Goalie effectiveness formula?

[Jason MacIsaac]

I'm not sure that Hurme will ever play again to prove you right or wrong (severe back injury).
But the SV%'s of those 3 goalies didn't change when they moved from/to better systems to Florida. More shots didn't help them look better at all.

Backups are a little harder to gauge then starters. Backups let in goals that they shouldn't. Lets see if it is the case for Lalime in the future. Untill then I am going to sleep.

 

[habs9]

An excellent magazine for goalie rating and analysis

http://www.goaliesworld.com/

 

[trentmccleary]

I read both methods and they depend mostly on SV% which is a poor stat. The two worst seasons ever where by an Ottawa early 90's team and a washington team early 70's. That itself shows a flaw.

I don't care what anyone says....how good a player truely was is shown by his championships. Brady is a winner, Jordan was a winner, Roy was a winner and so is Brodeur. He won 3 stanley cups, 1 world cup (something Roy couldn't do) and an Olympic gold medal. It doesn't have to be fancy, as long as he gets the job done.

I find it ironic that you call SV% a flawed stat, even though (good shot/bad shot) the goalie is involved in the play.
But then you cling to wins, which are also very dependent on offense... which the goalie has nothing to do with.
If anybody should be getting a lecture on flawed statistical reasoning, it's you.

 

[Jason MacIsaac]

I find it ironic that you call SV% a flawed stat, even though (good shot/bad shot) the goalie is involved in the play.
But then you cling to wins, which are also very dependent on offense... which the goalie has nothing to do with.
If anybody should be getting a lecture on flawed statistical reasoning, it's you.

Because I value winning over sv%. Let the Marc Denis's, Jose Theodore's, Roverto Luongo's of the world lead in that stat. I will take Brodeur and his Cups any day of the week.

 

[BuppY]

Brodeur the best goalie...

 

[littleHossa]

(Stanley Cups / 6 * 5 + Vezina*Harts)/ 2 - 6

 

[mymkovski]

Because I value winning over sv%. Let the Marc Denis's, Jose Theodore's, Roverto Luongo's of the world lead in that stat. I will take Brodeur and his Cups any day of the week.

As was previously mentioned by other posters, winning has a lot more to do with the team than the save %. Save % is primarily goaltender controlled. Roberto Luongo on the Devils over the same amount of time Broduer had with similarly built teams would amount to similar results for Roberto Luongo. And I was bet it would be fairly close with Theodore and Denis.

Yes, winning is a very important statistic when it comes to judging an individual's career. However, winning is a TEAM sport and a TEAM stat whereas Save % is more GOALIE orientated (it does have a team aspect to it as well).

 

[Kickabrat]

Because I value winning over sv%. Let the Marc Denis's, Jose Theodore's, Roverto Luongo's of the world lead in that stat. I will take Brodeur and his Cups any day of the week.

I don't get what you are arguing about? Outsider's list includes some of the winingest goalies in history. Roy, Plante, Dryden, Hall, Smith...all goalies that would be described as "money" goalies. Stanley cups, Vezina's, etc. So how can you dis his methodology if the majority of the goalies on the list fit your criteria: winning games, regardless of their SV%.

 

[mymkovski]

I don't get what you are arguing about? Outsider's list includes some of the winingest goalies in history. Roy, Plante, Dryden, Hall, Smith...all goalies that would be described as "money" goalies. Stanley cups, Vezina's, etc. So how can you dis his methodology if the majority of the goalies on the list fit your criteria: winning games, regardless of their SV%.

 

[meehan]

As was previously mentioned by other posters, winning has a lot more to do with the team than the save %. Save % is primarily goaltender controlled. Roberto Luongo on the Devils over the same amount of time Broduer had with similarly built teams would amount to similar results for Roberto Luongo. And I was bet it would be fairly close with Theodore and Denis.

All speculation (which seems to be predominant when talking about losing goalies). There is no way you could know how a player would play in another environment unless he actually plays in a similar environment. Yes, it's possible Luongo could do what Brodeur did if the situation was reversed, but until he actually does do that, there can be no comparison between the two as far as actual NHL accomplishments are concerned. For now, the only way to judge the two against each other is on the subjective level.

Yes, winning is a very important statistic when it comes to judging an individual's career. However, winning is a TEAM sport and a TEAM stat whereas Save % is more GOALIE orientated (it does have a team aspect to it as well).

