Saturday, April 25, 2009

Shooting High

I am tracking shots based on ESPN's shot charts for the playoffs, but I have also been regularly checking out CBS Sportsline's shot charts because they add another interesting piece of information: Where the shot was targeted on net. For each shot they record one of five target zones: High glove, high blocker, low glove, low blocker, or 5-hole.

I haven't compiled any numbers other than for this year's playoffs, but Hockey Numbers did a post a while back that gives the CBS data broken down by shot location for the 2006-07 season. There seemed to be a lot of shots missing in that sample as the average save percentage was much lower than the official NHL numbers, but the results still conclusively prove what all hockey players know: That shooters are much more likely to score if they shoot high.

It doesn't seem to matter much where a shot is targeted on net from right to left (e.g. high glove or high blocker are about the same), the key difference is whether the shot is high or low. There are of course some goalies who are better on the blocker side or on the glove side, and all goalies probably have slightly less success on five-hole shots than shots that are low to either side, but those differences are small compared to the difference between top shelf and along the ice.

The 2006-07 data has a save percentage difference of .054 between high shots and low shots. This year in the playoffs, there has been a .942 save percentage on low shots and an .872 save percentage on high shots, for an even more extreme gap of .070. Those differences make a pretty good case that shot height should be included in measurements of shot quality.

As is unfortunately often the case with real-time stats, however, the CBS reporting system is pretty suspect. They seem to have fixed the earlier problem of missing shots, but there are large variances in high shot frequency from rink to rink. Here are the stats by series:

BOS vs. MTL: .917 low, .857 high, 27% high shots
WSH vs. NYR: .946 low, .868 high, 19% high shots
NJD vs. CAR: .961 low, .907 high, 41% high shots
PIT vs. PHI: .961 low, .800 high, 14% high shots
SJS vs. ANA: .926 low, .933 high, 19% high shots
DET vs. CBJ: .919 low, .840 high, 20% high shots
VAN vs. STL: .961 low, .875 high, 16% high shots
CHI vs. CGY: .928 low, .845 high, 35% high shots

We would expect high shots to be correlated with space on the ice. The more time and space a shooter has, the more likely he is going to be able to shoot high. I wouldn't be surprised that a tight-checking series like Vancouver/St. Louis might have a below-average amount of high shots. However, even in that particular series the number of actual high shots is almost certainly understated. In game 4 in St. Louis every single one of Roberto Luongo's 49 shots against were recorded as being low shots, which is extraordinarily unlikely and seems to be merely a case of an indifferent scorekeeper. I observed several other similar games where all or nearly all of the shots were booked as low shots.

The scorers in New Jersey and Carolina appear to be the opposite, booking too many shots as high. The high shot rate in that series is almost double that of all the other series combined, and the save percentage against high shots is much higher than average. The Calgary series also has an abnormal ratio, but the save percentage on high shots is just .845. Either the shooters are really managing to go upstairs that often, or else Kiprusoff and Khabibulin are doing a very poor job of handling high shots.

In most series both goalies have faced a pretty similar number of high shots. One series stood out by having a large gap between the teams, Detroit's 4 game sweep of Columbus. Based on shot distances and shot locations, it looks like Columbus allowed easier shot quality against than Detroit. Over half (54%) of the shots against Steve Mason came from the point or were perimeter shots, while the same areas accounted for just 37% of the shots against Chris Osgood. As a result, Detroit's outshooting advantage is counterbalanced by a longer than average shot distance. I have Detroit with an expected goals figure just 0.3 ahead of Columbus for the series. That is similar to the shot quality results at Hockey Numbers (Detroit +0.5).

When we consider where the shots were targeted, it becomes a different story. The rate of high shots against Steve Mason was twice as high as the rate against Chris Osgood (26% to 13%). If we adjust only based on the average save percentages for low and high shots, that means we would expect Osgood to have the easier job with a .933 expected save percentage compared to .924 for Mason. If we recalculate the expected goals based on those save percentages, Detroit would have been expected to score 3.4 more goals than Columbus over the 4 game series.

What seems possible is that even though Columbus' shooters were getting into good shooting locations, they were mostly shooting under pressure from defenders. In contrast, Detroit was setting up more open shots, which allowed their shooters to snipe up high against Mason. I must confess I wasn't able to catch any of the games of that series, however, so if you did follow that series then feel free to comment on whether you believe the shot quality figures (both based on ice location and target location) seem correct.

If we adjust for both where the shots were coming from and where they were targeted, based on playoff averages so far, I estimate that Mason had a .933 expected save percentage while Osgood was at .922.

We can use those figures to conclude that the much-maligned Chris Osgood did surprisingly well in round 1, but it wasn't a very good playoff debut for Steve Mason. The stats suggest that the likely Calder Trophy winner had the worst overall performance of any goalie in the playoffs, although at least Jose Theodore ranks below him on a per-game basis.

