Tuesday, November 3, 2009

Dominik Hasek's Decline

It is very difficult to be a great player for a long time at the highest level of hockey. Aging, the force of competition, and overall evolution of the game usually end up gradually pulling them back to the pack, a process that can be accelerated by injuries, coaching strategies or team environments. We have seen some of the league's best players go from special to ordinary because of injuries, Eric Lindros being one of the most extreme examples. When that happens it is obvious to everyone that the player is a shadow of their former selves.

Sometimes the descent is a little more difficult to pick out. There have been players who went from special to merely very good because of injury. Wayne Gretzky's injury in the 1991 Canada Cup has often been used to explain his late career scoring drop, for example. For Gretzky, a "drop" still meant being among the league leaders in assists every year, but in his later career the Great One was not nearly the same goalscorer or even-strength offensive player that he was in his prime.

I think one overlooked example of this latter type of injury was Dominik Hasek's groin injury in 2000. The numbers suggest the injury was what ended the Dominator's peak. Many people consider the 2000, 2001, and 2002 seasons to be part of Hasek's prime, but I don't believe that is correct.

To demonstrate this point, we need to look at the situational breakdowns. There were a couple of things going on in the early '00s that were inflating Hasek's numbers: A league that was gradually becoming more low-scoring, and a Buffalo Sabres team with improved team discipline that took fewer penalties. Let's look at Hasek's even-strength save percentages:

1998-99: .946
1999-00: .923
2000-01: .924
2001-02: .925

We don't have official even-strength save percentages from before 1998-99, but based on my estimates from Hasek's overall save percentage and the number of power plays the Sabres faced I'm quite sure they were much closer to .946 than .925. In 1997-98, for example, NHL.com has the goals against by situation, although the shots faced totals aren't correct. However, if we assume that Hasek faced the same number of even strength shots against in 1997-98 as he would in 1998-99, his even strength save percentage would have been .938. That assumption is unrealistic, given that Hasek played more minutes and faced more shots per game in 1997-98 than he did in 1998-99. Hasek was probably performing at .940+ in terms of EV SV% in '97-98, and he was likely in or near that range for the entire period from 1993-94 to 1998-99.

If old age was the reason for the drop from '90s Hasek to '00s Hasek, then we would have expected a more gradual decline. There was essentially a sudden drop from a high peak to a lower plateau, which fits the pattern of a player reduced by an injury.

Hasek's even-strength numbers recovered post-lockout to the .930 range. Perhaps that year off allowed him to heal some of the old battle scars and revert to the Dominator of old, at least for half a season. After that, another injury at the Olympics and the effects of old age caught up with him, even though Hasek was still an effective goalie pretty much right up until the end.

I have heard it argued that comparing Hasek's Buffalo numbers with his numbers in Detroit show that it is more difficult to dominate as a goaltender on a strong team, but I think that is wrong. Hasek's numbers dropped in Detroit partly because the Red Wings took a lot more penalties than the Sabres, and partly because he was not the same goalie that he was at his peak.

It is not surprising that Hasek ran into injury trouble, given that his style had to have been one of the most physically demanding styles of any goalie to ever play. He usually played a straight butterfly on the first shot, which applies the usual stress on the groin and hips, and then he also added his special brand of post-save moves and close-save tactics that required extreme agility and flexibility. Unfortunately, injuries often follow athletes who play a style nobody else is able to play (think of guys like Orr, Lindros, Forsberg, etc.), and the same thing was true for the Dominator.

34 comments:

Anonymous said...

Maybe Detroit took more penalties than early 00s Buffalo, but they would have had (a) more effective penalty-killing than the Sabres and (b) much better even-strength shot prevention.

overpass said...

Good post, CG. I agree that injuries are an underrated cause of reduced effectiveness in hockey players, and the evidence is certainly strong that Hasek wasn't the same after that groin injury.