I think earlier on in the tread someone made the remark that just because a company is successful doesn't mean an individual employee is a great worker. Though that is often true, is it rare to find a highly successful company that suffers from incompetance on the executive level. The nhl goaltender is like the CEO of the company. Winning and losing are more dependent his actions then on the actions of any other individual player. As such, when a team is consistently a championship calibur team, the goalie gets the bulk of the credit (save a few cases like Fuhr in Edmonton), just like how when a company is determined to be highly successful, the CEO ends up on the cover of forbes. For all this, in the end, a goalies ability to generate wins for his team, means alot more then his ability to stop alot of shots in loses. A tough save in the closing minutes of a tightly played conference finals game is of much more value then an easy save on some dump in, even though statistically, they have the same value. Most people think a goalies job is merely to make saves; a goalies real job is to give his team the best chance of winning the game. That job encompasses more then making saves. A goalies ability to effectively play the puck thereby reducing opposition forechecking pressure is an immensely important aspect of the position, even though it can't be accounted with statistics. In fact, it probably hurts a goalies stats, as the best puckhandling goalies don't get many easy dump in saves. Also, a goalies attitude and demeaner play a huge role in team success. How much did it help the Detroit players in 2002 knowing they had Hasek back there and not Osgood? Or how much did it effect Philly all those years they had Cechmanek? That's another very important aspect of goaltending that can never be relayed with statistics. As such, the best way to know how good a goaltender is, is to watch how he plays in the biggest games. And not just watching the saves, but watching all aspects of his game.

 

[Malefic74]

Statistics are helpful and all but there is NO formula that will ever accurately measure how good any goalie is. Period. Any formula that goes through a list of all time goalies and puts John Vanbiesbrouck in the top 10 is fatally flawed IMO. In spite of all numbers and varying validity of certain statistics it only comes down to one question:

Given the choice of any goalie, who would YOU want in net?

That's it. Because virtually every stat for goaltenders is affected by game situations having nothing to do with the goalie. And yes that includes save percentage. 5 flip shots from centre ice have the same statitical weight as 5 breakaways. Not the best example of a clean stat is it?

 

[trentmccleary]

Sorry that Method sucks.....if a goaltender sees alot of shots and plays alot of games he will be seen as the best goaltender that year which is not the case. Winning games, holding leads, clutch saves, goals allowed are not included in that that formula.

I'm also sorry but you math dorks need to get a life and watch a hockey game. The higher the shots the better the sv% usually is.

I have to say, Jason's method is starting to grow on me. I think I'll try that out for awhile to see if I like it.

I'll practice with last season:
All of these goalies, were better than Martin Brodeur last season:
Khabibulin, Kiprusoff, Esche, Nabakov, Belfour, Aebischer, Theodore, Joseph, Lalime, Raycroft, Hedberg, Dipietro, Osgood, Cloutier and Auld.

There is no shame shame in being 16th Marty!

 

[Amen evil king]

I have to say, Jason's method is starting to grow on me. I think I'll try that out for awhile to see if I like it.

I'll practice with last season:
All of these goalies, were better than Martin Brodeur last season:
Khabibulin, Kiprusoff, Esche, Nabakov, Belfour, Aebischer, Theodore, Joseph, Lalime, Raycroft, Hedberg, Dipietro, Osgood, Cloutier and Auld.

There is no shame shame in being 16th Marty!

Just out of interest, how did you come to that conclusion? (ie. how did you use the stats?)

 

[trentmccleary]

Just out of interest, how did you come to that conclusion? (ie. how did you use the stats?)

Clutch saves allowed these vastly superior goalies to win more playoff games by preventing goals and holding leads. In the case of a tie, goalies whose clutch saves prevented more goals than Martin Brodeur were considered better.

 

[Hockey Outsider]

Any formula that goes through a list of all time goalies and puts John Vanbiesbrouck in the top 10 is fatally flawed IMO.

My goalie rating system has Vanbiesbrouck 9th. Daryl Shilling has him tied for 3rd with Jacques Plante (http://members.shaw.ca/hbtn/player_study/goaltender_rating.htm). Klein & Reif rate him 11th (Hockey Compendium pg 146). They're all based on different stats and different premises. This leads me to believe that perhaps Vanbiesbrouck has been underrated throughout his career. He's played very well over a very long period of time.

Most of the objections here seem to take the form: "statistics can't tell you everything, so we shouldn't bother using them at all". I agree that stats can't tell you everything, and that other factors, like intangibles do need to be taken into account--I've never said they don't. However, stats are the sole objective record of performance, and they do count for something. Ignoring them entirely in favor of subjective opinions doesn't prove anything and lacks reliability and validity. I guess my point is you need a combination of the two; focusing only on one or the other is shortsighted.

