Tuesday, August 19, 2008

Goalie Fatigue

I have argued for a long time that games played for goaltenders is an overrated stat. A goalie does not become better than another goalie simply because he played more games. He could potentially be described as more valuable, but there are many factors that determine games played and nearly all of them are out of the goalie's control. That is not to say that the games played stat is useless - it helps us determine whether a performance was likely to be a fluke or not. For example, if we are evaluating whether Ty Conklin is likely to repeat his 2007-08 performance where he posted a save percentage 20 points above his career average, we know the fact that he did it in only 33 games implies that it is much more likely it was a lucky season rather than a true measure of his skill.

The justification for the importance placed on games played seems to be that it is more difficult to play more games. I remain unconvinced that playing additional games is significantly more difficult. I think that virtually every goalie in the NHL is capable of handling the workload. If they weren't, then they would have failed on some lower level (junior, college, minor-league) when they were asked to play 70-75 games a season.

But what do the numbers say about it? Yahoo Sports provides breakdowns of month-by-month stats, so I thought it would be interesting to see if goalies tired as the season went on, which is what would be expected if fatigue is an issue. To ensure that I had a full selection of statistics and that era issues wouldn't come into play, I chose a group of goalies that have played a lot of minutes in recent years and that began their careers relatively recently (e.g. late '90s or later).

Here is the list of goalies in the sample: Roberto Luongo, J.S. Giguere, Marty Turco, Tomas Vokoun, Evgeni Nabokov, Miikka Kiprusoff, Jose Theodore, Nikolai Khabibulin, Rick DiPietro, Marc Denis, Ryan Miller, and Henrik Lundqvist.

I broke down their statistics by month to see what the results were (minutes, win %, GAA, save %, shutouts):

October: 543 GP, 2.71, .905, .497 win %, 28 SO
November: 730 GP, 2.53, .912, .531 win %, 53 SO
December: 774 GP, 2.48, .914, .547 win %, 54 SO
January: 768 GP, 2.58, .910, .521 win %, 53 SO
February: 653 GP, 2.56, .912, .535 win %, 53 SO
March: 817 GP, 2.53, .913, .539 win %, 53 SO
April: 242 GP, 2.52, .913, .566 win %, 14 SO

What do the data show us? That the goalies tended to start slowly in October, and after that kept up a similar level of play the rest of the season. Their best month was December while their worst month was January, but there was only a .004 difference in save percentage and a 0.10 difference in GAA between the two. Even though these goalies played big minutes all throughout the season, they did not get worse as the season went on.

Not only that, but the goalies played the most games and minutes in March, late in the season, and their performance did not deteriorate. In fact, they were more likely to win games and stop the puck in March and April than in any month except for December. The correlation between save percentage and games played was essentially zero.

However, there could be some goalies that deviate from this trend. Perhaps some iron man goalies improve as the season goes on and get better when playing big minutes. If anybody fits those criteria it would be Martin Brodeur, who year after year plays more games than anyone else. However, Brodeur's splits are are virtually identical to the splits of his goaltending peers. October is easily his worst month (2.45/.904). He peaks in January (2.09/.920), and also does well in April (2.08/.919). The rest of the months are pretty similar, ranging from 2.12-2.22 in GAA and .911 to .916 in save percentage (gaps of .10 and .005 respectively, again virtually exactly the same variances as the larger group).

So goalies apparently do not get worse as the season goes along, and Martin Brodeur is no different than anybody else in terms of monthly splits. This suggests that goalie fatigue is not really a major issue.

To look at this a bit further, I looked at the goalies who took on large workloads (75+ games in a season) to see if they tired late in the year.

I compared how well they did in games 71 or later compared to games 1-70. Here is the breakdown of the average results:

Games 1-70: 2.54 GAA, .910 save %, .567 win %
Games 70+ 2.59 GAA, .903 save %, .571 win %

So the goalies did slightly better in the earlier games, however it was not by much. And if there was a fatigue effect, it was again apparently something that affected Martin Brodeur just as much as everyone else - after game 70, Brodeur slipped from 2.18/.918/.595 to 2.20/.913/.588.

So it appears that fatigue may have an effect on goalies, but if so it appears to be a very small one. Furthermore, goalies who routinely play a lot of minutes like Martin Brodeur have not been better at maintaining their level of play late in the season than other goalies facing heavy workloads. The fact that goalies did far better late in the season than they did in the month of October implies that playing a lot of games may actually be beneficial to goalies, in that it is harder to overcome a summer off than it is to overcome playing 70+ games in a season.

This means that playing a lot of games is probably more accurately characterized as a choice, rather than a skill, at least for the large majority of NHL goalies. If this is correct, that means that Vezina Trophy voters have placed an inappropriate weighting on games played in recent years, as they have repeatedly voted for goalies that played more games over others that played fewer games at a high level.

43 comments:

Anonymous said...

I never really bought into the whole fatigue thing for goaltenders. The reality is that while the goalie may be on the ice a full game and he has to pay attention there is a good portion of the game where he's just standing there watching the play at the other end of the ice. And I would argue that like most of the better skaters in the league who play 25 mins per game, a goalies actual "playing time" is not much more than that.

Plus with every athlete there are some who are conditioned better than others, so fatigue may be less of a factor with some who are better conditioned. And I would put Brodeur in with those goaltenders who are better conditioned.

Fatigue is probably more of a concern going into the playoffs for the forwards and defensemen more so than the goaltenders, and just maybe the slight drop off in playoff performance for goaltenders may just be that the players playing in front of them are not as fresh as they were due to the rigors of the regular season.

Tom said...

Fantastic analysis, both in this article and previously. If TV commentators did this kind of homework, we wouldn't have to sit through the agonizing cliches.

www.bostonblueline.blogspot.com

Bruce said...

Q1. What's this?

11-39-50
6-35-41
12-30-42
11-46-57
9-36-45
7-45-52

A1. That would be Nicklas Lidstrom's points totals from his six Norris Trophy winning seasons, IF his minutes were regularized to the mean (which for a defenceman is 20:00 per game). Decent numbers, but hardly way out in front of the pack.

Q2. Wait a minute, you say, Lidstrom plays way more than that, does that make him better, or simply more valuable?

A2. Say what??

The best players play the most, and the same holds true for goalies.

A goalie does not become better than another goalie simply because he played more games.

No, he plays more games because he's the better goalie.

Here is the list of goalies in the sample: Roberto Luongo, J.S. Giguere, Marty Turco, Tomas Vokoun, Evgeni Nabokov, Miikka Kiprusoff ...

Here is the list of the best goalies in the league: (Martin Brodeur), Roberto Luongo, J.S. Giguere, Marty Turco, Tomas Vokoun, Evgeni Nabokov, Miikka Kiprusoff ...

The best players play the most.

there are many factors that determine games played and nearly all of them are out of the goalie's control.

Then why do all the best goalies wind up being the ones who play the most?

The justification for the importance placed on games played seems to be that it is more difficult to play more games.

