Thursday, October 29, 2009

Playoff Wins Are a Bad Stat

It has been quite noticeable in the post-lockout NHL that the teams with playoff success have not generally been the teams with the best goaltending. That hasn't made much of a difference for many hockey fans, who continue to consider goalies with Cup rings to be the most clutch in the league. I'm not saying that goalies like Marc-Andre Fleury, Cam Ward, or Chris Osgood have played poorly in the playoffs. Not at all. However, the difference between them and a bunch of other guys is nothing more than the quality of their teammates.

There have been some goalies who have been very good in the regular season, but have not had the same level of recent playoff team success. Three of the best examples would be Roberto Luongo, Martin Brodeur, and Henrik Lundqvist. I decided to compare the numbers since the lockout for each of these three goalies with the supposedly clutch group of Fleury, Ward and Osgood, to see whether that might explain the discrepancy in their win/loss records. The numbers given are goal support per 60 minutes, shots against per 60 minutes, and save percentage:

Osgood: 3.12, 24.9, .928
Fleury: 2.95, 29.1, .916
Ward: 2.61, 28.7, .917
Luongo: 2.06, 29.9, .930
Lundqvist: 2.43, 28.8, .907
Brodeur: 2.56, 30.1, .917

We can use these numbers to calculate the playoff win threshold for each goalie, that is the save percentage they would need to record for their team to score as many goals as they allowed. Not surprisingly, the top three on the list are the guys who haven't been winning, and the bottom three are the guys who have.

1. Luongo .931
2. Lundqvist .916
3. Brodeur .915
4. Ward .909
5. Fleury .899
6. Osgood .875

Chris Osgood's advantage in Detroit is downright unfair. Marc-Andre Fleury is a fine young goalie, but there are not many other teams that would have won the Cup with a .908 save percentage from their starter. On the other hand, Martin Brodeur is not getting much help from his teammates, and somewhat amazingly Roberto Luongo's team has been outscored in the playoffs despite his .930 save percentage.

If I was about to play game 7 of the Stanley Cup Finals, I'd take any of Luongo, Lundqvist or Brodeur ahead of Ward, Fleury or Osgood. Team records don't matter, only the quality of the individual goaltender.

18 comments:

Anonymous said...

sure you can make the case for luongo and brodeur, but lundqvists loses because he doesnt stop the puck. a .907 save percentage is terrible, especially considering that league average save percentage generally goes up from regular season to post season. his goes down. both brodeur and luongo s save percentage numbers rise, and interestingly enough so do ward, fleury, and osgoods numbers. everyone on that list improves their play when it counts, except for lundqvist who gets significantly worse. the only difference is he gets a free pass in new york because of guys like redden, drury, and rosival being so bad.

Anonymous said...

and before anyone wants to make the excuse that his numbers are skewed because he was deplorable against nj in 06, kind in mind that they were also bad because he gave up 11 goals on 56 shots in the final 3 games of the washington series.

The Contrarian Goaltender said...

There are always more reasons why a team loses than just the play of the goalie. Lundqvist had the second-lowest goal support of any of the goalies, which was another big reason for his team's lack of success. Lundqvist may be at a .907 save percentage, but as I pointed out Fleury was at .908 in the 2009 playoffs and his team won the Cup. The point with win thresholds is that a certain mark is enough to win on one team but not enough to win on another.

Lundqvist has certainly had some poor playoff games, in both 2006 and 2009, but he has also had some terrific performances. You can't mention the last three games in '09 without mentioning the three games that the Rangers won in the series, where Lundqvist gave up just 4 goals on 109 shots and was probably the difference between winning and losing in all of them.

One final issue with Lundqvist in the playoffs is strength of opposition. Goalies on lower seeded teams generally have worse stats than goalies on higher seeded teams. Lundqvist has never started a playoff series on home ice. I also don't think he should be blamed for putting up a .908 against Washington with the way the Caps were playing. After all, Ovechkin and Co. killed Fleury in the next round as well, to the tune of a mere .878 save percentage.

There are reasons to prefer Ward's or Fleury's playoff record over Lundqvist's, but even so I think The King is a better goalie overall so I'd rather have him on my team.

Bruce said...

Hey CG, you've been a busy boy lately. October was a prolific month!

I've been busy myself over at the Copper & Blue. I try not to pimp my own stuff too often but thought you and your audience might be interested in this piece since it's all about goaltending and specifically one of the all-time greats of the position.

