Kukla's Korner

Kukla's Korner Hockey

The Curious Case of Chris Osgood

First, let me say I am really happy that there’s a 3 year waiting period for election to the Hockey Hall of Fame, since that gives everyone a better chance to cool down and reflect on a career, rather than operating on emotion.

That said, there’s still a lot of discussion going on, and some alleged pundits must have had “No Way Osgood Gets Into the Hockey Hall Of Fame” columns stored away on their disks, hands hovering over the “send” command, just waiting….

Me? I’ve been running the numbers, and working through the history on Chris Osgood.

First, let’s see what we’re grading on. Here’s what the Hockey Hall of Fame has to say about candidate attributes:

“Playing ability, sportsmanship, character and their contribution to the team or teams and to the game of hockey in general.”

Looks like it’s all covered: talent, work ethic, avoiding trouble,  and how well do you play with others. Looks suspiciously like a performance review, doesn’t it?

I then looked at the NHL goalies who have been voted into the Hockey Hall of Fame, specifically those who have played with modern equipment. That means at least basket masks, and a the new lighter pads and protected gear. That group includes: Billy Smith, Grant Fuhr, Patrick Roy, and Ed Belfour.

image

Osgood is on the lighter side of games played, but not appreciably so. And he also showed the same pattern of being able to tighten up his goatending during the playoffs. Osgood compares favorably with his closest peer in the group, Belfour.

Smith and Fuhr are definitely from the “don’t worry—we’ll score more” era of the game.

And let’s take a look at two other peers who played around the same time Osgood did, Curtis Joseph and Dominik Hasek:

image

Unless it gets pretty weird, Hasek will be in the Hockey Hall of Fame. Joseph is another name that gets punted around as “should he or shouldn’t he”, with more falling on the side of “close but not HHOF material”.The big hole in Joseph’s resume is a lack of a Stanley Cup, but his stats are just far off from the rest of his peers who have made the cut to show where the line is likely drawn.

Osgood’s numbers are much more in alignment with his peers. And when you look at some of the usual arguments, you find that Roy, and to a much closer extent, Belfour, had similar environments. Belfour left the Blackhawks just as they were sinking, and after a partial season with the Sharks, he arrived at Dallas just in time for a Stanley Cup. Roy arrived in Colorado when what the Avs needed was a goaltender who could lead them to the Cup. Osgood? He was traded to the Islanders and the Blues at a time when both teams were past their heights. Yet he managed to play his game, and come back to the Red Wings after their experiment in Goalie D’Jour.

What’s colored Osgood’s career is that first year playoff loss aganst the Sharks, which if you think about it doesn’t give the Sharks any credit for their hard work, a rookie’s reaction, and his coach pulling the rug out from under him in public. How many players would have come back from that, much less come back to have to prove themselves repeatedly? Prove themselves, and still rack up numbers that show him in the same class with other Hockey Hall of Fame goaltenders?

Interesting questions, but we’ll have to wait a few years to find out. But I get the feeling the vote is going to go in Osgood’s direction.

Filed in: NHL Talk, Laurie Sefton, | KK Hockey | Permalink
  Tags: chris+osgood, goalies, hhof

Comments

Avatar

This is among the most cherry-picked, apples to oranges arguments I’ve ever seen. I should have stopped reading when you tried to say Osgood wore similar equipment to Billy Smith and Grant fuhr, which is stupidly outrageous. Here’s some pictures. Take a good look at the chest protectors, blockers and pads you’re grouping together as ‘modern’.




And if you want to include Belfour and Roy as modern-day comparables because they wore the same sort of pads for part of their careers (the other guys in your first group are out), you can do that, but have to use the numbers from their career overlap. Team factors aside, Roy’s save percentage and GAA in 1986 pads, behind a 1986 system, has nothing at all to do with Osgood’s save percentage or GAA in 2008.

If you adjust for career overlap (which is STILL skewed in Osgood’s favor, because you would now be comparing Osgood’s prime against Belfour and Roy’s respective twilights), these guys dust Osgood’s numbers badly.

And that’s without even looking at the primary knock against Osgood: stats inflated by team factors. In 53% of the seasons in Osgood’s career, the other goalie on the team had higher winning%, better save% and better GAA, indicating that, more often than not, whoever he was playing for would have done better than they actually did with the other guy in the net more often that Osgood. Heck, his save percentage and GAA are below the league mean for 94-2011. Can any of that be said about Belfour or Roy? Absolutely not.

Posted by steviesteve on 08/04/11 at 01:32 AM ET

Avatar

Lidstrom had more to do with Osgood’s win percentage than Osgood did.

Posted by kmad from Vancouver on 08/04/11 at 03:30 AM ET

Nathan's avatar

Geez… I think the truth lies in the middle guys… he’s not as good as the OP makes him sound, and he’s not as bad as the comments make him sound. I wouldn’t vote for him for the HOF, but I also don’t think you can freak out if he were voted in.

Posted by Nathan from the scoresheet! on 08/04/11 at 09:35 AM ET

Avatar

Watch the games.  Nobody should be voted into the Hall of Fame based on reading the stats page. 