Anon - yes to both, but if Detroit is better at shot prevention at ES and on the PK, then a) their percentage of shots against taken on the PP doesn't necessarily change, and b) it won't affect save percentage unless there is a shot quality effect.

Anonymous said...

I was saying that I think his age/injury decline was much more significant than any penalty effect on the Wings. Detroit may have allowed more power plays, but the quality of the average power-play shot they allowed would have been less than what Buffalo allowed.

The Contrarian Goaltender said...

Sure, I agree that the decline in play was the biggest factor, but power plays did have an effect. I don't think there is much to suggest that Hasek's play declined from 2000-01 to 2001-02. As I showed in the post his EV SV% were pretty much identical. Yet his overall save percentage went from .921 to .915. Part of this was because Hasek's PK save percentage also dropped slightly, from .897 to .878, but the main reason was that he had to face more shots on the PK.

Going from a very disciplined team to an undisciplined team can have a save percentage effect of something like .003-.005. The injury apparently cost Hasek more in the range of .015 in overall save percentage, so that was for sure a more significant factor, but with respect to comparing 2001 to 2002 you have to take into account the special teams situation.

Statman said...

Gretzky looked great in the 91 Can Cup, prior to the Cheap Shot Suter incident. I think he was player of the game each game. Back in '94 or '95 we watched some of the '91 Can Cup on tape & noticed that Gretzky seemed much quicker than he was by '94 or '95. He was pumped about the 91-92 season, the Kings having done well in 90-91 & then they picked up Kurri that summer.

But, then the back injury & his dad's stroke & Kurri showing up out of shape after a year in Europe.

Sometimes injuries can inhibit players for awhile, & by the time they are fully healed a year or 2 has gone by & even though fully healthy the player is that much older, & naturally declines. In that sense the decline isn't caused by the injury, but the injury robbed a player of a certain amount of healthy time when they should've been at/near their prime.

Bruce said...

Going from a very disciplined team to an undisciplined team can have a save percentage effect of something like .003-.005.

CG: Turning that around, do you think the goalie himself can have an effect on the (in)discipline of his team?

The Contrarian Goaltender said...

Bruce: Looking at that relationship is on the to-do list. I'd say it is reasonable that there could be a slight effect in terms of fewer penalties against for a goalie who can prevent shots and reduce time spent in the defensive zone. I don't expect it to be anything that has much of an impact in the big picture, though. All the non-save stuff is pretty much about gaining slight edges. However, when two goalies are very similar in save skills those variables can cumulatively become the deciding factor.

Anonymous said...

"Gretzky looked great in the 91 Can Cup, prior to the Cheap Shot Suter incident."

Is this the same Suter who tried to decapitate Paul Kariya in the spring of 1998, and only got four games for it?

Anonymous said...

"In that sense the decline isn't caused by the injury, but the injury robbed a player of a certain amount of healthy time when they should've been at/near their prime."

Not to mention the fact that goaltending was very rapidly evolving in the early '90s and it quickly was becoming tougher to score.

Statman said...

Yep, the same jerk: Gary Suter.

I don't know if by the early 90's goaltending was evolving so much as the goalie equipment (particularly after about 1996) was getting bigger & bigger... & the NHL for some odd reason allowed goalie leg pads to each go from 10" to 12" across. Since the mid/late 90's it has become almost impossible to score on a "Guy Lafleur slapshot" from the far wing.

Scoring also dropped due to an increased tolerance for clutching & grabbing.

I guess a way to estimate how much the improvement in save % is due to goalie equipment size vs. changes in 'style' would be to see how individual goalies' save % changed during this period. I suspect most of the goalies who exhibited a new 'style' would be those early in their pro career who developed the style as kids/teens, vs. those established goalies who might be less likely to change their 'style' part-way through their NHL careers.

Anonymous said...

I think TCG said it best in the Allan Bester post--Bester was one of those goaltenders caught up on the wrong side of the goaltending style revolution, and because of this wasn't given another chance at the NHL level. In the late 80s and early 90s, you suddenly got many goaltenders playing at previously unforeseeable performance levels (St. Patrick, Cujo, Belfour, and eventually Hasek) due to radical new styles of goaltending.