Anonymous said...

i still think a lot of these excuses for lundqvist are bogus. you mention the 3 games the rangers won. lets be honest here, game won was hardly lundqvist winning it for them, nearly as much as jose theodore playing miserably and blowing the game. the other 2 games they won, he was certainly solid, however very lucky. washington hit an incredible number of posts, the rangers were blocking an incredible number of shots, and did a very good job keeping quality chances to a minimum. look at game 2. lundqvist stopped 38 shots. but everyone watching that game saw how new york shut down the middle of the ice and kept things to the outside. but the media and fans always seem to do a great job of trashing the rest of the team to hype up lundqvist. he's very overrated and has an above average defensive team helping him out, even though he gets all the credit.

Anonymous said...

I love this blog and you usually get things right, but if you're talking about goalies who are byproducts of their teammates play, Osgood is NOT one of them. He was a wall for Detroit where most goalies would've failed miserably.

Now Fleury, HE is the perfect example of a kid who isn't good at all, and only looked good because of him teammates(who were aided by a league who were dead-set on screwing over Detroit and giving them a Cup anyway).

Anonymous said...

Dude, the Penguins took more penalties than the Wings in the SCFs. Which pretty much obliterates your stupid theory.

Agent Orange said...

The problem with basically all of the statistical analysis you do is that you assume a good team is ALWAYS good. This article is a perfect example of it.

In this article you are stating that because the Red Wings are good Osgood doesn't have to be good to win. We are left to interpret that because he doesn't have to be good he isn't good.

In playoff losses last year the Wings scored 1.75 goals per game. Based on the 25.7 shots against Osgood would need a 0.931 sv% to win these games.

Its also worth noting that in 8 losses the Wings lost a total of 2 games by more than 1 goal. One of those games had an empty netter. In the other game the 2-goal lead goal came with 6 minutes left.

What can we conclude from this? Despite only scoring 1.75 goals/game in their 8 losses the Wings were in literally EVERY game late in the 3rd period. They had a chance to win every single game.

While I respect the performances and talent of Lundqvist, Broduer and Luongo they had some very uneven performances last year (in addition to some great performances). While this was going on Osgood had his team in every game.

Had Osgood gotten 3 goals per 60 minutes in his losses his record would have been 5-3 (I gave the 4-2 loss with an empty netter to the L column).

Anonymous said...

I agree. Osgood had a poor regular season but he clearly took it up a notch (or five) in the playoffs last year. The idea that he didn't have to be good to win is false--one needs look no further than Manny Legace in 2006 to see that. He had an easier job than Jonas Hiller or Simeon Varlamov to be sure, but with the exception of one game (the last one of the Columbus series), he was anywhere from solid to outstanding in every single outing and gave a Wings team that was not as good as the 2008 one a very good chance every night.

If he had performed similarly to Luongo (in 2009) over the Stanley Cup Finals, the Pens would have swept Detroit. Pittsburgh completely outplayed Detroit in the first two games and should have scored at least four goals in each. Don't forget his performance in the first period of game 5 also, which was similarly wholly dominated by the Pens. The fact that the Pens did not sweep the finals, and the series went to seven, is very largely due to Ozzy.

Agent Orange said...

I quickly checked losses for the Wings in 2008 as well. Again this is a pretty small sample size (only 4 games for Osgood) but at this point its what we have. In losses Osgood would have needed a 0.914 sv% to win these games. This is interesting because most consider the 2008 performance by Osgood to have been better than the 2009 but your calculations would lead you to believe otherwise.

A lot of this is due to the defensive performance of the Wings in these 2 seasons. In 2008 the Wings allowed 20.5 SA/60 minutes and in 2009 25.7 SA/60 minutes. This is logical because when comparing the 2008/2009 teams the 2009 team scored a ton more goals. As least one would assume. Averages are based on a per 60 minute calculation

Year GF(All) GF(Losses) GAA(losses)
2008 3.12 1.45 2.48
2009 3.22 1.61 2.75

As stated Osgood would have been 5-3 if he got 3 GF/60 minutes. In 2008 he would have been 3-1 in losses. In 2 playoffs he was 29-12. If he had gotten 3 GF/60 he would have been 35-4.

Its also worth noting that the Wings would have swept the Ducks and would have likely avoided much of the injury trouble that hit them (Dats/Zetts/Hossa/Lids). Literally the Wings top 4 players were injured in the playoffs last year.

To the point. Your analysis assumes that a good team is always good. This is simply not the case. If the Red Wings were always good and just needed a serviceable goalie in net they would have won the last 2 Stanley cups and would have only lost 4 games over the 2 years.

Why is a per 60 GFA important and meaningful for an offense but a per 60 GAA is meaningless for a goaltender? The apparent answer is that GAA is a team number. Where your argument gets out of hand is where you claim that goaltendering isn't a team game. Every aspect of hockey is a team game. Offense, defense and goaltending.

The Contrarian Goaltender said...

"The problem with basically all of the statistical analysis you do is that you assume a good team is ALWAYS good. This article is a perfect example of it."