 

[kenabnrmal]

This thread has pretty much boiled down to everyone vs. Mac, and Jason's done a good job of ensuring that it be that way. Jason, with every post you've made in the thread, you've made it more and more clear that the only real objection you have to the "theory" is that it doesn't place Brodeur in the top 10. Goalie stats are a very difficult think to qualify, and of course they should be taken with a grain of salt. Outsider has done a good job of considering that, but understanding that the stats do have their, albeit limited, place and came up with a creative and interesting theory. A hell of a lot better than yours, which seems to indicate that anyone who is on a team that just so happens to win a championship is clearly the best. Winning is massive in team sports, but its the TEAM'S job to win. Everyone does their own job, and the team has a better chance of winning. Hell, your own team the Devils perfected this. Everyone doing their job to perfection stifles the other team. Luongo's a lesser goalie because he's a more recent draft choice and thus is on a losing team? Come on now. I don't agree often with you, but I know you're smarter than that.

Outsider's theory is like all comprehensive statistical analyses...needs to be taken with a grain of salt, but is very interesting...and he might be on to something.

 

[Jason MacIsaac]

This thread has pretty much boiled down to everyone vs. Mac, and Jason's done a good job of ensuring that it be that way. Jason, with every post you've made in the thread, you've made it more and more clear that the only real objection you have to the "theory" is that it doesn't place Brodeur in the top 10. Goalie stats are a very difficult think to qualify, and of course they should be taken with a grain of salt. Outsider has done a good job of considering that, but understanding that the stats do have their, albeit limited, place and came up with a creative and interesting theory. A hell of a lot better than yours, which seems to indicate that anyone who is on a team that just so happens to win a championship is clearly the best. Winning is massive in team sports, but its the TEAM'S job to win. Everyone does their own job, and the team has a better chance of winning. Hell, your own team the Devils perfected this. Everyone doing their job to perfection stifles the other team. Luongo's a lesser goalie because he's a more recent draft choice and thus is on a losing team? Come on now. I don't agree often with you, but I know you're smarter than that.

Outsider's theory is like all comprehensive statistical analyses...needs to be taken with a grain of salt, but is very interesting...and he might be on to something.

It turned out that way but honestly ask yourself this question....do you think there were around 30 to 40 (whatever number it is) goaltenders better then Brodeur last year in the regular season.

 

[kenabnrmal]

It turned out that way but honestly ask yourself this question....do you think there were around 30 to 40 (whatever number it is) goaltenders better then Brodeur last year in the regular season.

No, definitely not, which is why it needs to be taken with a grain of salt. I think intangibles play a BIG part in Brodeur's career, as well as the team in front of him, AND his abilities. I think the intangibles play more a part in his career than any other goalie I can think of, and as such, his stats are going to be more unexpected. I think he's better than what the stats indicate...however, I think he's an exception. I think stats in general provide a decent overview, and I think outsider's done a pretty intriguing job.

 

[Hockey Outsider]

It turned out that way but honestly ask yourself this question....do you think there were around 30 to 40 (whatever number it is) goaltenders better then Brodeur last year in the regular season.

Statistically or not, I don't think Brodeur had that great a year last year. But please keep in mind: it's just one year. Overall Brodeur does extremely well according to my formula and is easily on pace for a Hall of Fame career. He's already in the top 20 in Career Value and top 20 in Peak Value. Since Brodeur's entered the league he's 4th overall (behind Hasek, Roy and the underrated Burke) and has many seasons among the top 10, top 5 and top 3.

Also, I would be very interested to consider how Brodeur's puckhandling affects the Devil's ability to score goals and/or give up fewer shots. I've toyed around with a few ideas but I haven't been able to come up with anything. Any ideas?

And thanks for the support, kenabnrmal.

 

[acr*]

A couple questions-

1.Can I be one of the first to jump on the Hockey Sabermetrics Bandwagon?

2.Is there any way to sort out easy perimiter shots from breakaways in this stat?

3.Does this mean that when I've argued with people about Moog being better than Fuhr, I wasn't talking out of my ass?

 

[Hockey Outsider]

1.Can I be one of the first to jump on the Hockey Sabermetrics Bandwagon?