I would offer the opinion that the WEIGHT of the goaltending pads, more than their size, is what made high save percentages much easier. Lighter blockers, combined with much improved application of recent human kinesthesiology into goaltender coaching and the butterfly, suddenly made it a whole lot score even for elite players to light the lamp.

Bruce said...

CG: A somewhat-related question: do you think team/management/coaching philosophies on discipline might have an effect on Sv%? If it's a priority to stay out of the box, there might be a few more five-star opportunities at evens. More high-percentage shots against, balanced against fewer shorthanded situations.

You know which team I'm thinking of here.

The Contrarian Goaltender said...

Bruce: Maybe, I don't know, I'm not really sure what to make of the whole business of shot quality right now. Subjectively, from watching that team, they didn't give up very many five-star opportunities at evens. Minnesota is in the same boat, high EV SV%, low penalties against. Generally, opposing power plays are going to result in better scoring chances than even strength plays will, so I think not taking penalties is the better strategy and is going to lead to better goalie stats.

The Contrarian Goaltender said...

I agree that the decreased weight of goalie pads was a big factor, along with improved technical ability. Long slapshots stopped going in because the goalies got better, not because the pads got larger. The modern butterfly technique entered the league before the pads ballooned, and we saw the guys who were at the cutting edge of the technical revolution have great success (Roy and Hasek being of course the two exemplars).

I'm not saying that larger pads had no effect, they allowed some of the bigger, blocking-style goalies to have some success at the highest level. However, I don't understand why there is so much focus on leg pad width, because that extra inch or two really does not make much of a difference in terms of stopping the puck. Increased use of the butterfly allowed for greater efficiency, of course, but another major change was that goalies in the 1980s relied on their gloves and rarely caught pucks against their chest whereas in today's game you see shooters hitting the goalie's crest 15-20 times a night.

The increased reliance on the chest and upper body to make saves might be the most significant technical development in goaltending over the last two decades, because it greatly increased the amount of blocking surface and meant that goalies started filling a lot more net. That goes back to the equipment weight issue, too, because goalies needed better and lighter chest protectors to be able to play that type of game.

Statman said...

"However, I don't understand why there is so much focus on leg pad width, because that extra inch or two really does not make much of a difference in terms of stopping the puck. "

True, a leg pad of 20inches/72 inches (width of net) = 28% net coverage, vs. 24/72 = 33%... doesn't seem like much of a change... but 2 inches wider x 2 pads = 4 more inches across... watch a replay of some Lafleur slapshots from the wing that just catch the space between the leg pad & the post.. a lot of those wouldn't go in after the mid 90's. I've read Lafleur quoted where he said that the bigger pads means that nowadays players have to try to work the puck in closer to get a decent scoring chance, which of course is harder to do.

Although it's not just the leg pads... the gloves, blockers, shoulder pads, pants circa approx. late 90's to current compared to previously are noticably bigger. Brett Hull said that after the mid 90's goalies didn't so much make saves as they let the puck hit them. Watch all the pucks that are caught near the outer edge of today's ginormous gloves -- 15+ yrs ago that was in the net.

But yeah, equipment weight & size enabled goalies to play different styles, take more chances & cover more of the net. No doubt the change in equipment facilitated the styles you guys are talking about. Goalies used pretty much the same equipment from the 40's or 50's until the 90's so there wasn't much allowance for style changes until recently. I'm know I'm talking to a few goalies on here, but it seems pretty obvious that the big increase in goalie equipment size over the past 20 yrs has more to do with lower scoring/increased SV% than goalies suddenly discovering new techniques/better training to stop shooters. Prior to the mid 90's it was only the non-goalie skaters who trained, practiced, watched video etc?

Anonymous said...