Say what? "More difficult" doesn't enter into it. If the guy is an outperformer, the more games he can outperform, the better. Oooops, not "better", just "more valuable".

e.g. In 2007-08 Marty Brodeur not only had a Sv% .011 better than the league norm and a GAA 0.43 better, he outperformed at these elite rates for 77 GP, not 41, nearly doubling his advantage in goals saved over an "average" goaltender. To normalize on just the Sv%, if Brodeur had maintained that .920 through just 41 games, he would have saved about 12 goals more than expected by a half-time goalie with a league avg. .909 Sv%; by doing so over 77 games that advantage to his team was -22 goals.

The Contrarian Goaltender said...

The best players play the most, and the same holds true for goalies.

I agree with you in a general sense. But my post is not about comparing 1st liners to 3rd liners, or starters to backups. In that case games played or TOI can do a pretty decent job of telling them apart. It is about comparing the best of the best, and whether extra games played make any difference at all in that analysis. Lidstrom played a lot of minutes, so it's pretty safe to say he is a very good player. But Lidstrom was only 10th in the league in total minutes played last season. Does that mean there were 9 guys better than he was?

Here is the list of the best goalies in the league: (Martin Brodeur), Roberto Luongo, J.S. Giguere, Marty Turco, Tomas Vokoun, Evgeni Nabokov, Miikka Kiprusoff ...

The best players play the most.


Yes, those are the best goalies in the league, which is why I specifically chose them for the sample. A goalie that plays 75 games per season over a decade is better than one who averages 25, that is clear. But what about a goalie who averages 50 games per season over the span? Or 55? 60? Can we still say that the first goalie is better by default?

Look at the above numbers. The fatigue effect is essentially non-existent. All of these goalies played a lot of games every year, yet they didn't get worse as the season went on. Brodeur's relative numbers are very similar to everyone else's. There appears to be no recurring pattern of more games played leading to substantially worse performance.

By my interpretation, it means that good starting goalies could all play 75+ games, if they were only allowed to do so. The fact that some of them have never done so despite far outplaying their backup goalies and ranking among the league leaders in goalie statistics implies that choice is a very significant variable in this equation.

Here's an example of a couple of goalies who had back-to-back seasons of almost identical play. They were established goalies, doing what they always do (compare their career averages), only in the second season they played a lot more games. And it had a huge effect on the way they were rated and seen around the league:

Nabokov, 2006-07: 50 GP, 2.29, .914, no Vezina votes
Nabokov, 2007-08: 77 GP, 2.14, .910, Vezina runner-up
Nabokov, career: 430 GP, 2.37, .911

Fuhr, 1986-87: 44 GP, 3.44, .881, 3rd place in Vezina
Fuhr, 1987-88: 75 GP, 3.43, .881, won Vezina, 2nd for Hart
Grant Fuhr, career: 868 GP, 3.38, .887

Is there any reason to think that Nabokov in '06-07 or Fuhr in '86-87 would have been done much worse with an extra 25 starts? And if not, then why were they "better" the following year?

e.g. In 2007-08 Marty Brodeur not only had a Sv% .011 better than the league norm and a GAA 0.43 better, he outperformed at these elite rates for 77 GP, not 41, nearly doubling his advantage in goals saved over an "average" goaltender. To normalize on just the Sv%, if Brodeur had maintained that .920 through just 41 games, he would have saved about 12 goals more than expected by a half-time goalie with a league avg. .909 Sv%; by doing so over 77 games that advantage to his team was -22 goals.

J.S. Giguere not only had a Sv% .013 better than the league norm and a GAA 0.48 better, he outperformed at these elite rates for 58 GP. But since 58 < 77, Giguere is worse than Brodeur according to the "he played more so he is better" logic?

Why is that again? Because Giguere's coach didn't run him out there every day? Because Anaheim has a far more demanding travel schedule than New Jersey? Because Giguere's backup goalies are talented up-and-comers, not 32 year old journeymen? Because you think he is going to tire out over the extra 19 starts? Because Giguere couldn't possibly sustain such a high pace, given that he was an unrealistic .005 above his career save percentage average, and a whopping .002 ahead of his previous season's performance?

The question is not whether, say, Ty Conklin (.923, 33 GP) is better than Marty Brodeur (.920, 77 GP). We both agree that he isn't and that games played is a strong indicator of that. The question is more like which of these goalies is better:

A: 55 GP, .706 win %, 1.72, .932
B: 73 GP, .623 win %, 2.02, .914

Both played for strong defensive teams. Both were excellent puckhandling goalies. Goalie B won the Vezina Trophy because he had 18 more starts. To me, that logic is insane, and this post outlines the reasons why.

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

CG: Interesting comp between Brodeur/Giguere. Leaving aside such considerations as who played in the higher scoring conference or who had Hall of Fame defenders in front of him and who had Mike Mottau and Johnny Oduya, their percentage stats are comparable. But that is exactly where the extra games of outperformance do matter.

Grossly oversimplifying:

Brodeur: -0.43 GAA * 77 GP = -33 GA
Giguere: -0.48 GAA * 58 GP = -28 GA

Brodeur: +.011 * 2089 SOG = -23 GA
Gigure: +.013 * 1508 SOG = -20 GA

So Brodeur's slightly more modest outperformance over fully a third more games results in more goals actually saved.
***

I'm in full agreement that Turco got hosed in 2002-03. He led the league in all three major "percentage" categories, Win%, Sv% and GAA. Using the same methodology, outperformance over league norms:

Brodeur: -0.51 * 73 = -37 GA
Turco: -0.81 * 55 = -45 GA

Brodeur: +.005 * 1706 = -9 GA
Turco: + .023 * 1359 = -31 GA

... which to my mind says Turco made a bigger difference than Brodeur that particular season.

Of course as you know there are other considerations that go into the Vezina Trophy, and lifetime achievement is one of those things. At that point in his career Brodeur had been shut out of the award despite having posted an enviable career record and some spectacular individual seasons, e.g.

1996-97: 67 GP, .680, 1.88, .927,
led league in points, shutouts and GAA

1997-98: 70 GP, .691, 1.89, .917,
led league in wins and points

... so right or wrong, some of the voters felt he was due to be recognized. You (and I) could argue that Turco deserved that one in '02-03, but if you look at Brodeur's overall career acocmplishments I don't think it's too far out of line that he has compiled four Vezinas over the course of his sparkling career. His is a rare combination of quality and quantity.

The Contrarian Goaltender said...

Bruce, we keep talking past each other in this debate. Yes, Brodeur saved more goals than Giguere. I don't care, because the only reason he did so is that he got 19 extra starts. In my mind, goalies should be rated based on factors of skill, not luck/opportunity/teammates/etc. You may believe playing more games is a skill, but do so you would have to prove that Brodeur possesses an ability to play a high number of games that other good NHL starting goalies like Giguere don't. I thought the above post did a pretty good job of challenging that reasoning.

