd GF, 15th SA
2008-09: 11th GF, 10th SA

If you want to make the case that the Canucks were a good team last year, I'm not going to argue with you. But it was still far from a huge failure for the 8th best team in terms of win threshold to make it to the final 8 in the playoffs before losing to a superior opponent.

The Collector said...

What about the Panthers? Florida made an annual playoff run every year up to 2000 with Vanbiesbruck they stopped making the playoffs with 7uongo in net; why is that? In fact they almost got a Stanley Cup with Vanbiesbruck. Any person would conclude that Vanbiesbruck was the better goalie than 7uongo.

The only goaltender you like that I can at all respect, is Hasek, all the others are fly by nights. Tell me what 7uongo, Sieve M8son, Giguerrible, Turc0, et all have done for the hockey world lately. End of the day, Martin Brodeur and Patrick Roy have 7 cups between them, compared to 1 for those four goaltenders your blog promotes.

The Contrarian Goaltender said...

1. If you want to count Cups, there are lots of places on the Internet where you are free to do that, and where people will agree with you. This, however, is not one of those places.

2. If you think I'm promoting Steve Mason and Marty Turco, you obviously haven't read very much on this blog.

3. Florida's annual win thresholds from 1996-2004 are the following:

1995-96: .891
1996-97: .901
1997-98: .908
1998-99: .903
1999-00: .895
2000-01: .919
2001-02: .924
2002-03: .926
2003-04: .922

The Collector said...

1. I do not deny that it is a whole team that wins a cup but one bad player, the goaltender, can cost you the cup, more than any other player. Martin Brodeur, Patrick Roy, even Hasek, put together fine careers and avoided blowing big games the way your hero, 7uongo, has. 7uongo blew the series against the Mighty Ducks, in 2007, because he was calling the referee. He missed the playoffs in 2008 by completely falling off the last ten games. And, we all know what he did in 2009 with the Hawks, it's how he got his name. :-)

2. Mind if I refresh your memory?

http://brodeurisafraud.blogspot.com/2009/10/situation-adjusted-save-percentage.html

"What that doesn't show, however, is that Rinne faced an unusually low number of shots against on the penalty kill. If we multiply their EV SV% by 80% and their PK SV% by 20%, Mason edges ahead .917 to .914."

How is that not plugging Sieve M8son, a goalie who makes 7uongo look downright good in comparison?

3. Of course the team is going to have a higher win threshhold if the goalie plays badly, that is why Luongo never took Florida to the playoffs, so I am not sure what your numbers prove, that is not already common sense.

The Contrarian Goaltender said...

1. Martin Brodeur and Patrick Roy have never blown a big game? Seriously? I mean I can see why it would be hard to remember particular examples, you would have to go all the way back to the last playoff game both of them have played. Especially if your definition of blowing big games is so broad that it somehow includes Luongo's outstanding performance in game 5 of the 2007 WCSF.

2. Interpreting a player's situational context is not "plugging" a player, it is simply pointing out the facts. Steve Mason probably had the most overrated season of any player or goalie in the NHL last season, and I'm not in the least bit surprised by his sophomore slump this season. And yes, you will find that written in many places on this website if you actually take a look.

3. You don't understand win thresholds. Win thresholds have nothing at all to do with the goalie and everything to do with the rest of the team. High numbers mean a weak team, and that's what Luongo had in Florida.

Anonymous said...

1. Last playoff game Marty Brodeur played was April 09 against the Canes, he played a tight seven game series, all the games by one goal, he played damn tight against Carolina, what's your point man?

When 7uongo lost the series to the Mighty Ducks in 2007, did you forget, he was looking up at the ref, trying to get a penalty, and didn't see a baby-soft shot by Neidermayer.

2. You said Sieve M8son deserved the calder over Peka Rinne. I don't follow the Preds at all but I doubt M8son was better than Rinne. He had one lucky year because he had good defense in front of him but come the playoffs, Detroit and Chris Ozzy, another goalie you disrespect, handed his ass to him!

3. I guess I do not understand your chart then, but the Panthers were never a good solid team, they were always a bottom barrel expansionist team, they got lucky a few years because of their great goaltender Vanbiesbruck, but that is it. Vanbiesbruck and Hasek at least carried sh@tty teams, the other goalies you like never did. How about you give Van some credit man?
Promote someone besides just 7uongo and Craig Andersen, Tim Thomas, JS Giguerrible, and Allen Bester (ROTF,LMAO!!! :-D)

The Collector said...

That above post was me man, I wasn't paying attention when I typed in the captcha and hit submit at the bottom, sorries.

The Contrarian Goaltender said...

1. Look, if you're ripping one guy who lets in a stoppable shot after making the save on 56 of the previous 57 shots against the eventual Stanley Cup champions, and not ripping another guy who lets in two stoppable shots in the last 90 seconds of game 7, then you must have some curious criteria.

Perhaps you are unaware that the 2007 Anaheim-Vancouver series was very tight as well, with 4 out of 5 games decided by one goal, 3 of them in overtime (and 2 of those in double OT)?

2. I never said that. I said that when you adjust for situation, Mason had a slightly higher save percentage. I never made any other adjustment for team effects, I never considered the amount of games either of them played, and I never looked at any other possible contributions to winning. As far as I can tell, Rinne and Mason were about even last season, it's pretty tough to pick between them.

I recommend you read this post: http://brodeurisafraud.blogspot.com/2009/04/steve-mason-vezina-candidate.html.

3. Vanbiesbrouck was quite good, sure, both in the 1996 playoffs and in general, you don't have to convince me of that. Yet you're still giving way too much credit to the goalie.

Win threshold says that the skaters on the 1995-96 Panthers were about as good as the skaters on the 2008-09 Canucks. Again, the 2009 Canucks were 11th in goals for and 10th in shots against. The 1996 Panthers were 12th in goals for and 7th in shots against. Both teams were pretty solid defensive teams.

nightfly said...

In my rec league it's easier to tell who the bad and good goalies are - since none of the teams carry backups, the league permits a fill-in from anyone else's roster to jump into your net if your guy can't make it. You just look for the guys who get asked the most. Failing that, it's easier to compare how the same guy does when he has the superior team, or the inferior one.

It would be like having 40 games of Ondrej Pavelec (just to grab a guy at random) with Atlanta, plus another 20 of him with Carolina and 20 of him with Chicago. Small sample sizes, but it would be a little something.

Perhaps it would be constructive to run something similar, only with the splits being games where the team was outshot vs. outshot their opponent, and games vs. playoff teams vs. non-playoff teams, and compare those against the league averages... something like ERA+ for baseball (call it GAA+ or sv%+). A keeper's performance in those situations would probably be instructive.

nightfly said...

It would be like having 40 games of Ondrej Pavelec (just to grab a guy at random) with Atlanta, plus another 20 of him with Carolina and 20 of him with Chicago. Small sample sizes, but it would be a little something.

To clarify, since the context is a little weak - that would be all over the same time period, and intermixed; as in on Tuesday you play for your team, Thursday you fill in, Friday your team again, Sunday your team and a couple hours later fill in again. We usually only play one night a week but there are six games that night, I've actually had scheduled doubleheaders, and with fill-ins, I've managed to play four times in a night. Probably why my knees have aged in dog years.

John said...

Ahh, thanks for the memories, CG. I wonder how long that game would have went had Louie had his eye on that.

Wish I had it taped.

John said...

And I forgot to mention that he was signalling the referee, ie, had his glove up. No reason to expect him to not make that save.