I think it is pretty evident that shot quality would be improved if the target location of the shot was accurately tracked. By combining that information with where the shot was coming from on the ice, it should be possible to get a more accurate scoring probability. Unfortunately CBS Sportsline's tracking system seems to too untrustworthy to be useful at the moment. To evaluate all goalies on a level playing field it is necessary to standardize the reporting to remove or at least drastically reduce rink reporting bias.

26 comments:

jl said...

hey dude - i probably have a bit too much time on my hand, but i watched every game of the detroit series ... here is a breakdown of the goals, from what i remember. it does seem that an unusual proportion of the goals scored on mason went high, and most of the goals scored on osgood came from the slot/crease area, which is also consistent with your thoughts. by my count, the wings scored 5 times from the point or perimeter, and the BJs not at all.

Game 1

10:48Jiri Hudler
Assists: Valtteri Filppula, Mikael SamuelssonRed Wings 1-0

Rink Location: slot
Shot location: low, middle/right side of goaltender
Comments: Mason was not in position to make the save – it was a 2 on 1 and filpulla made a nice head fake. Hudler essentiall shot into an empty net

11:40R.J. Umberger
Assists: Jakub VoracekTied 1-1

Rink location: crease
Shot location: think this went low to the left side of osgood
Comments: off of a kronwall turnover, umberger received the pass in the crease and shot it backhanded. Osgood may have been interfered with just a bit by umberger who was in the crease before receiving the pass.

14:21Jonathan Ericsson
Assists: Jiri Hudler, Valtteri FilppulaRed Wings 2-1

Rink location: point
Shot location: high (not sure which side)
Comments: this is the puck that malhotra(?) tried to glove down instead of letting mason play it. The player’s glove may have actually tipped it lower.

15:09Niklas Kronwall (power-play)
Assists: Mikael Samuelsson, Johan FranzenRed Wings 3-1

Rink location: point
Shot location: unsure (think high though)
Comments: another deflection, this time off of the thigh or leg of a BJ

2:54Johan Franzen
Assists: Henrik ZetterbergRed Wings 4-1

Rink location: crease
Shot location: high, right side of mason
Comments: a wrap around that beat mason on the short side

Game 2
13:33Brian Rafalski (power-play)
Assists: Nicklas Lidstrom, Tomas HolmstromRed Wings 1-0

Rink location: point/perimenter – really on the far side of the facoff circle
Shot location: high, unsure which side
Comments: Mason was screened by Holmstrom

7:18Pavel Datsyuk (power-play)
Assists: Marian Hossa, Nicklas LidstromRed Wings 2-0

Rink location: slot
Shot location: low, left side of mason
Comments: deflection off of a fake shot/hard pass from hossa

15:30Henrik Zetterberg
Assists: Johan Franzen, Daniel ClearyRed Wings 3-0

Rink location: slot
Shot location: high, right side of mason
Comments: franzen draws two guys to the corner, hands off to Z, who has plent of time/space to pick the high, right side

3:38Jiri Hudler (power-play)
Assists: Niklas Kronwall, Brad StuartRed Wings 4-0

Rink location: perimeter
Shot location: right side of mason, about halfway up the goal
Comments: shot from a bad angle in the corner bounced off of mason and into the goal (not mason’s best moment of the series)

Game 3
1:07Tomas Holmstrom
Assists: Marian HossaRed Wings 1-0

Rink location: crease
Shot location: didn’t matter, it was sitting on the goal line (low I guess)
Comments: Hossa beats mason, hits the crossbar, puck is sitting on the goal line for Homer

19:14Daniel Cleary
Assists: Johan Franzen, Brian RafalskiRed Wings 2-0

Rink location: slot
Shot location: low to mason’s left side
Comments: Cleary does a 180 and as mason moves left-to-right, clearl puts the puck behind him on the left side.

13:55Henrik Zetterberg
Assists: Daniel Cleary, Johan FranzenRed Wings 3-0

Rink location: slot
Shot location: high, right side of mason
Comments: commodore goes to mug franzen, leaving Z all alone in the slot with plenty of time to pick the high corner.

16:07R.J. Umberger (power-play)
Assists: Rick Nash, Jason WilliamsRed Wings 3-1

I forget how he scored this one, but the wings couldn’t clear the puck on a late PP. game was already decided at this point.. I think it was scored from the slot or crease though (definitely not the point) and off of a rebound from a nash shot.

19:29Henrik Zetterberg
Assists: Daniel ClearyRed Wings 4-1

Empty netter

Game 4
2:58Nicklas Lidstrom (power-play)
Assists: Brian RafalskiRed Wings 1-0

Rink location: Point
Shot location: low, right side of mason
Comments: Mason makes a pad save but the rebound bounces off of Tutyin and back into the net

6:12Kristian Huselius (power-play)
Assists: Rick Nash, Kris RussellTied 1-1

Rink locatoin: slot
Shot location: high, to osgood’s right
Comments: good high shot coming off the rush on the PP

7:09Tomas Holmstrom
Assists: Brad Stuart, Pavel DatsyukRed Wings 2-1

Rink location: crease
Shot location: think he actually chipped this up high from very close in (to mason’s left side)
Comments: rebound off of a stuart shot from the point, homer there to pick up the garbage