A few comments and questions on the issue of unique style and resultant injuries:

--I don't necessarily think that Lindros fits into the mold of playing a unique style that inevitably leads to injury. He was targeted for head shots again and again because he was widely disliked, and seen as an easy mark (at least based on what I gather; I was young and didn't follow the game in the '90s).

--I think that J-S Giguere arguably should be included, in his stead. He plays an extreme, pure butterfly that relies almost exclusively on lateral leg propulsion to make saves, and has suffered tremendously from groin and hip problems in recent years.

--What about Michael Peca? I think he fits in as well.

--Finally, what is your take on Markus Naslund? Starting in late 2003, he began a very linear dropoff in scoring at a very steady rate. Some Canucks diehards would blame the Steve Moore hit, but I would argue he had already peaked. In five years he went from one of the two best players in the league to an average second-liner. Why did he decline so precipitously?

Statman said...

More than any other NHL'er I ever saw, Lindros kept his head down when he had the puck... probably because as a kid/teen he was so much bigger than anyone else that it didn't matter if he got hit with his head down.

Plus, there might be a genetic pre-disposition to concussions; his brother's career ended very early due to concussions.

Statman said...

Actually, a bad-angle "Guy Lafleur shot" would've meant that leg pads 4" wider (2x2") would've covered e.g. 24/36 = 67% vs. 20/36 = 56%(if the angle means 1/2 the 72" net is available)... if only 1/3 of 72" is available due to wide/bad angle shot: 20/24 = 83% coverage vs. 24/24 = 100% coverage.

Big advantage to having wider leg pads.

Anonymous said...

"More than any other NHL'er I ever saw, Lindros kept his head down when he had the puck... probably because as a kid/teen he was so much bigger than anyone else that it didn't matter if he got hit with his head down."

Yes, this was arguably his downfall, but it's pretty undeniable that he was widely reviled around the league for his attitude. Nobody liked him and everyone wanted to let him have it.

Statman said...

If you mean Lindros' attitude towards not playing in Quebec, or some perceived slight against the fans... I doubt other players cared about that. In fact I bet a lot of non-French players would've rather not played in Quebec City.

If you mean Lindros' general snarly/aggressive on-ice behaviour, then sure other players would be more likely to try to hit/stick him, as he'd do the same to them.

The Contrarian Goaltender said...

If there is a big advantage to having wider leg pads, then how come the average save percentage in the NHL today is the same as it was in a more defensive league in 2002-03, despite a reduction in leg pad width from 12" to 11", as well as a reduction in glove and blocker sizes?

Whether a goalie has 8" or 10" or 12" pads makes essentially no difference at all on a long shot from the wing, because that's a reaction save. Anything above the pads should be caught by the glove or deflected by the blocker or the goalie is doing something wrong. Old goalies used to get beat because of inefficient technique, but even back in the day that was a low-percentage play.

Technique is far more important than equipment size. If size was more important than anything else, every goalie would be 6'6". There are goalies that big out there, but very few of them are in the NHL. Equipment size is what kept someone like Garth Snow in the league, but look at the top guys, goalies like Roy and Hasek, they were wearing the same thing stuff as everybody else in the early to mid-'90s and they were still dominating.

In the 1990s a whole generation of standup goalies got replaced by a generation of butterfly goalies. Guys like Fuhr and Hextall were among the best goalies in the league in the 1980s, and yet they were well below average in the 1990s. This despite everybody switching to bigger and better equipment. If it was just the equipment, then the best guys in the 1980s should have been the best guys in the 1990s as well, correct?

Bruce said...

I think not taking penalties is the better strategy and is going to lead to better goalie stats.

CG: You mean like more Wins? :)

I agree with you generally, but if you narrow the focus to a single key stat like EV Sv% it might be a little worse on a team that preached no penalties.

Statman said...

There has to be an advantage to using larger pads, otherwise why would goalies use them? Imagine how fast/quick a goalie could be with today's light materials applied to the pre-90's size of equipment.