If Giguere had played in an extra 19 games to match Brodeur's total of 77, we can estimate he would have faced about 494 extra shots. As I showed in the above post, fatigue does not seem to matter, but let's say for the sake of conservatism that Giguere's save percentage drops by .010 in those extra minutes because of fatigue or some other reason. That would leave him at .920 for the season, exactly the same as Brodeur. The difference in goals saved between them would then be less than one goal over the entire season, i.e. a virtual tie. I think it is reasonable to say that if Giguere played in 77 games his performance would likely match or surpass Brodeur's. Do you disagree with that assessment? If so, then I don't know why those "extra games of outperformance" matter at all. To me it is like looking at two defencemen who score at an equal rate on the power play, seeing that the one averaging 4 minutes per game outscored the one averaging 3 minutes per game, and declaring the first guy to be better. He isn't better, he just had more chances to score.

In the Turco vs. Brodeur example, both of them played in over two-thirds of the games that season. That would meet anybody's cutoff for a starting goalie workload. And given that, the evidence that says Turco made a bigger difference than Brodeur in my mind is that .932 > .914, not who has more total goals saved.

You always focus on the number of starts a goalie actually got and use that to calculate the total goals saved, and I keep saying that the number of starts each goalie made DOESN'T MATTER AT ALL (above a certain minimum game threshold). Brodeur played in 77 games, but with a different team/coach/backup goalie maybe he would only have started 65. Giguere played in 58 games, but does that mean he couldn't have played in 75? I haven't seen any proof that goalies get a lot worse by playing in more games. If there is some evidence that they do, I would love to see it because my position on this issue has always been based on what the stats are telling me.

If goalie fatigue doesn't exist, and the evidence suggests that it is at the very least highly overrated, then every team in the league should play their best goalie in every single game. That's what they did back in the day of Sawchuk/Plante/Hall and they handled it fine even without facial protection. It looks to me like New Jersey is better at utilizing their goalies than the other teams in the league. You seem to be giving Brodeur all the credit for having a smart coach and GM, which again doesn't make any sense to me.

Anonymous said...

I suspect the reasons why Brodeur plays so many games per season is that:

1) His backups have generally been poor to mediocre.
2) His team defense has generally allowed few shots/game, therefore enabling him to play more games (vs. if he was facing 35/game).
3) He wants/likes to play a lot of games.
4) He has avoided injuries.
5) His salary is huge.
6) Coaches/GM's, faced with the above points, want him to play 70+ games per yr.

If any of the above points changed, Brodeur would likely play fewer games. Brodeur has had little impact on the first 2 points (other than being better than poor-to-mediocre).

Bruce said...

Bruce, we keep talking past each other in this debate.

That's cuz I insist on talking logic and common sense. Can be a barrier at times. :)

If Giguere had played in an extra 19 games to match Brodeur's total of 77, we can estimate he would have faced about 494 extra shots.

If my aunt had balls she'd be my uncle. Point is, he didn't play those 19 games and Brodeur did. As a result, Brodeur accomplished more than Giguere did in 2007-08, and it's not that close.

As I showed in the above post, fatigue does not seem to matter, but let's say for the sake of conservatism that Giguere's save percentage drops by .010 in those extra minutes because of fatigue or some other reason.

I don't care about fatigue, I care about results. Both players had similar percentage stats, except this one: Brodeur played in 94% of New Jersey's games, Giguere just 71% of Anaheim's. That's a huge difference, whether you think it matters or not.

I don't know why those "extra games of outperformance" matter at all.

Because Brodeur actually played those games, whereas Giguere sat on his padded stats trying to look good in a baseball cap. Thanks, I'll take the guy with 300 innings pitched over the guy with 220. His ERA and WHIP might be about the same but he'll have more wins and shutouts and quality starts and strikeouts, and will have helped his team more over the course of the season.

To me it is like looking at two defencemen who score at an equal rate on the power play, seeing that the one averaging 4 minutes per game outscored the one averaging 3 minutes per game, and declaring the first guy to be better. He isn't better, he just had more chances to score.

Nice analogy, CG. Let's extend it to compare two defencemen on a strictly percentage basis. This from 2006-07, all numbers expressed as per/60:

Player A: 0.35 G, 1.34 A, 1.69 Pts, +1.09
Player B: 0.34 G, 1.52 A, 1.87 Pts, +1.97

... and Player B clearly made more of his ice time. Not merely "as good" but clearly "better". Which is fine until I tell you Player B is Tom Preissing and Player A is Nick Lidstrom. Preissing's coach saw fit to give him just 15 minutes a night, clearly robbing him of a chance to win the Norris Trophy; meanwhile Lidstrom ran up his counting numbers playing that fourth minute on the PP. Yes of course I'm being facetious, the difference between Brodeur and Giguere is not so large, but it is large nonetheless and their other stats are so similar that it's a no-brainer to vote for Brodeur in that comp. X = X, Y = Y, and 77 > 58.

In the Turco vs. Brodeur example, both of them played in over two-thirds of the games that season. That would meet anybody's cutoff for a starting goalie workload.

Yup, just like a defenceman playing 23 minutes is getting top pairing ice time. It still doesn't make him the equal of the guy getting 30 minutes, whatever imaginary "cut-off" standard you wish to apply.

And given that, the evidence that says Turco made a bigger difference than Brodeur in my mind is that .932 > .914, not who has more total goals saved.

As mentioned above, I would've voted for Turco. He made a bigger difference despite playing in fewer games. He got hosed.

I keep saying that the number of starts each goalie made DOESN'T MATTER AT ALL

And no matter how loud you yell it, it does matter, especially when it comes to such things as voting for Vezina Trophy and All-Star teams. The guy who can handle the big workload with a consistently high quality performance is more valuable than a similar performance with medium workload. The first guy will help his team more by wearing a mask than a ball cap, he'll post the bigger underlying numbers, and he will attract votes. Brodeur certainly would have attracted mine in 2007-08.

I haven't seen any proof that goalies get a lot worse by playing in more games. If there is some evidence that they do, I would love to see it because my position on this issue has always been based on what the stats are telling me.

The stats that you actually pay attention to, you mean. You throw out evidence like 77 > 58 because of certain assumptions. Fact is we don't know if Giguere could handle 75 GP, he's never played more than 65 and only once over 60. Maybe his coach(es) know his limitations and have to go with their second choice more often to keep their starter fresh. Nor do we know the long-term effects of playing 70+ games year after year; looking at the evidence some guys start to wear down under such a huge workload, not within a single season necessarily but in subsequent years. Of the guys who have played two such seasons back-to-back, Miikka Kiprusoff, Arturs Irbe and Tommy Salo all saw their play deteriorate thereafter; on the positive side of the evidence Roberto Luongo is a quality heavy-workload goalie (although he struggled down the stretch this past season for perhaps unrelated reasons). And then there's the freak of nature that is Marty Brodeur, who has played 70+ games ten seasons in a row. So let's just assume because he can do it, they all "could" do it??!! Sorry, not buying.


If goalie fatigue doesn't exist, and the evidence suggests that it is at the very least highly overrated, then every team in the league should play their best goalie in every single game. That's what they did back in the day of Sawchuk/Plante/Hall and they handled it fine even without facial protection. It looks to me like New Jersey is better at utilizing their goalies than the other teams in the league. You seem to be giving Brodeur all the credit for having a smart coach and GM, which again doesn't make any sense to me.