10:02Daniel Cleary
Assists: Tomas Kopecky, Valtteri FilppulaRed Wings 3-1

Rink location: crease
Shot location: low, right side of mason
Comments: wrap around that bounces off of mason’s skate and into the net

1:44Rick Nash
Assists: Rostislav Klesla, Kristian HuseliusRed Wings 3-2

Rink location: crease
Shot location: low to Osgood’s left
Comments: a great shot-pass from klesla that nash stretches out to tip in

5:38R.J. Umberger (power-play)
Assists: Raffi Torres, Steve MasonTied 3-3

Rink location: slot
Shot locatoin: low to osgood’s left
Comments: another one coming off the rush, great individual effore by umberger, osgood couldn’t cover his left side

6:59Marian Hossa
Assists: Niklas Kronwall, Brad StuartRed Wings 4-3

Rink location: slot
Shot location: high over mason’s right shoulder (mason had gone down)
Comments: hossa gets his own rebound

11:26Marian Hossa (power-play)
Assists: Tomas Holmstrom, Nicklas LidstromRed Wings 5-3

Rink location: slot
Shot location: high to mason’s left
Comments: no-look pass from behind the net by homer finds Hossa with plenty of time to make his shot

15:45Kris Russell
Assists: Raffi TorresRed Wings 5-4

Rink location: slot
Shot location: osgood’s left, neither high nor low really
Comments: 2 on 1 rush, shot beats osgood far-side

18:04Fredrik Modin
Assists: Aaron Rome, Jason ChimeraTied 5-5

Rink location: crease
Shot location: think this was actually five-hole
Comments: terrible rebound off of a shot from the point, modin left all alone in the crease

19:13Johan Franzen (power-play)
Assists: Jiri Hudler, Niklas KronwallRed Wings 6-5

Rink locatoin: crease
Shot location: low, right side of mason
Comments: Hudler crashes the net with the puck, franzen comes in late and finds the puck sitting there for him with half of the net wide open.

JLikens said...

Interesting post, particularly the part about Detroit generating twice the frequency of 'high' shots compared to Columbus.

Based on the figures at hockeynumbers, it would appear that the Wings outperformed their expected GF this season.

Perhaps generating a large percentage of high shots enabled them to do this.

As you said, shot height ought to be factored into any shot quality analysis.

Kent W. said...

Great stuff. Shot height never even really occurred to me when it comes to SQ.

Scott said...

I'd add my voice to those that agree shot height should definitely be considered when looking at shot quality. This is really fantastic work. I am, however, a bit confused by your conclusions about Mason. Could you clarify these two statements for me:

"If we adjust only based on the average save percentages for low and high shots, that means we would expect Osgood to have the easier job with a .933 expected save percentage compared to .924 for Mason."

"If we adjust for both where the shots were coming from and where they were targeted, based on playoff averages so far, I estimate that Mason had a .933 expected save percentage while Osgood was at .922."

Thanks.

The Contrarian Goaltender said...

jl: Thanks for that, that does support the numerical evidence.

Scott: I ran three different adjustments based on the shot data. Shot location only (the typical SQ method), shot height only (the first set of numbers you quoted), and both shot location and shot height (the second set of numbers, the .933/.922).

Mason faced more high shots, but he also faced more long shots, so his shot distribution was probably a little easier overall than Osgood's.

Scott said...

Thanks for clarifying CG. I guess I just figured the expected save percentage to be pretty high at .933. Do you have any idea what would be "normal" for an expected save percentage?

The Contrarian Goaltender said...

I am using the overall averages from this year's playoffs to adjust the numbers, so the numbers might be a bit high because the goaltending has been pretty good so far (.924 league average). Ideally I'd use a multi-year sample but I don't have one, at least not yet. The adjustments will actually be changing as the sample size increases and the overall averages change.

The average should put the numbers a bit more into context, though. My estimates have Mason as facing shots that were 12% easier than average, and Osgood facing shots 3% harder than average.

You are right that Mason's number is pretty high. By shot location he is tied for the easiest shots against, and by shot location and height combined he has faced the second easiest.

Anonymous said...

the washington nyr series was completely changed once washington started shooting high. games 1-4 they were just throwing everything at the net. games 5-7 they started being more selective with their shots, thus their shots/game dropped significantly, but they were almost exclusively shooting high glove; to the point where the nbc announcers where even commenting on it, and thus lundqvist looked terrible. more proof the butterfly style makes it easier for less talented goalies to occasionally put together solid numbers, especially in the regular season.

The Contrarian Goaltender said...

more proof the butterfly style makes it easier for less talented goalies to occasionally put together solid numbers, especially in the regular season.There you go about less talented again. Lundqvist has occasionally put together solid numbers? Really?

Since Henrik entered the league:

Lundqvist: 2.31, .917
Brodeur: 2.32, .917

High shots are not some unique weakness of the butterfly style. High shots are simply more likely to go in. That is true for Lundqvist, that's true for whichever goalies you consider to be talented, that was true for Grant Fuhr, that was true for Ken Dryden, that was true for Terry Sawchuk. I bet it was probably true for Georges Vezina, if anyone was even getting the puck off the ice back then.