The mandated reduction in pad size in 02/03 was too tame, in my opinion. The pads are still too big. (It's amusing to hear announcers scream "what a save!" when the goalie hasn't even moved as the puck goes into his big glove/pads.) As for the comparison in SV% of today, I haven't looked at the numbers so I don't really have an answer to that.

Why wouldn't having wider leg pads help on a hard, accurate shot that is aimed for just inside the post a couple inches off the ice? It would be a huge help.

Whether goalies of the 80's maintain their #'s/skill/whatever into the 90's would be somewhat dependant on how old they are, correct?

Has anyone seen the picture of Roy 80's vs. Roy late 90's in The Death of Hockey. Big difference in equip't size & therefore net coverage.

A study of goalies who straddled the 'equipment inflation' would be interesting.

Bruce said...

Towards the end of his career it appeared that Roy purchased his jersey and pants at Big Top Tent & Awning.

The Contrarian Goaltender said...

Sure, there is an advantage to larger pads. I just don't think it is a particularly large one. Focusing on area coverage oversimplifies the issue because most saves involve a reaction by the goalie. And the areas that are covered by big equipment also happen to be the areas that are closest to the goalie and therefore require the smallest reaction distance.

If a shooter rips a one-timer from the slot that is 12" off the ice and just inside the post, you're right that it goes in today but would not have gone in five years ago. But that's what, something like 1 shot out of every 500 taken? I think big equipment adds a few points onto a goalie's save percentage, just like being say, 6'3" is an advantage over being 5'11", but the difference between goalies of the late 1980s and goalies of today goes far beyond equipment size.

"A study of goalies who straddled the 'equipment inflation' would be interesting."

OK, you got it. I'm working on something right now that I think demonstrates the point about goalie equipment size. Hopefully we can have a good discussion about it.

Statman said...

I also think that (1) the increasing prevalence of 'clutch & grab' since the... late 80's? mid 90's?... (Crackdown on clutch & grab over the past few yrs reversed this somewhat.)

& (2) increased use of video improved team defensive coverage,

reduced the quality of shots taken, which improved SV%'s.

I just find it hard to believe that for some reason goalies suddenly began to exercise/train better (that is, train more effectively than the training that skaters partake in) & figure out what style will stop shots more effectively; unless we're referring to some styles that are only facilitated by using bigger, lighter equipment.

The Contrarian Goaltender said...

"I just find it hard to believe that for some reason goalies suddenly began to exercise/train better (that is, train more effectively than the training that skaters partake in) & figure out what style will stop shots more effectively; unless we're referring to some styles that are only facilitated by using bigger, lighter equipment."

Whether you find it hard to believe or not, the level of exercise and training done by goalies today is simply light-years ahead of what goalies did 2 decades ago.

Goalies started to train better because they started to receive much better coaching. Throughout most of hockey history, the goalies have been pretty much left on their own to their own devices while the rest of the team got put through its paces. Most coaches didn't understand goaltending play, and to be honest most of them still don't. Having a full-time goalie coach on staff is a relatively modern innovation, and naturally it has led to improvement in goalie training and goalie performance. For example, in the mid-'80s Francois Allaire became the first full-time goalie coach in the history of the Montreal Canadiens, the franchise that has produced more outstanding goalies than any other. Patrick Roy gives Allaire a lot of credit for his own personal success.

The same thing happened in Finland in the late '90s - they invested in goalie coaching, and Finland has since become a goaltending hotbed. Similar story in Sweden a few years back, they weren't producing any elite goalies. That caused the occasional international embarrassment, heads started to roll, and so they invested resources into top-level training and today we see Lundqvist and Gustavsson in the NHL and a group of great prospects led by Markstrom and Lehner on their way.

Part of this is that better athletes have decided to play goalie. Just compare the way today's goalies skate and handle the puck compared to the guys in the 1980s, for example, that is not just a difference of style or coaching.