Nope, I'm just giving him credit for what he accomplishes. Heavy workload, big (or small where appropriate) numbers every year. He provides his team with a stable, consistent advantage every year and virtually every night.

5) His salary is huge.

Anonymouse: in 2008-09 Brodeur's salary of $5.2 MM represents the tenth largest cap hit in the league among goalies. This for a guy who has won four of the past five Vezina Trophies, who has backstopped a contender year after year, who has won three Stanley Cups, and who is in the process of rewriting substantial portions of the NHL record book. Do you think he is overpaid?

Anonymous said...

5) His salary is huge.

I'm mainly referring to how much greater his salary is as compared to his backups.

He's 10th among all goalies in the NHL? When is his contract up? Will his salary be ranked higher next contract? Didn't he agree to take less $$ to stay with the Devils? He must like playing in that system, facing among the fewest shots/game in the league :)

Anonymous said...

...I'm mainly referring to how much greater his salary is as compared to his backups.... this is added pressure on coaches/GM's to play him 70+ games per season.

Bruce said...

I'm mainly referring to how much greater his salary is as compared to his backups.

Outside of Chicago you see that quite a lot, e.g. Vancouver, Calgary. But let's just stick with the current comp:

J.-S. Giguere -- cap hit = $6.0 MM
Jonas Hiller -- cap hit = $1.3 MM
------------------------------------------
Anaheim goalies cap hit = $7.3 MM

Martin Brodeur -- cap hit = $5.2 MM
Kevin Weekes -- cap hit = $0.7 MM
------------------------------------------
New Jersey goalies cap hit = $5.9 MM

Because Brodeur can be relied on to play virtually all the games, they don't need to spend much on a backup either, leaving the Devils an additional $1.4 MM to pay on other positions. Giguere takes 82% of the goalie payroll to play 71% of the games; Brodeur takes 88% of the (smaller) goalie payroll to play 94% of the games. Accepting for a moment the insane economic system, New Jersey's got a real bargain there. Affordable, long term stability in an extremely important position.

He's 10th among all goalies in the NHL?

Yes.

When is his contract up?

2012. Midway through his "contract" year (2005-06) he quietly re-upped for 6 years at probably $2 MM below market value.

Will his salary be ranked higher next contract?

Well it ought to be, don't you think? At the time I was shocked he signed for less than Roberto Luongo was refusing in Florida. This despite Brodeur having years of seniority and far greater achievements than Luongo, who had never even made the playoffs to that point in his career.

It's even money there will even be a next contract, Marty will turn 40 the year this one expires. I wouldn't be surprised if he just lived up to this commitment and retired. That said, he might be one of those guys who plays to 45, who knows? He sure doesn't seem to be aging very rapidly, for all the mileage on the chassis.

Didn't he agree to take less $$ to stay with the Devils?

Yes. Marty is all about stability, that's what he provides the Devils and that's what the Devils provide him. He's a perfect fit there, why would he want to move? Brodeur appears almost certain to be one of those rare superstars who will play his entire career for a single team, a la Steve Yzerman, Joe Sakic or Nick Lidstrom, three pretty decent comparables in certain respects. Classy, team-first players, all of them.

Looking at the short list of Brodeur's peers, all of Roy, Belfour and Hasek jumped teams at least once, almost always under disruptive circumstances. All three did so to further their own careers, and were successful in that they all won Cups in their new homes, but they weren't always team players in the process. Meanwhile Curtis Joseph's pursuit of the mighty buck over any sort of team loyalty is legendary.

He must like playing in that system, facing among the fewest shots/game in the league :)

I'm sure he does. The system is good for him, and he's good for it. That said, the only two constants in "the system" are Lou Lamoriello and Martin Brodeur, while coaches, captains and checkers come and go.

As for facing the fewest shots per game in the league, the Devils finished a respectable but hardly overwhelming eighth overall in that category in 2007-08, while Brodeur personally faced the fourth most shots of any netminder and made the second most saves. Do you think he is underworked?

Anonymous said...

Interesting. I wonder how NJ's shots against/60 rank, per season, over the past 10, 15 yrs?

I wonder if it is harder to face 30 shots/60 over 55 games (total shots = 1650) then 25/60 over 70 games (total shots = 1750)?

Bruce said...

Interesting. I wonder how NJ's shots against/60 rank, per season, over the past 10, 15 yrs?

While I'll leave it to CG to put an actual number on it, I'll cheerfully concede that the Devils would rank at or near the top of the league over the course of Marty Brodeur's distinguished career. Maybe Dallas comes close, and Detroit's had some sustained success at both ends of the rink, but most teams have had at least a few bad years over the past decade and a half. Lou Lamoriello has built Jersey's success on excellent goaltending and team defence, of that there can be no doubt.

I wonder if it is harder to face 30 shots/60 over 55 games (total shots = 1650) then 25/60 over 70 games (total shots = 1750)?


This former goalie thinks the second example would be much tougher. For one thing, that's 900 more minutes of concentration. Sure your example is only 100 more shots but maybe it's 300 touches of the puck, and 15 more games of your defencemen knowing exactly what their goalie can and will do to make their jobs easier.

And that's without accounting for other things the goalie himself may do to reduce the number of shots allowed. I don't want to get into that again, but I would hope at the very least we could agree that contribution to the flow of play is non-zero for any goalie, and a positive value for Marty Brodeur.

Indeed, quite independent of GP there are many goalies who would prefer those extra 5 shots a night. It's axiomatic that the shot a goalie faces against the flow of play after an extended break from the action is the toughest one to stop. (All the colour guys say so, so it must be true!)

This may be a mild contributor to the observed (by me) effect of how goalies on low shots-allowed teams rarely achieve exceptional Save Percentages. But maybe that's a new topic for another day, eh CG?

Anonymous said...

I wonder if it is harder to face 30 shots/60 over 55 games (total shots = 1650) then 25/60 over 70 games (total shots = 1750)?

An analogy would be a shorter work week... e.g. 4 days per wk x 9 hrs per day (36 hrs per wk), vs. 5 days per wk x 7.5 hrs per day (37.5 hrs per wk). I suppose some people might prefer the second option, even though they work more total hrs, because then each individual day is more manageable. But, some may prefer option 1. (I'll leave aside the issue of wages & salary.)

The Contrarian Goaltender said...

This may be a mild contributor to the observed (by me) effect of how goalies on low shots-allowed teams rarely achieve exceptional Save Percentages. But maybe that's a new topic for another day, eh CG?

I have discussed this issue before in a previous post. The data do not support your perception, i.e. goalies with exceptional save percentages are more likely to have achieved it facing a lower than average number of shots than a higher than average number.

Many others have agreed with me on the finding that there is not a clear linear relationship between shots faced and save percentage. Having said that, there are some teams that have a pattern of allowing fewer shots with a higher shot quality, but they tend to be weaker teams like Tampa Bay rather than New Jersey or Detroit. So it depends a bit on the situation, but in general goalies usually put up good save percentage numbers on teams that are good at shot prevention.