The reason that high shots get focused on so much for someone like Lundqvist is that he is so terrific down low. Goaltending is not just about using your gloves - your legs are very important elements, and Lundqvist's are very good. That's talent, just the same as catching a puck is talent.

If someone makes up for a deficit in athleticism through positioning and excellent technique, then I say more power to them. I'm certainly not going to downgrade them because they aren't "skilled".

Having said all that, I do actually agree with you that there are more skilled goalies in the league than Henrik Lundqvist. I don't particularly care though, though, because skill does not equal performance. That is something I have been arguing on this blog since the beginning.

There are thousands of hockey players all around the world who can skate faster and shoot harder than Tomas Holmstrom. I think it is fair to say they are more skilled than he is. But none of them are putting up 20 goal seasons in the NHL. If all you value is skill, then you can have all the pretty boy AHL scorers you want. I'll take the guy scoring goals at the highest level. In the same way, goaltending is a results business.

Speaking of results, you may have noticed a certain highly skilled goalie getting burned last night and in game 6 on a few low shots that a goalie with quick legs and a wide butterfly makes look easy. Someone like Lundqvist does not give up Jokinen's tying goal with 1:20 left. All Brodeur had to do was get his pad down and he would have had it, but he came across like he was paddling down without getting the paddle down, and got beat five-hole. If Brodeur had done a simple butterfly slide, you know, a skill that apparently requires no talent whatsoever, his team would be in the second round of the playoffs.

Anonymous said...

Contrarian,

What is your analysis of the Vancouver-St. Louis series?

Personally, I do NOT think Luongo stole the series, at least not to the degree that everybody else thinks. Even if St. L outshot the Canucks, (a) the Canucks' shots were of higher quality (meaning Luongo faced easier shots) and (b) the Canucks have a more coherent and reliable defense, as shown by their perfect penalty-killing.

As for the other "great" goalies in the first round, I think that by far Jonas Hiller was the best. He faced not only the most, but the best-quality shots against by a considerable margin over any other "hot" goalie. Varlamov was great but he was not seriously tested, facing the worst offensive team in the playoffs in the Rangers. Osgood was not really tested until game 4, and then he was really very mediocre.

Anonymous said...

PS: I don't agree that shot distance is the main factor in evaluating shot quality (in regards to the DET/CBJ series). It is very possible to have a peripheral, weak, or hasty shot from right in the slot or crease (look at San Jose in games 4 and 6) and to have a very good shot from the point area (look at the Getzlaf/Perry goals in that series).

Osgood may have faced more close shots but most of the close shots he faced were peripheral, hurried, or weak owing to the great defensive presence in front of him, plus he wasn't screened the way Mase was for much of the series. A good point, or even blue-line shot, especially if screened, counts more than a point-blank shot that the shooter can't get much on.

Anonymous said...

"Speaking of results, you may have noticed a certain highly skilled goalie getting burned last night and in game 6 on a few low shots that a goalie with quick legs and a wide butterfly makes look easy. Someone like Lundqvist does not give up Jokinen's tying goal with 1:20 left. All Brodeur had to do was get his pad down and he would have had it, but he came across like he was paddling down without getting the paddle down, and got beat five-hole. If Brodeur had done a simple butterfly slide, you know, a skill that apparently requires no talent whatsoever, his team would be in the second round of the playoffs."

Yes, and if he also had the size 38 goalie pads and the XXL chest protector 3 of the 4 goals scored never happen. You say it makes no difference the size of the equipment, but mere inches, maybe even less, made the difference between all of those goals. So really who cares. You live with what got you there, and New Jersey lost that game far before they gave up the lead, and there are far more things in New jersey to worry about then Brodeur. Such as the 3rd line consistently getting nearly 20:00 of ice time, and then defenseman consistently failing to get pucks out of the zone at the end of games.

If Brodeur was in net last night for the Rangers, that top circle glove high wrister from Fedorov is stopped, and same with the 6 mph deflection on the first Capital goal. But who knows, maybe if Brodeur was in there Fedorov throws a pass to Ovechkin for a cross ice one timer. So its fair to say that simply substituting this goalie for that goalie isn't accurate because the game plan obviously changes depending on who is in goal. All I am saying is that there are certain glaring weaknesses with a lot of the so called "elite" butterfly goalies, that in a 7 game series can be targeted, much more so than with a guy like Brodeur because the butterfly style gives away most of the top of the net, and relies on defenseman to keep the puck to areas in which quality shots up high can not be taken. Luongo I'd say is the exception, but outside Luongo, Brodeur, and recently Cam Ward, many of the so called "elite" butterfly goalies have been torched because their defenses can not continue to make them look good, and their overall skill levels are poor... Nabakov, Kiprusoff, Lundqvist.

"The reason that high shots get focused on so much for someone like Lundqvist is that he is so terrific down low. Goaltending is not just about using your gloves - your legs are very important elements, and Lundqvist's are very good. That's talent, just the same as catching a puck is talent."