There is no question that goalie styles have changed in the last two decades and that the modern style is more successful. You don't have to take anybody's word on that, just watch some Youtube videos of old goalies in action.

This is where an equipment argument could be made, that lighter and more protective equipment allowed them to play a modern style. That's probably partially true, not literally true but more practically true, in the same way that a goalie could have played a butterfly style in 1950 without wearing a mask, yet he would have probably been killed doing so.

I still think that the equipment changed to fit the optimal style, rather than vice versa, or at the very least they developed in tandem. I've gone through the experience of playing a style relying heavily on the down game in old equipment. There's no question it can be done, you just need to adjust the pad setup a bit. You are of course going to be slower, you can't slide as well, and you'll get dinged on a few shots here and there.

The pros are naturally going to demand the best possible equipment and they want it custom to their own specs. The technical revolution being spread by Roy and Allaire was happening at the same time as the improvement in goalie equipment. Roy started endorsing Koho in 1992-93, and one of the conditions of his endorsement was that they hire equipment designer Michel Lefebvre, who Roy had been working with for some time already. It certainly was not the case that some design genius in a lab somewhere turned out a new design and goalies the world over changed their style of play in response.

Anonymous said...

So this begs the question--why were the elite goalies of the '80s, like Fuhr and Hextall, unable to take advantage of the vastly better equipment of the '90s?

Statman said...

"...the level of exercise and training done by goalies today is simply light-years ahead of what goalies did 2 decades ago."


"...better athletes have decided to play goalie."

True, I think for decades the avg goalie was less fit & less athletic than the avg skater, & at the higher levels of hockey goalies probably lagged by a few years getting into serious athletic preparation as compared to skaters. In that sense I think that goalies have only started to catch up, not that they have surpassed skaters in terms of prep. That's why I don't really marvel at goalie prep; it's about time, haha.

"It certainly was not the case that some design genius in a lab somewhere turned out a new design and goalies the world over changed their style of play in response."

True; there really isn't anything difficult about imaginingdesigning a general (approx) 10-25% enlargement of leg pads, pants, blocker, glove, shoulder/elbow pads, chest protector. It ain't rocket science.

The Contrarian Goaltender said...

So this begs the question--why were the elite goalies of the '80s, like Fuhr and Hextall, unable to take advantage of the vastly better equipment of the '90s?

They were able to take advantage of it, actually. It is just that style is more important than size. Fuhr and Hextall both had decent seasons in 1995-96, which was I believe when goalie equipment really started to balloon. They just couldn't sustain it, however. Part of it was aging, as they were both getting up in the mid-30s, but part of it was that the game had passed them by and they just weren't good enough anymore. Teams were better off with a young French butterflyer than with an aging standup goalie, and that would have been true no matter how big the equipment was.

Anonymous said...

Do you have any kind of formula for translating expected save percentages for the '80s into modernity and accounting for better style? Fuhr, Moog, Hextall, and Bester were four of the best goaltenders of the '80s who all more or less failed to make the jump into a modern style and modern SP%s.

Let's just take Bester, a good goalie on a fairly bad team. His best "prime" year was .890, excellent for the pre-butterfly era, especially considering the very high shot quality Toronto allowed. Almost a decade later, on the Dallas Stars, he posted .899, showing that although his play had not declined at all, he had not kept pace with modern methods.

What sorts of general numbers might Bester have put up post-lockout, on a similar quality team (assuming that in order to make it as a starter today, he would have had to have mastered modern goaltending theory)?

The Contrarian Goaltender said...

As far as translating save percentages, I'm using comparisons against average for now, but I think the period between 1985 and 1995 deserves a closer look because this method may not properly assess what is going on. I have a post coming up on this issue, so I won't go into too much more detail now.

Host PPH said...

It is a pity that we can't fight against aging. Because Hasek had a great consecutive seasons.

netha said...

k2k80p7z53 s4v71p4r28 u9j36d4k35 y3q33e6r61 d8o74x4x64 a7l07n7e04