The Contrarian Goaltender said...

That's cuz I insist on talking logic and common sense. Can be a barrier at times. :)

Your logic, as far as I can tell, is that the number of games started matters because the number of games started matters. You haven't even considered variables such as the impact of the talent of the backup goalie, the preference of the coaching staff, the travel schedule of the team, or the evidence I presented showing goalie fatigue to be a non-factor, and always keep coming back to "Brodeur started 94% of New Jersey's games." If I am missing any cogent statistical or logical arguments, please enlighten me.

"If my aunt had balls she'd be my uncle."

If you aren't even open to considering hypotheticals, how do you manage to compare goaltenders, given that every goalie is facing a different quantity and quality of shots with different teammates in a different team system over a different length of career in an ever-evolving NHL with varying league parity, scoring rates, penalty definitions, and tiebreaker rules?

As a result, Brodeur accomplished more than Giguere did in 2007-08, and it's not that close.

Yes, Brodeur accomplished more than Giguere, because he played more games. But Anaheim's goaltending accomplished more than New Jersey's:

Anaheim 47-27-8, 2.14, .922
New Jersey 46-29-7, 2.23, .918

Isn't that really the main goal in a team sport? Now, of course you are going to say that Ilja Bryzgalov and Jonas Hiller are better than Kevin Weekes. Yes they are, and that is exactly the point.

Which is fine until I tell you Player B is Tom Preissing and Player A is Nick Lidstrom. Preissing's coach saw fit to give him just 15 minutes a night, clearly robbing him of a chance to win the Norris Trophy; meanwhile Lidstrom ran up his counting numbers playing that fourth minute on the PP.

You are, presumably, familiar with Behind the Net's Quality of Competition measure?

Nicklas Lidstrom: 0.16
Tom Preissing: -0.04

And thus, the reason why you can't compare playing increased minutes as a skater with playing increased minutes as a goalie. More minutes for a skater generally means more minutes against the opposition's best. Tom Preissing would get destroyed if he started playing a few extra minutes per game against the Iginlas and Crosbys of the world. For a starting goalie in the NHL, more starts do not mean they are playing against harder opponents (in fact, the opposite is more likely). The only possible factor that would make it tougher is fatigue, the effects of which are still uncertain.

And no matter how loud you yell it, it does matter, especially when it comes to such things as voting for Vezina Trophy and All-Star teams.

Of course games played has an impact on Vezina Trophy voting and All-Star Teams. That is obvious. Whether it should or not is another matter. The Vezina is, after all, supposed to be awarded to the goalie "adjudged to be the best at this position", which certainly leaves the door open for interpretation, a lot more so than if it was for "the most valuable" or for "the greatest contribution to his team".

Nor do we know the long-term effects of playing 70+ games year after year; looking at the evidence some guys start to wear down under such a huge workload, not within a single season necessarily but in subsequent years.

This is a good point, and something of a separate issue given that it is more important in terms of evaluating a career rather than a single season. I think it is quite reasonable that goalies will wear down over time. I do think it is more likely that the decline would be related to shots faced rather than games played, just as number of carries is more important for estimating durability of football players than games played.

However, your sample is still too small to draw conclusions from. For example, Kiprusoff has suffered a steep decline in recent years, going from .923 to .917 to .906. But "freak-of-nature" Martin Brodeur also had a similar stretch where he went from .927 to .917 to .906. This is something that would be interesting to investigate further.

Nope, I'm just giving him credit for what he accomplishes. Heavy workload, big (or small where appropriate) numbers every year. He provides his team with a stable, consistent advantage every year and virtually every night.

Which is fair. That is certainly a valuable thing to provide a team. I would just like to point out that J.S. Giguere does exactly the same thing, only 2-3 times fewer per month. And Giguere has done it more consistently than Brodeur (between .911 and .922 every season as a starting goalie in the NHL, while Brodeur has ranged from .906 to .922 over the same span). Giguere also has a better save percentage than Brodeur in both the regular season and the playoffs. Yet I know you are still going to rank Marty far ahead of Giggy because of the difference between 70-75 games per season and 60-65.

I agree it is possible that J.S. Giguere is not capable of playing 75 games per season. However, he played 86 games in 2002-03 including playoffs, and the last 21 included some of the best goaltending in NHL history. He has also played his whole career with starting-caliber backups (Gerber and Bryzgalov, primarily), which undoubtedly factors into the equation.

I am not saying Giguere is better than Brodeur. They play in different team contexts, after all, and even though two different shot quality measures say that Anaheim allowed more difficult shots than New Jersey last season, the defensive talent and excellent backup goalie performance in Anaheim suggest that it is not the league's most difficult place to play goal. I do think Brodeur and Giguere are closer than conventional wisdom suggests. There are many reasons why one would prefer Brodeur, but in my book an extra 10-15 starts per season is far from the #1 reason to do so.

Bruce said...

//But maybe that's a new topic for another day, eh CG?//

I have discussed this issue before in a previous post.


OK, an old topic for another day which happens to be in the past. I'm not surprised you have done something already, which is why I suggested it rather than do a lot of work on it. Thanks for the link.

Not sure I see anything there to gainsay my general point about low-shot goalies and *exceptional* Sv%. They're just kinda average. There's not much to support the assumption that teams that allow fewer shots also allow lower-quality shots. Generally they do the one and not the other.

Kipper was an exception in 2003-04 but in just 38 games and just 966 shots. Turco really did have an exceptional season in 2002-03, and as a legitimate starter, but still 55 GP is not 70. Furthermore, both Kipper and Turco have seen their Sv% come crashing back to Earth (or at least to the league average) after their one transcendent season. So maybe there was some of that Ty Conklin smallish sample size fluke to it.

I guess the danger is if we start putting on too many stipulations on our comps (say >70 GP, <27 SoG/60) and we wind up comparing a guy like Brodeur against himself and then draw whatever conclusion we want and they're basically meaningless. So I'll stop my trip down that road for now ... I have some ideas of things to explore in terms of goalie's workloads and may come back to this topic later in the season.

Bruce said...

I do think Brodeur and Giguere are closer than conventional wisdom suggests.

I agree with this for the most part. There's not that big of a gap. Giguere is one of the league's elite, has a Cup and a Conn Smythe to his credit. He does, however, play fewer games, even missed time in both the 2006 and 2007 playoff runs; but he's a top notch, Top 5 type goalie, no question. Brodeur's major edge on him statistically is generated by those extra 12-20 GP per season, which obviously I put a higher value on than you do. But TOI is a huge measuring stick for players of all positions, and I don't know why goalies would be any different.

Trivia note: Brodeur played in 77 games last year and got 77 decisions; Giguere played 58 GP and got 58 decisions. I thought that was pretty unusual so I checked out a few sample goalie-seasons to confirm that most goalies usually have a handful of no-decisions. Then I found out this odd fact: that the last time Marty Brodeur played in a game in which he didn't get the decision was 2000-01. In the six seasons since he has played 73-78 GP every season, and averaged over 60:00 TOI/GP. Now that's an iron man.