Come on. For somebody who supposedly likes the get to the "why" part of goaltending, I think you are selling yourself short here. Why is he terrific down low? A regulation goal is 72x48. When you horizontally stack 2 pads, that are 38 inches long and 11 inches high, not much is going to go in down low. In fact, it would not be that difficult to actually mathematically break down the advantage a goalie wearing 38's has over a goalie wearing 35's when out lets say 3 feet in the butterfly by simply referring to the percentage of the net that is covered. Especially when the goalie stays far back in the net like Lundqvist does. Top that with a strong Rangers defense that is excellent at keeping shots to the outside, and maybe you'll start to see why Lundqvist gets away with dropping to the butterfly so quickly most of the time.

Statman said...

Bring back the Rogie pads!! haha

http://www.legendsofhockey.net:8080/LegendsOfHockey/jsp/SearchPlayerGallery.jsp?player=18724#photo

http://www.legendsofhockey.net:8080/LegendsOfHockey/jsp/SearchPlayerGallery.jsp?player=18724&photo=44#photo

The Contrarian Goaltender said...

"Osgood may have faced more close shots but most of the close shots he faced were peripheral, hurried, or weak owing to the great defensive presence in front of him, plus he wasn't screened the way Mase was for much of the series."

Yes, this is what I'm trying to get at with looking at whether shots are high and low. I didn't see much of the Detroit series, but I'm not surprised that this was the case.

The numbers don't support your view that distance is unimportant, however. Both Osgood and Mason had much worse save percentages on close shots than on long shots. Overall in the playoffs, goalies have a save percentage of .819 on shots from the crease area and .952 on shots from the point. I do agree that some long shots are very dangerous, and that height should be factored in, but on average the closer the shot the better the scoring chance.

The Contrarian Goaltender said...

"You live with what got you there, and New Jersey lost that game far before they gave up the lead."

That's an interesting interpretation of events. In my book New Jersey wouldn't have lost that game if Brodeur never allowed the other team to score on him and take the lead. He didn't let them tie it up while under pressure in game 5, and he's bailed out his teammates at key times before in his career, as have most good goalies. This time Marty simply wasn't able to pull it off, even though I think he probably should have given the scoring chances that beat him at the end of the game.

"You say it makes no difference the size of the equipment"

I never said that. I said that it is an advantage to be a bigger goalie. I just said I don't really care if a goalie wears huge equipment or not, because as long as it's legal it's simply a matter of choice.

"If Brodeur was in net last night for the Rangers, that top circle glove high wrister from Fedorov is stopped, and same with the 6 mph deflection on the first Capital goal."

You might be right. Every style has a tradeoff. Unfortunately for both Lundqvist and Brodeur, the other team was able to take advantage of their technical limitations at the most inopportune time.

"In fact, it would not be that difficult to actually mathematically break down the advantage a goalie wearing 38's has over a goalie wearing 35's when out lets say 3 feet in the butterfly by simply referring to the percentage of the net that is covered."

If the goalie is set and square to the shooter with appropriate depth, there is not much difference in low coverage. I've worn 32s, 34s, and 36s, so I know that from experience. That is because the net is only 6 feet wide, and with top-of-the-crease positioning a goalie makes himself bigger and reduces the shooting angle even further.

It is easier to close your five-hole with longer pads, but you can make up for that using your stick and a narrower butterfly. The biggest advantage that I see with longer pads is that is easier to reduce holes when sliding or moving across the net. That's what Lundqvist does particularly well with his long pads, quick legs, and deep stance.

In addition, goalies are not set for every shot nor square to every shooter. When I'm talking about Lundqvist's coverage down low, I'm not just talking about him stopping long-range shots that he sees all the way, I'm talking about saves that require lateral movement, extension, sliding, etc. Lundqvist made several toe saves against Washington that required him to react to shots and dekes. We know from his shootout record that he is very tough to deke. And like I said, Jokinen most likely doesn't score that late goal with the same shot against someone like Lundqvist.

All styles have strengths and weaknesses. If a shooter is teeing it up from the slot, I'd rather have Brodeur in net. If a guy comes free on a breakaway, I'd rather have Lundqvist. All the different situational and technical factors are difficult to evaluate subjectively, and that's why I prefer to focus mainly on the results.

Anonymous said...

"All the different situational and technical factors are difficult to evaluate subjectively, and that's why I prefer to focus mainly on the results."

They are difficult to evaluate period, but the does not mean they are not completely relevant. By your own admission, results are often heavily influenced by team play. For instance in the Carolina NJ series, games 2,4,5,7 ended with New Jersey allowing Carolina to maintain possession in the defensive zone late in the game, for periods of at least 2-4 minutes. 3 of the game winning goals where scored because of botched opportunities to clear the zone. Whereas in the Washington NY series, the games in which Washington won, were because of Washington shooters skating in and taking wrist shots short side glove high, through out the game. A good defense can clear the zone, however, every team over the course of a game is going to the opposing team wrist shots from the circle. The key is being able to adapt, and whereas the team can find ways to clear the zone, having a goalie that has shown he is consistently vulnerable in the same spots is much more detrimental. As I have said this is hardly a Brodeur vs Lundqvist thing, as much as it is a perfect example of the styles they play.