There's a small but growing number of hockey analysts. If you're at all interested in the topic, you should check out:
-- The Hockey Analysis Group (http://sports.groups.yahoo.com/group/HAG_list/)
-- Puckerings (http://www.puckerings.com/)
-- Hockey Analytics (http://members.rogers.com/hockeyanalytics/Research.htm)
-- The Hockey Project (http://members.shaw.ca/hbtn/)
-- The Hockey Outsider (my humble site) (http://www.geocities.com/thehockeyoutsider/)
-- "The Hockey Compendium" by Jeff Z Klein and Karl Eric Reif (available for under $10 in every bookstore I've gone to--can't recommend it enough)

2.Is there any way to sort out easy perimiter shots from breakaways in this stat?

Depends on what year you're asking about. For 2003 and 2004 I've included shot quality into my project. The work was done by Alan Ryder (Hockey Analytics); he was generous enough to give me that data and let me use it. So, yes, those years take into account shots from the blueline, breakaways, wrist-shots vs slap shots, etc.

I believe (not totally sure) that there's enough data to take shot quality into account for most of the 90s-2002. Sadly the NHL hasn't released it, so obviously I nor anyone else has been able to use it.

3.Does this mean that when I've argued with people about Moog being better than Fuhr, I wasn't talking out of my ass?

Ha. Good question. Well, Fuhr was a good goalie during the dynasty years in Edmonton. He had one amazing season (1982) and in other years he was well above average. However, he had some really awful years later in his career. From '95 onward he was just brutal. So overall, Fuhr was a very good goalie in his peak, but he hung around way too long and this really hurt his career stats.

During their Edmonton years, Moog was about as good as Fuhr. The difference between Moog and Fuhr is that Moog was able to play at a decent level past his prime, whereas Fuhr was just awful.

 

[acr*]

Wow, this is interesting stuff. I've been reading through this big long pdf file about player contribution from the hockey analytics site, and now I believe that "sabermetrics" in hockey is a lot more possible than I would've thought before.

This thing is long too, I've been reading through it for a half hour, I'm on Step 7, and according the the scoll bar on the side of the screen, I'm just halfway through it! It shows that these guys have put a lot of thought into this idea.
I'll look for that book you recommended and I'll join the following of using better stats in hockey

 

[Malefic74]

Most of the objections here seem to take the form: "statistics can't tell you everything, so we shouldn't bother using them at all". I agree that stats can't tell you everything, and that other factors, like intangibles do need to be taken into account--I've never said they don't. However, stats are the sole objective record of performance, and they do count for something. Ignoring them entirely in favor of subjective opinions doesn't prove anything and lacks reliability and validity. I guess my point is you need a combination of the two; focusing only on one or the other is shortsighted.

I agree. There are however a LOT of people on these boards who live and die by statistics alone. My point was more directed at those unfortunate souls.

 

[BM67]

One big problem with most of these formulas is they reward goalies for stopping more shots. Facing fewer shots also brings with it it's own difficulties. Would you rather face a shot after making a flurry of saves a minute ago, or after not seeing a shot for 3-6 minutes?

That's a minor thing next to this "shot quality" thing. He says himself the data is wonky, then when you add in all the factors that aren't taken into account, it's value is very much in doubt. PP shots are of a "higher" quality, but what about shots just after a penalty expires? What kind of shot is it when a shot bounces off two legs and goes to a wide open player beside the net on the other side from the goalie? Is a player more likely to score if the goalie hasn't made a save in several minutes?

Also why if you're going shot by shot, don't you break it down goalie by goalie rather than just lumping them by team?

 

[trentmccleary]

One big problem with most of these formulas is they reward goalies for stopping more shots. Facing fewer shots also brings with it it's own difficulties. Would you rather face a shot after making a flurry of saves a minute ago, or after not seeing a shot for 3-6 minutes?

That's a minor thing next to this "shot quality" thing. He says himself the data is wonky, then when you add in all the factors that aren't taken into account, it's value is very much in doubt. PP shots are of a "higher" quality, but what about shots just after a penalty expires? What kind of shot is it when a shot bounces off two legs and goes to a wide open player beside the net on the other side from the goalie? Is a player more likely to score if the goalie hasn't made a save in several minutes?

Also why if you're going shot by shot, don't you break it down goalie by goalie rather than just lumping them by team?

But facing more shots tires a goalie out. Who could stop a shot better? A fresh goalie or a tired one?

Shot quality would be a real problem in a smaller sample ... but should be much less of a difference (between goalies) as the sample of games increases.