Bruce said...

On the subject of goalie fatigue, I took a look at the 11 seasons in NHL history that a goalie has played over 75 games. I then compared their Sv% against the league average for that season.

Interesting results:

Goalie - - | - Season - | - GP - | Sv% | Sv% +/-
-----------------------------------------------------------
Brodeur - | 2006-07 | 78 GP | .922 | +.017
Luongo - - | 2006-07 | 76 GP | .921 | +.016
Brodeur - | 1995-96 | 77 GP | .911 | +.013
Brodeur - | 2007-08 | 77 GP | .920 | +.011
Fuhr - - - | 1995-96 | 79 GP | .903 | +.005
Irbe - - - -| 2000-01 | 77 GP | .908 | +.005
Nabokov - | 2007-08 | 77 GP | .910 | +.001
Miller - - - | 2007-08 | 76 GP | .906 | -.003
Kiprusoff -| 2007-08 | 76 GP | .906 | -.003
Denis - - - | 2002-03 | 77 GP | .903 | -.006
Ranford - | 1995-96 | 77 GP | .885 | -.013


Brodeur has three of those eleven seasons, and three of the top four by the Sv% metric. In them he has maintained a Sv% some .014 above the league norm. The other eight goalies (one season each) were collectively average against the norm. Only Luongo came close to Brodeur in delivering on both quality and quantity in a huge-workload season.

Anonymous said...

Too bad for Brodeur that he couldn't keep up that kind of differential throughout his career... he's had some mediocre & below-avg yrs as well.

Bruce said...

Too bad for Brodeur that he couldn't keep up that kind of differential throughout his career... he's had some mediocre & below-avg yrs as well.

What a load of rubbish. If you want to count two seasons where Brodeur was all of .002 below the league average as "below average", fine. And if you want to count those two seasons where he was .001 and .003 above the league average as "mediocre", be my guest. (I would call all four of those seasons mediocre by this metric)

At the same time I hope you are able to give credit for the other ten (10) seasons of clearly above-average results.

To review:

-.005 to -.001: 2 seasons
=.000 to +.004: 2 seasons
+.005 to +.009: 3 seasons
+.010 to +.014: 4 seasons
+.015 to +.019: 1 season
+.020 to +.024: 2 seasons


For his career Brodeur's Sv% is +.009 above the league norm. Meaning that based on Sv% alone -- probably his single weakest metric -- he still has allowed 10% fewer goals per shot than the standard set by goalies across the NHL throughout the Dead Puck Era.

Anonymous said...

Why would it be "rubbish" to say, "Too bad for Brodeur that he couldn't keep up that kind of differential throughout his career... he's had some mediocre & below-avg yrs as well." ?

Provide some context to ".001 and .003 above the league avg"... is this mediocre? In a typical year, among goalies with a reasonable number of minutes, where does .001 & .003 above the league avg place a goalie? What percentile?

What percentile is:

-.005 to -.001: 2 seasons
=.000 to +.004: 2 seasons
+.005 to +.009: 3 seasons
+.010 to +.014: 4 seasons
+.015 to +.019: 1 season
+.020 to +.024: 2 seasons

?

What percentile is "+.009 above the league norm"? Is this amazing? An all-time great? Is allowing 10% fewer goals per shot fantastic?

Anonymous said...

(by the way, when dealing with 2,000 shots in 4,000 minute season, & 10,000+ shots in a career, I think that limiting save pct to 3 decimal places - e.g. ".009" - instead of at least 4 oversimplifies things.)

The Contrarian Goaltender said...

What percentile is "+.009 above the league norm"? Is this amazing? An all-time great? Is allowing 10% fewer goals per shot fantastic?

Anonymous: I'd say +.009 would be normally be in the range of "very good starting goalie". Here some other guys who are within a couple thousandths of Brodeur in save percentage over league average according to my quick and dirty calculations: Curtis Joseph, Ed Belfour, J.S. Giguere, John Vanbiesbrouck, Tomas Vokoun, Manny Legace, and Guy Hebert.

Over the last couple of decades, there have been two goalies who have been dominant in the save percentage vs. league average metric: Dominik Hasek (+.020) and Patrick Roy (+.016). That would be the "all-time great" level.

The only other guy to even approach that duo is Roberto Luongo, who is at +.013. Then Brodeur is in a pack of guys that are in the +.007 - +.010 range. Of course this all fails to take into account team factors, which probably benefit Brodeur at least as much as anyone else, but on the surface +.009 is a solid number.

I don't think it is unfair to say Brodeur has had some mediocre seasons, but he has also had some great ones as well. I agree with Bruce that both need to be taken into consideration.

Bruce said...

(by the way, when dealing with 2,000 shots in 4,000 minute season, & 10,000+ shots in a career, I think that limiting save pct to 3 decimal places - e.g. ".009" - instead of at least 4 oversimplifies things.)

Right you are:

Brodeur: .913465
League*: .904557

So the difference is actually just +.0089. Obviously that negates my point.

(* - "league average" expressed as the simple mean of the league wide Sv% of the 14 seasons; due to lockout and expansion, calculating by the actual gross numbers of shots and saves would yield a slightly different result.)

What percentile is "+.009 above the league norm"? Is this amazing? An all-time great? Is allowing 10% fewer goals per shot fantastic?

"Amazing"? "Fantastic"? I'm not sure I'd be that effusive. OTOH I certainly wouldn't be throwing around terms like "mediocre" or "below average" either when describing his body of work. I think the term w.r.t. Marty Brodeur is "routinely excellent". And "workhorse".

Bruce said...

I don't think it is unfair to say Brodeur has had some mediocre seasons, but he has also had some great ones as well. I agree with Bruce that both need to be taken into consideration.

Thanks, CG! For now let's just take those those "mediocre" seasons into consideration and examine them a little closer.

1998-99: Sv% .906 (-.002 to NHL mean)
70 GP (1st in NHL)
39 W (1st)
21 L (16th !)
88 Pts. (1st)
.629 Pt% (4th NHL)
2.29 GAA (T-11th)
***

2001-02: Sv% .906 (.-002 to NHL mean)
73 GP (1st in NHL)
38 W (2nd)
26 L (T-12th)
85 Pts (2nd)
.582 Pts% (8th)
2.15 GAA (6th)
***

1995-95: Sv% .902 (+.001 to NHL mean)
40 GP (T-6th)
19 W (T-5th)
11 L (T-19th)
44 Pts (6th)
.611 Pts% (10th)
2.45 GAA (10th)
(Won Stanley Cup)
***

2000-01: Sv% .906 (+.003 to NHL norm)
72 GP (T-3rd)
42 W (1st)
17 L (T-30th !!)
95 Pts (1st)
.679 Pts% (T-3rd)
2.32 GAA (13th)
***


Hmmm. At or very near the top of the league in Wins, well outside the top (bottom) 10 in Losses in each of these "mediocre" years. Routinely excellent GAA. Yeah yeah I know, I know you don't believe in Wins and you don't believe in GP, but some of us do consider both an important part of the larger picture. Considering how mediocrely he was playing based on that all-important Sv% metric, Brodeur still posted some pretty enviable numbers pretty much across the board in those seasons. Just think of the numbers he might post in a "good" year ... oh wait, you can just look them up, there's lots of examples. :)

Anonymous said...