As a goalie, I am sure you can confirm that nothing is more fatiguing for a goalie than constant cycling by the attacking team down behind the net. Carolina scored quite a few goals by doing this, which again shows signs of a defensive let down. Again comparing the styles, Brodeur tends to hug the posts, whereas recently goalies have adopted the "slam the skates against each post" method which covers both sides at the same time, and cuts down on scrambling. Both Tim Thomas and Jonas Hiller do this really well, and in this case I'd say it is the better approach.

My whole point, which is where we seem to disagree, is that with all things being equal in terms of team variables, it is much harder to play a non butterfly style. While you seem to only look at the results, I do not know how you can evaluate a goalie and not take skill into consideration. There have to be reasons for the results, and if team elements are removed, I do not see why you would want a goalie with a weakness that can so easily be exploited, especially if the results are somewhat similar. This is not saying all butterfly goalie suck, but it is saying that with all things equal, who would not want the guy with the most talent. Just using common sense, a guy who relies on reflexes to stop pucks, will be a lot more consistent regardless of the team he is on, than a guy who relies on team defense to make his style effective. The problem is, when we refer to a standup goalie in todays game, the sample size is pretty limited as they seem to be a dying breed. Thus using Brodeur as the example to compare to other goalies may be unfair, however IMO, whether the goalie relying on relflexes is good, bad, or average, his results will be more consistent than a goalie reliant on outside circumstances.

Anonymous said...

I never said that distance is not important, just that it isn't the only factor. A point-blank shot from Manny Malhotra or Raffi Torres simply isn't going to be as deadly as one from Zetterberg or the Mule even WITHOUT the world-class D in front of Osgood, let alone with. Most of the Getzy/Perry goals against Nabokov were not point-blank (I don't know how much of the series you watched), but obviously they were very good shots.

I concede that Osgood handily outplayed Mason (and I was really quite surprised to see Mase play as poorly as he did), but I am not sure that the result of the series would have been that much different had the two goalies switched teams. With Ozzy in goal for Columbus facing a Mason Red Wings, the series might have lasted five games instead of four, but Detroit would still have taken it for all the obvious reasons.

FatMan said...

My whole point, which is where we seem to disagree, is that with all things being equal in terms of team variables, it is much harder to play a non butterfly style. While you seem to only look at the results, I do not know how you can evaluate a goalie and not take skill into consideration.Simple: All the skill in the world doesn't mean a thing if it doesn't mean anything in terms of on-ice contribution. A weak goal, at least in my book, isn't excusable because they make a "tough" save later-it's a weak goal, and all the tough save later does is somewhat erase the weak goal. Similar to the Homer comment - there are tons of guys more skilled that can't do what he does, and as some teams (like my Sharks) have shown this year, all the talent in the world doesn't mean s*** if there isn't performance.

There have to be reasons for the results, and if team elements are removed, I do not see why you would want a goalie with a weakness that can so easily be exploited, especially if the results are somewhat similar.As CG has tried to explain with the "Replacements" posts, teams try to cover up the deficiencies. I think we have to remember here that the Rags were a pretty bad team this year, and realistically this series was close to being absolutely stolen by Henrik. A better team may have been able to cut down on the time of the Caps shooters to go glove-side, which is an important consideration to take into account.

This is not saying all butterfly goalie suck, but it is saying that with all things equal, who would not want the guy with the most talent. Just using common sense, a guy who relies on reflexes to stop pucks, will be a lot more consistent regardless of the team he is on, than a guy who relies on team defense to make his style effective.The problem is that this makes it sound like the elite butterfly goalies, like Lundqvist and Luongo as examples, ARE talented. As the post explains, most goalies have trouble with the glove-hand side, and both goalies havemade absolutely amazing saves in the playoffs using there skills and reflexes, such as the Luongo 5v3 PK save in game 1, and the save by Lundqvist yesterday in the 3rd on Semin I believe it was.

The problem is, when we refer to a standup goalie in todays game, the sample size is pretty limited as they seem to be a dying breed. Thus using Brodeur as the example to compare to other goalies may be unfair, however IMO, whether the goalie relying on relflexes is good, bad, or average, his results will be more consistent than a goalie reliant on outside circumstances.I would actually think it's the other way around, that a butterfly goalie would be more consistent, considering he doesn't need his skill to be "on" each day to make his saves. Would like to see an argument against that however, as I don't have much personal goaltending experience, so I don't know quite the relative eases/difficulties of the position :)

Anonymous said...

"I would actually think it's the other way around, that a butterfly goalie would be more consistent, considering he doesn't need his skill to be "on" each day to make his saves."

The problem with this is that it is reliant on the defense in front of him playing to his strength. Lundqvist relies on the Rangers keeping other teams to the perimeter, from which it is i credibly difficult to get off a decent high shot. Thus as every hockey player is taught, they throw it low at the net, hoping for a rebound of tip in. Contrarily, we have all seen what happens to Lundqvist when the Rangers defense is not stellar. Guys walk in and pick the upper corners at will.