Where do I assume that? I'm looking at averages. Averages include good games and bad games. I never said there is no variance from game to game. I assumed that everyone reading would understand that obvious point.

You can cherrypick the losses or wins for whatever goalie you want to, and the results are going to be the same: The win threshold is going to be a lot higher in losses than in wins. Why? Because the team scored fewer goals. Simple.

If the Red Wings scored 3 goals in every game that they lost, then their overall goals for average wouldn't be 3. It would be 3.5 or 4 or something like that. All games affect the average.

"In this article you are stating that because the Red Wings are good Osgood doesn't have to be good to win. We are left to interpret that because he doesn't have to be good he isn't good."

Maybe that explains some of your disagreement, because that's a poor interpretation. That does not logically follow from the information presented. Win thresholds have nothing at all to do with a goalie's performance, and merely describe the context. For example, Ken Dryden didn't have to be good to win, but he was still an all-time great.

My point here is that playoff wins are a bad stat to use in evaluating goalies. We need to look further at their performance, at things like save % and other better metrics, to understand their performance.

"This is interesting because most consider the 2008 performance by Osgood to have been better than the 2009 but your calculations would lead you to believe otherwise."

I'd be very surprised if most people agree with that statement. I'd say Osgood was much better in 2009 than in 2008. The Wings defence was a fortress in 2008 and nobody was able to touch them, the Penguins included.

"Where your argument gets out of hand is where you claim that goaltendering isn't a team game. Every aspect of hockey is a team game. Offense, defense and goaltending."

Goaltending is not a team game. There is only one guy out there wearing a glove and a blocker. You can argue that the numbers are dependent on the rest of the team, and I'm not going to argue with you, that's what I've said all over this blog. But team offence is a team game because 18 guys contribute to it and team defence is a team game because 19 guys contribute to it. On the other hand, the goalie stands alone. He's the only guy that does his job, and that's why he should be evaluated separately.

The Contrarian Goaltender said...

"The idea that he didn't have to be good to win is false--one needs look no further than Manny Legace in 2006 to see that."

In the 2006 playoffs the Red Wings scored 2.5 goals per game and allowed 22.8 shots per game. That's an .890 win threshold. Legace's save percentage was .884. I fail to see how Legace's performance proves me wrong. He did worse than his team's win threshold, and his team lost more games than they won.

"If he had performed similarly to Luongo (in 2009) over the Stanley Cup Finals, the Pens would have swept Detroit. Pittsburgh completely outplayed Detroit in the first two games and should have scored at least four goals in each."

You're ignoring the role of the shooters in the equation. Pittsburgh probably should have had 3-4 goals in each of those games, but the reason they didn't certainly wasn't all Osgood. It had a lot to do with the goalposts, bounces and Henrik Zetterberg as well.

Every hockey game has a few chances that are what some in the blogosphere refer to as "25-cent chances", i.e. about a 25% chance of going in. They're from close enough in that if the shooter makes his shot, he scores. All the goalie can really do is fill the net. When you're hot, those shots hit the post or go wide or hit you in the chest. When you're not, the shots keep finding their way around you and into the corners. Part of goaltending is playing the percentages, and that's why you see save percentages bounce all over the place from game to game.

I could just as well turn it around and say that if the Chicago Blackhawk shooters had shot like the Penguin shooters did in the Stanley Cup Finals, then the Canucks would have advanced to the Conference Finals. Sidney Crosby's line scored 1 goal on 48 shots in the 2009 Stanley Cup Final. That was not all because of the play of Chris Osgood.

Anonymous said...

"In the 2006 playoffs the Red Wings scored 2.5 goals per game and allowed 22.8 shots per game. That's an .890 win threshold. Legace's save percentage was .884. I fail to see how Legace's performance proves me wrong. He did worse than his team's win threshold, and his team lost more games than they won."

.890 is a very subpar save percentage, especially for the playoffs, especially for an outstanding team like the '06 Wings. There's no real excuse for Legace not being able to meet it.

"You're ignoring the role of the shooters in the equation. Pittsburgh probably should have had 3-4 goals in each of those games, but the reason they didn't certainly wasn't all Osgood. It had a lot to do with the goalposts, bounces and Henrik Zetterberg as well."

In my opinion, the 2009 Stanley Cup Finals were very similar to the Oilers-Ducks series of 2006. One team pretty decisively outplayed the other, but went down 2-0 almost entirely due to goaltending. Yes, the Oilers caught lucky breaks in that series (i.e. pushing the net loose when Lupul, Selanne or Beauchemin were about to score), but by far the biggest factor was Roloson's show-stopping performances.