If I was a GM/coach & I had a goalie who was at or below the league avg in svpct, I don't think I'd want him playing 70+ games... unless the alternative (e.g. his backups) were worse.

Which is the obvious case for Brodeur & the Devils.

A little odd that you'd keep pointing to his wins, pts, GAA etc. even for the the seasons where he had a mediocre/below-avg sv pct. All that indicates is that he played in front of a very good team.... & that if he had an outstanding/great svpct he would have more wins, pts & a lower GAA.

Anyway, my original pt was that it was too bad for Brodeur that the rest of his career hasn't been as personally good as the past couple, where he has been very very good. I don't know of too many all-time great players (at any position) that had at least 4 mediocre/middling/below avg seasons while in their prime (it would be different if we are talking about poorer seasons when one hits the end of a career).

But, during those 4 or so seasons he played a lot of games in front of a team that allowed few shots & outscored the opposition, so had a lot of wins & a low GAA, & therefore he is a fantastic goalie during those seasons. Er, yeah.

Bruce said...

A little odd that you'd keep pointing to his wins, pts, GAA etc. even for the the seasons where he had a mediocre/below-avg sv pct. All that indicates is that he played in front of a very good team.... & that if he had an outstanding/great svpct he would have more wins, pts & a lower GAA.

Anonymous: As one who doesn't put all my faith in a single column of goaltending statistics, what it indicates to me is that the Devils of 1998-2002 were a different team than the notorious "trapping" teams of the Lemaire years. The Devils of Robbie Ftorek and Larry Robinson played more offensively, took more chances, gave up more chances, and relied on their world-class goalie to limit the damage.

In the three years 1998-2001 there were more goals scored (for and against) in the average Devils game than the NHL average, a clear departure from the team's usual defence-first posture that has seen them (well) below the league average in total goals in each of Brodeur's other 11 seasons. Of course even in their high-scoring years they remained below the league average in GA -- they have been at least 20 GA below the league average in each of Brodeur's 14 seasons after never having achieved that standard even once in the previous history of the franchise. But for those three seasons in they were a puck-possession, outscoring type of team. Such teams may or may not allow bunches of shots, but in my observation they do tend to give up a better quality of scoring opportunity when their players are tending towards the other end of the rink.

One predictable result of this altered team philosophy is a lower Sv%, and Brodeur's were indeed merely "average" throughout those years. Yet he continued to rack up very, very good numbers in other categories, both counting numbers like GP and Wins and percentages like GAA and Pts%. That's a successful goalie in my view.

The Devils won the Cup in 2000 and returned the Finals the next year. In 2001-02 the (tired?) team lost its offensive edge, were decent but not outstanding defensively, and Robinson got fired. Kevin Constantine took over mid-season to re-establish the team's defence-first philosophy and the next year Pat Burns rode a re-worked game plan to yet a third Stanley Cup on a team that didn't even score the league average number of goals. Predictably, Brodeur's Sv% soared and his GAA dropped further still.

High-scoring high-risk team or close-to-the-vest defensive club, Brodeur kept on playing, and winning, more games (and Cups) than anybody. He's been doing it for a decade and a half, long enough to earn the stamp "All Time Great" in my hockey passport.

Anonymous said...

In the following example, one goalie is clearly far superior, correct? I mean, how did the 2nd goalie even get the opportunity to play 70 games with those kinds of poor "counting numbers"?

GP 71 70
MIN 4255 4205
W 40 27
L 19 37
T 12 6
PTS 92 60
Win% 0.65 0.43
GAA 2.21 3.10

Anonymous said...

...The first guy looks like an all-time great, while the second guy might be playing in the minors before he knows it & clearly his team needs to upgrade their goaltending... (same year, same league...)... yep, those counting numbers tell most if not all of the story... no need for analysis here...

Bruce said...

This seems to me a different discussion, but in the interest of moving it forward, tell me about their respective Save Percentages. After all, that's the only thing that really matters.

Anonymous said...

Regardless of their Svpct's, I suspect you would deem the first goalie to be better (perhaps far better) than the second... & even if goalie #1's Svpct was the same or less than goalie # 2's, I suspect you'd still attach some sort of highly positive intangibles to #1 as a result of his counting numbers, which would at least bring him to #2's level, if not surpass #2.

Correct?

Bruce said...

My only assumption at this point is that #1 clearly played on a better team than #2. Since the goalies are part of their team, the working hypothesis would be that #1 probably had a better season than #2. But without further context, I'm grasping at straws. Anonymous goalies in anonymous leagues cited by anonymous posters are pretty nebulous territory.

Anonymous said...

But Bruce, you have repeatedly cited MB’s “counting numbers” as proof that he is a good goalie even during seasons when his svpct is at or lower than the league avg...70+ games, 40+ wins, a low GAA... he MUST be a good goalie, even during those mediocre svpct years... If you insist on attaching value to those counting numbers, irrespective of his svpct, then you logically must declare him better than goalie #2, especially if his svpct is not lower than #2’s. After all, he alone was largely responsible for that 40+ win, low GAA season, right? How on earth could he have 40+ wins & a low GAA unless he was (at least) a “good” goalie, right? Plus his team played him in 70+ games! The counting numbers tell us he is good!

The anonymity of the goalies is to prevent subjective judgement as to their abilities. Otherwise, it is easy (but not correct) to pick one over the other by pointing to other seasons as proof of ability.

[My anonymity has nothing to do with anything, although you keep mentioning it in a scornful way, for some reason. (As if declaring myself "Bruce" or "Bob" or "Tracey" means anything, haha!)]

Bruce said...

But Bruce, you have repeatedly cited MB’s “counting numbers” as proof that he is a good goalie even during seasons when his svpct is at or lower than the league avg.

Anonymouse: I explained above why I think his Sv% may have dipped below his established career rate for those few seasons, even as his success rate at winning (the actual objective of a team player) continued unabated. You have not addressed this, you just continue to pound on your single drum that a mediocre Sv% = a mediocre goalie, no matter what other factors may or may not be at play. As long as you continue to hold on to this one stat as a "Single Point of Failure" then we don't have much more to discuss.

Still I'd be interested to know the unrevealed Sv% of those anonymous goalies from that anonympous league to see if there was any actual point to your post.

As for names, there is generally only one Bruce or Bob or Tracey and there may be hundreds or thousands of Anonymice. It's hard to tell you all apart. For all I know you are somebody who uses a different name on another blog, and we have a back history of which I'm not aware. So it's pretty hard to address you respectfully (try as I might) when I can't establish any continuity to you or your arguments.

Anonymous said...