What I am saying here, is that over the course of a few seasons or so, you take 2 goalies, one a butterfly goalie, the other a standup goalie, and have them split games for the same team. Then take them both and put both on a another team with a different system. The guy relying on talent, will be more consistent than the guy relying on circumstance, because he is relying on something in his control. Thomas Vokoun is a guy I can think of who is more so a hybrid than a butterfly goalie. His stats when going from Nashville to Florida are relatively consistent. Same with Belfour. Hasek was another guy who relied on reflexes and was consistent where ever he was. Even Osgood has been consistently mediocre. Meanwhile with the modern breed of goalie, a Toskala, or a Huet can be dominant on one team, and then lousy on another. Fluery or Price can be great in one system, and then lousy with another one. Same with Theodore. Same with almost all the so called top tier butterfly goalies outside Luongo who was good in Florida as well as in Vancouver. Thats the theory behind the way Ken Holland runs the Red Wings. Spending money on so called "elite" goalies now is much more of a gamble now then it was decades prior.

The Contrarian Goaltender said...

Anonymous: You are both underrating Lundqvist and overrating the ability of other goalies to make similar saves. If a team gives the opposition time and space to pick corners from the slot, they are going to light up every goalie, not just Lundqvist. You are making one of the league's best goalies sound like a helpless cardboard cutout. Lundqvist does not rely on keeping the other team to the perimeter. When he is playing well, he is a dominant goaltender even if his defence is playing poorly.

According to my playoff shot location data, it was actually perimeter shots that were Lundqvist's undoing this year, not close-in chances. He had a higher than average save percentage against shots from the crease, slot, and mid-range areas, although part of that is the MSG arena effect. He also let in 56 of 62 shots (.903) from the point and perimeter areas, way below the average of .966.

All NHL goalies are good enough to routinely stop perimeter shots, and they are all going to be reacting to those shots which makes style pretty much irrelevant. A goalie letting in a higher-than-normal share of perimeter shots is usually a sign of a goalie who isn't on his game.

To be honest, I don't buy the whole "The Caps were shooting high" thing as a reason for their success. No doubt they were shooting high, and of course shooting high is the advisable tactic against Lundqvist, but he let in some soft goals in games 5 and 6 that he wasn't letting in over the first 4 games. I'd say it's more of a case of a goalie coming back down to earth than a team "solving" a goalie.

"Thomas Vokoun is a guy I can think of who is more so a hybrid than a butterfly goalie. His stats when going from Nashville to Florida are relatively consistent. Same with Belfour. Hasek was another guy who relied on reflexes and was consistent where ever he was."

You have been consistently mixing up your terms. Butterfly goalie does not mean blocking goalie, and hybrid goalie does not mean reflex goalie. Dominik Hasek relied on his reflexes and used a textbook butterfly to make the first save. The best goalies know when to react and when to block and use both tactics effectively.

Lundqvist's problem is not his style, but his glove hand. There are lots of butterfly goalies who are very difficult to beat high glove. For example, Cam Ward made a number of terrific glove saves against the Devils, and the Blues seemed to be shooting blocker side whenever possible against Luongo.

On long range shots, everyone reacts, and on close-in chances, everyone blocks. The areas where goalies differ in their tactical responses is the mid- to high slot and on angle shots, depending on their size and their style. I don't know what percentage of shots this represents, but it is certainly the minority.

About one-quarter of the shots are from what I call the "mid-range" zone, that might be a fair estimate. If so, that works out to maybe a half-dozen shots per game where we get to see the real tactical difference between a Giguere and a Brodeur.

"Meanwhile with the modern breed of goalie, a Toskala, or a Huet can be dominant on one team, and then lousy on another."

Toskala was never dominant, he played on a dominant team. Huet was a late-bloomer, and he hasn't been lousy this year in Chicago. I disagree with your assessment of Belfour's consistency, he had some very poor seasons and twice lost his starting job.

Down years for butterfly goalies are often because of injuries. Hip injuries are very common for goalies who go down on every shot. That's likely the main reason Toskala was so bad this year, for example.

"Thats the theory behind the way Ken Holland runs the Red Wings. Spending money on so called "elite" goalies now is much more of a gamble now then it was decades prior."

No, that's not correct. Holland's view is not because of risk, it is because of parity. He's made that clear in the quotes I've seen from him. Goaltending talent is very deep today, there are lots of decent goalies out there, probably more than there are spots available for them to play in. That means that average talent has little marginal value, since you can easily replace it with someone else.

There are a few elite goalies in the league today, just like there has always been. The best guys in the league have big paycheques, and that has always been the case because top goaltenders are always valuable.

The difference is that there are far fewer bad goalies today. The value of an average goalie is determined by the number of bad goalies - the more bad goalies there are, the more teams don't want to get stuck with one of them and the more value there is in at least having an average guy who won't cost his team games.

If you can get 95% of the performance at, say, 20% of the cost, then why wouldn't you? And that is the reasoning behind Holland's position.