For that matter, MAF wasn't really bad in the first two games either, just like Bryzgalov--they just caught a couple of very unlucky breaks at the wrong times. The reason why the Pens clawed back and won, whereas the Ducks did not, is Bryzgalov's confidence was totally destroyed after G2, whereas MAF's wasn't. In games 3 and 4 of both series, Roloson and Osgood did not do as well, enabling the Pens to tie the series and almost allowing the Ducks to, save for Bryz' meltdown.

The Contrarian Goaltender said...

".890 is a very subpar save percentage, especially for the playoffs, especially for an outstanding team like the '06 Wings. There's no real excuse for Legace not being able to meet it."

You're right, and that's the point: .890 is a very subpar save percentage, because the Wings were an outstanding team. Legace did not need to post a high save percentage for Detroit to win. If he had been better than .890, his team would probably have won.

"One team pretty decisively outplayed the other, but went down 2-0 almost entirely due to goaltending."

I simply disagree. My sense was that the Pittsburgh shooters were missing rather than that Osgood was taking goals away from them. Of course that's subjective, everybody's going to see it differently, but that's what I saw and from what I remember most of the hockey fans whose judgment I trust saw it in similar terms. Again, not that Osgood played poorly, I thought he played well, but I didn't see him as the difference-maker like you did.

Anonymous said...

"I simply disagree. My sense was that the Pittsburgh shooters were missing rather than that Osgood was taking goals away from them."

So, the Pens hit four posts over that two-game stretch. I do not see that as anything spectacular. Most hockey games feature a hitting of posts. Jonas Hiller, probably the best goalie of the '09 playoffs, also was the beneficiary of a lot of posts against the Sharks, but nobody brings that up.

The fact remains that the Penguins dominated those first two games in puck possession, shots on goal, and quality of shots, and Ozzy performed many jaw-dropping saves over them (and actually, most of the series).

"Again, not that Osgood played poorly, I thought he played well, but I didn't see him as the difference-maker like you did."

Over those two games, and arguably in the first period of game 5, he certainly was. I will grant you that Detroit was the grittier and defensively tougher of the two teams all series long, but Pittsburgh's dominance in scoring chances, puck possession, and quality of shots was undeniable.

If Zetterberg had not covered the puck when it was loose on top of Osgood's side those two times, it still was not guaranteed that Pittsburgh would have been able to capitalize. As I recall both incidents occured when the games were already out of reach in the third period, 3-1. Likewise, it's far from certain that Anaheim could have won those first two games when Michael Peca and Steve Staios knocked the net off when Joffrey Lupul had an open side. Roli still could have slid over and made the impossible saves just like he did all series. So, I don't think singling out a couple of "close calls" that were seemingly beyond the goalie's control really means that the goalie had no impact.

Anonymous said...


I simply disagree. My sense was that the Pittsburgh shooters were missing rather than that Osgood was taking goals away from them. Of course that's subjective, everybody's going to see it differently, but that's what I saw and from what I remember most of the hockey fans whose judgment I trust saw it in similar terms. Again, not that Osgood played poorly, I thought he played well, but I didn't see him as the difference-maker like you did.


I have to say that after reading this I can no long take any of your conclusions seriously. I will follow the statistical analysis for the sake of debate (not that I agree with it all) but your opinion has lost all credibility for me.

I am now convinced that there are certain goalies who you will never give credit to and others who can do no wrong.

The Contrarian Goaltender said...

That's the thing about subjective evaluations, people see things differently. I wouldn't expect anyone to agree with me 100% of the time, because everyone sees things through their own unique lens. So if you disagree with me, that's fine.

Again, the hockey fans that I respect most tended to agree with me. Here's a Calgary Flames fan (Kent from Matchsticks and Gasoline) from a comment left on this blog after game 2:

"Osgood has been extremely lucky by my eye this year as well."

Listen to what the Pittsburgh Penguins were saying after the first two games, they weren't talking about beating a hot goalie, they were talking about bounces and puck luck.

The other thing, by the way, about subjective opinions is that you remember things incorrectly. In game 2, it was actually 2-1 when Zetterberg made the save on the goal line. If that goes in it's a tie game and who knows what happens.

Anonymous said...

"Listen to what the Pittsburgh Penguins were saying after the first two games, they weren't talking about beating a hot goalie, they were talking about bounces and puck luck."

Yes, and Teemu Selanne said, after the Edmonton series, that "Edmonton didn't beat us, we beat ourselves." Team confidence would be much more furthered by promoting the idea that they just had some bad luck in the first two games than if they internalized that they were stonewalled by a goaltender standing on his head.

Osgood was great against a surprisingly strong Columbus, very good against the Ducks, superb against the Hawks, and arguably at his peak in the first half of the Stanley Cup Finals. Over a single series I might call him a recipient of puck luck, but the evidence is that he was red-hot for virtually the whole playoffs.