"I explained above why I think his Sv% may have dipped below his established career rate for those few seasons,"

Right, it had nothing to do with Brodeur's effectivness or ability to stop the puck during those yrs, it was his darn team changing strategies in front of him... (even though the shots against/60 was still lower than league avg)

"...even as his success rate at winning (the actual objective of a team player) continued unabated."

Ah yes, despite his team leaving him high & dry (supposedly) he valiantly continued on winning!

"You have not addressed this, you just continue to pound on your single drum that a mediocre Sv% = a mediocre goalie, no matter what other factors may or may not be at play."

But isn't it so obvious that if he had an increased (aka above avg) svpct during those yrs that his "counting numbers" would've been even higher? It's almost as though you believe that although he was not saving shots at an impressive % level, he somehow "knew" when to make the "big save" in order to win the game. E.g. his .906 sv pct is more impressive than someone else's .906.

[By the way, when did you find the time to watch all 240+ NJ games during those yrs when they supposedly drastically changed their system, leaving poor Brodeur to face those tough 22-24 shots/game? And then when did you watch the other 1,000's of NHL games to compare team systems & shot quality?]

I have no problem with recognizing when a goalie has had very good/great seasons, as Brodeur has had recently... but to just look at "counting numbers" & proclaim this as ipso facto evidence of an impressive goalie is more than a bit silly. What if the team only allowed 10 SA/60? (I can see it now: "but, he had 65 wins in 73 games & a .95 GAA!! He's amazing!! The best alltime season ever!! Facing only 10/60 is actually HARDER because of blah blah blah!")

Nope, I don't have another 'name' on any other site. Why would this matter in regards to respectfully addressing people? If you read your past posts, you frequently are over the top in terms of sarcasm & borderline personal insults.

Anonymous said...

I wanted these examples to be as objective as possible; no subjective name-checking leading to preconceived notions of goalie quality. If I had picked “real” goalies in a “real” league, you would’ve declared that obviously Brodeur is better than Goalie X because of their stats in other seasons, Brodeur has won so many games, won 3 Cups, has been an all-star X times whereas Goalie X hasn’t, etc etc etc.

Instead, I created an amalgam of Brodeur’s 98-99 & 01-02 seasons, with a svpct of .906 as he had in both those years. Other assumptions (to mirror his team’s real stats for those years):

23.5 SA/60, Team GF 238, Team GA 190, Team GF/GA = 1.25, TeamPythWinpct .6098, gross-up TeamPythWinpct by 1.067 (factor similar to MB’s 98-99 & 01-02 gross-up) = MB’s WinPct of .650, proportion .650 into W-L-T similarly to his 98-99 & 01-02 stats. This produces a representative "Brodeur season" for 98-99 & 01-02.

Then, I created Goalie #2 who played a similar number of games & minutes, with the identical svpct of .906, except that: his team allowed 33 SA/60, Team GF 220, Team GA 268, Team GF/GA = .8209, gross-up TeamPythWinpct by 1.067 = Goalie #2s WinPct of .430, proportion .430 into W-L-T. I'm sure there are real-world examples of Goalie #2 (Marc Denis a few yrs ago?) but I didn't want a Google search to produce a real name & then a subsequent argument as to why Goalie 2 is clearly worse than Brodeur, etc etc.

This simulates how a .906 svpct goalie would do under two very different team conditions.

Of course, in reality goalies do not often play 70 games for such poor teams… teams who consistently lose (even when the goalie is average or even good) tend to switch up their goalies, panic etc. as the team descends into turmoil, fans revolt, coach/GM fear for their jobs etc.

These goalies are, for experimental purposes, identical. However, I suspect you would refer to Goalie #1 as better since he owns these impressive “counting numbers” -- even when he is no better than Goalie #2. (unless Goalie #1 is not bearing down when "it doesn't matter" aka "making the save when he has to", etc etc.. that kind of B.S. often ascribed to goalies who play on very good teams but don't have impressive svpct's)

Anyway, you've always been quick to point out that Goalie #1 (aka Brodeur) must be good that particular season because of those counting numbers... that they give him the edge over someone with fewer wins, a higher GAA.

The Contrarian Goaltender said...

I decided to look a bit closer at Brodeur's results from 1999-2002 to see if there was some major change in the team dynamic. My conclusion: There wasn't. See my latest post for the details.

Bruce said...

Why would this matter in regards to respectfully addressing people? If you read your past posts, you frequently are over the top in terms of sarcasm & borderline personal insults.

Anonymous: It seems odd to see you use the word "personal". At your suggestion I did go back to one previous exchange (Vezina Trophy 2008) to see how over the top I may have been and found this:

Anonymous:
i cant even read your garbage......goes on and on and on.....
you need a life big time.

Anonymous:
I imagine this is what it's like trying to discuss the theory of evolution & the fossil record with a die-hard creationist, haha.

Bruce:
First of all, if this was my day job, maybe I would do that ... since it's so easy, maybe you want to do it. Oh wait a minute, scratch that idea, you don't even have time to sign up with even a pseudonym. (I hate never knowing which "Anonymous" I may be talking to around here)

Now I'm guessing that you are the second Anonymous and not the first, but as I keep mentioning, it's kinda hard to tell. As it stands, that crack comparing me, an agnostic of scientific bent, to a die-hard creationist was a little over the top too, which might account for my rather touchy response.

Nonetheless, I apologize for any comments which may have been out of line. I don't mind getting into a good kerfuffle, but my intention is to treat everyonbe with respect. I'm sorry that you feel I may have crossed that line.

Han said...

You say in "why aren't there more brodeurs' article when it suits your anti brodeur view that .003 in save percentage is a hug he number and here a .007 save percentage difference in other starters' goalies after 70 games is not huge. That rings alarm bells for me in terms of bias.

My main problem with this article is how you claim that you're not convinced that goalie fatigue does much and then go on to assume these starters on other teams would have put up similar numbers. All you have to is look at goalies' numbers at the beginning of the year and the goalie leaders and see what the numbers usually end up being. There's a reason goalie minimums for games played are observed in NHL stats, because even they know numbers get skewer when a goalie doesn't have to go through the grind of an nhl schedule. Players fatigue over a season, that is human nature. If your post had any truth to it Brian Elliot would be a vezina candidate.

I read the article on goalie fatigue you posted a link to, and I do not agree with your conclusion. You saw the starters' numbers decrease 7 save percentage points and 5 gaa points. You say that's not a big difference but the difference between a goalie who has a 2.20 gaa and a .922 save percentage and one who has a 2.15 and a .929 is a vezina trophy.

Tattered said...

With the current shortened season and a teams playing virtually every other day and at least 1 back to back a week Id like to see this looked at again. Using right now the mid point in the season and the end of the season. One goalie to look at in particular at the mid point is Bryzgalov who has played all but 1 game for his team so far and his team(Philadelphia Flyers) having played the most back to back games of another team in the league. This might be one situation where fatigue may actually be a factor

Hostpph.com said...

Well in that point of view they don't get better but at least players earn a little bit of experience that it can make the difference in the long run.