Anonymous said...

I still think you are failing to see the point. There are some butterfly goalies who have great blockers, gloves, reflexes, etc. But there are far too many, as I stated, who do not, and simply drop to their knees the second the puck enters the zone. Yes as I said guys like Luongo or Cam Ward are both highly skilled goalies in those regards, but for every Ward or Luongo, there are 10 Hiller's or Lundqvists, guys who show very little else besides the ability to sit on their knees for 60 minutes, and the ones who are fortunate enough to play behind a strong defense often even get confused for good goalies. If I stuffed a statue between the pipes and it posted a .920 save percentage would you being calling it a great goalie?

FatMan said...

If I stuffed a statue between the pipes and it posted a .920 save percentage would you being calling it a great goalie?

Dunno about others, but that would have been fine with me, as it would have represented a substantial upgrade over what my Sharks had in net against Anaheim. And again with the "skill" comments: Every goalie has trouble with stuff up high. The idea behind being a "butterfly" goalie isn't to cover up a lack of skill; it's to cover up the part of the ice where the majority of shots on net go to, and maximize the chance of stopping the puck from entering the net. What you also seem to fail to realize is that the type of goaltending you describe does, in fact, show it's flaws, THROUGH the statistical evidence, as any flaw in a sort of "b-fly and pray" system will show up as GA, and thus a reduction in SV% and GAA. I remember an example earlier in the season in a MIN-SJS game, where Backstrom was doing the style of goaltending you described. Early on it led to some saves, yes, but as SJ's shooters started to pick on the fact that Backstrom was going down early every play, they were able to pick him apart. His stat-line that day: 18/23 for .783 SV%,with a 4.72 GAA

Anonymous said...

"I remember an example earlier in the season in a MIN-SJS game, where Backstrom was doing the style of goaltending you described. Early on it led to some saves, yes, but as SJ's shooters started to pick on the fact that Backstrom was going down early every play, they were able to pick him apart. His stat-line that day: 18/23 for .783 SV%,with a 4.72 GAA
"

That is exactly my point though. Bacstrom still winded up with some pretty good numbers this year. Why? because his team's defense was good enough to keep the majority of teams from consistently doing what the Sharks did. there are a lot of butterfly goalies like that as we saw with Lundqvist and Price in the postseason. These guys rely on a good team defense to cover up their lack of skill. When these goalies then face teams that are good enough to capitalize on their obvious weakness, its like shooting fish in a barrel. It happens from time to time in the regular season, hut a lot more frequently in the postseason, hence the Sharks being bounced early every year, or Lundqvists career .907 save percentage. But again its easy to see what happens with Peter Budaj because Colorado is not a good team. With Backstrom or Nabakov it is harder to see because the team defense is good, and with Lundqvist it is even harder to see because not only is his team defense good, but his home scorekeepers also do an excellent job or padding his stats.

overpass said...

Anonymous - It sounds like you have a testable hypothesis. If you believe that the performance of butterfly goalies is more sensitive to team quality than the performance of other goalies, why not run the numbers to test it?

Anonymous said...

I'd like to get around to it, however there are 2 big hurdles. The first is how would we define a non butterfly goalie. Obviously there is Brodeur, and Osgood and a few others. But the thing is, every goalie uses the butterfly, its just a matter of how reliant they are on it. One possible way to go would possibly be by identifying which type of pads a goalie wears. If you notice Brodeur's pads, they are traditional goalie pads. When he goes into the butterfly, they stay underneath him, while the pads almost all other goalies use are now the swivel pads that rotate and create a barrier to the sides.

From what I have seen though, is that now even the non purely butterfly goalies like Vokoun use the swivel pads because it is an obvious advantage. Why wouldnt Brodeur or Osgood use those then? I have no idea, maybe for the reason old people do not learn to use computers or cell phones. Probably stubbornness to change and a pride in doing things the way they think things should be done. But who knows?

The second thing is that because of the popularity of the butterfly, there are not many standup or hybrid type goalies in the league anymore. So even if there was a sample size, there chances that they would have played for similar style teams may be an issue, as would be comparing the overall degree of change in the numbers. It would not be a matter of seeing "who's better" but seeing how much variation you would see from. Chris Osgoods numbers with the Red Wings are pretty consistently average at around 907. His period with a bad Islander team also saw his numbers at around .906. And even with the Blues, his numbers were again close to .905. Meanwhile Manny Legace, who pretty much played for the exact same Detroit team, was at around .919. Then goes to St. Louis and his numbers drastically drop to about .904. So two goalies who pretty much played in the same systems on 2 different teams resulted in a rate of change of maybe .002 for Osgood, while the butterfly goalie Legace, saw his numbers change roughly .015.

Anonymous said...

I would not say that Giguere, at least, was a "pure" butterfly. He is a butterfly goaltender, but is also capable of doing a lot of "flopping" like Hasek (just look at his performance in the 2003 playoffs, esp. Game 1 against Minnesota). He's also one of the very most consistent goaltenders in the league--this was his first bad year, really, and he had a lot of personal problems off-ice to deal with that distracted him.