And nobody ever watched Chris Osgood play for more than a period and said “Wow, this guy is one of the greatest ever.”  Or even “the greatest right now.”  Which isn’t to say he was a bad goaltender (except sometimes, and when he wasn’t playing for the Wings). 

He had an amazing career.  But there’s no denying he benefited from playing for a great team.  I don’t think there was any point in his career where he was the answer to “If you could pick any goaltender right now for a winner-take-all game, who would you pick?”  I don’t know if there was any point in his career where he was the answer to the same question if you changed it to “any 5 goaltenders right now.” 

He just wasn’t great compared to other contemporary goaltenders.  He held his own a lot of the time, but he didn’t set the bar.

Posted by jonquixote on 08/04/11 at 11:56 AM ET

awould's avatar

It is funny how I never hear about all the great teams Roy played for, as though he was the starting goalie for Columbus his entire career. You can’t lick the ass of Roy’s stats as though they stand on their own and then in the same conversation bash Osgood’s accomplishment’s as a gift from Lidstrom. I will be the first to admit Roy is a better goalie than Osgood, but these arguments I keep hearing against Osgood have no consistency. The criteria used to judge him seems to be whatever is most convenient to stack against him.

Not sure what is cherry picking about a direct comparison of GAA, Save% and playoff stats - last I checked those are the key metrics for gauging a goalie’s success. Comparing Osgood’s stats against Fuhr, Belfour, Hasek, Joseph and Roy is completely reasonable. Whenever a goalie is a legit HHOF candidate - which Osgood most certainly is - the conversation always goes to comparisons with the most recent HHOF/candidate peers. He compares well.

Posted by awould on 08/04/11 at 12:53 PM ET

J.J. from Kansas's avatar

And nobody ever watched Chris Osgood play for more than a period and said “Wow, this guy is one of the greatest ever.”  Or even “the greatest right now.”  Which isn’t to say he was a bad goaltender (except sometimes, and when he wasn’t playing for the Wings).

Be careful when you’re talking about “everybody” and “nobody” because there was a whole stretch of games in the 2008 playoffs where I wouldn’t have traded Osgood for anybody in the league.  I can’t say I speak for “everybody”, but I know I wasn’t alone in feeling that.

Nor was I alone in feeling it again during the 2009 playoffs.

Posted by J.J. from Kansas on 08/04/11 at 12:57 PM ET

Avatar

^ What JJ said. That is all.

Posted by wingsluver4ever from TC on 08/04/11 at 01:37 PM ET

Avatar

Not sure what is cherry picking about a direct comparison of GAA, Save% and playoff stats - last I checked those are the key metrics for gauging a goalie’s success. Comparing Osgood’s stats against Fuhr, Belfour, Hasek, Joseph and Roy is completely reasonable. Whenever a goalie is a legit HHOF candidate - which Osgood most certainly is - the conversation always goes to comparisons with the most recent HHOF/candidate peers. He compares well.

Posted by awould on 08/04/11 at 11:53 AM ET

Conditions change over time. Fuhr and Roy spent a large chunk of their careers behind, for instance, teams that didn’t block shots because doing so would result in a broken leg. Early in Roy’s career (and for all of Fuhr’s) tenders had to choose between making themselves tall to cover the top shelf and making themselves as small as possible to cover 6 holes as an either/or. There were no 6 holes in Chris Osgood’s career because the chest protectors grew.

Roy didn’t suddenly get much faster and far more coordinated at 37 than he was at 27 (when he dragged a crap Mtl team to a Stanley Cup). The league changed in the intermittent years. Look at what a league-leading save percentage was in Roy’s rookie year vs Osgood’s.
http://www.hockey-reference.com/leaders/save_pct_yearly.html

It’s 30 points higher.

It’ll be funny to see the same people singing the opposite song when Datsyuk retires. By your reasoning he should go shine Bernie Nicholls’ shoes.

I don’t think Nicholls is in Datsyuk’s class. I think the league just changed.

Posted by steviesteve on 08/04/11 at 01:54 PM ET

Avatar

Posted by jonquixote on 08/04/11 at 10:56 AM ET

I don’t think the eyeball test is a valid argument when considering candidates for the HHOF. Because it’s clear that a ton of the players who are already in (especially forwards) fail this test. Nobody ever thought Nichols, or Ciccarelli, or Anderson, or Federko, or Gillies, etc. were among the best who ever played the game. Most of those guys weren’t even the best players on their teams, and wouldn’t have been good enough to lead a team to a championship if they were the best player on it.

Yet they’re in, because their consistency and longevity produced career totals that met the Hall’s standard. Osgood, I think, is in the same category.

Posted by Sven22 from Grand Rapids on 08/04/11 at 02:19 PM ET

kcameron's avatar

I think Dino was one of the best to play the game. My mancrush will never die.

Posted by kcameron from Portland OR on 08/04/11 at 02:29 PM ET

Avatar

Sweater retirement is a team honor. It would seem, from what I have read, Red Wing fans are more than happy to have had Osgoode backstop their winning teams and contribute to all their seasons’ successes.

But the HHoF is for the whole NHL and not just the Red Wings fan base. This is where the measure of a player’s career needs to rest in favorable light.

Obviously, the selection committee will make the final decision.

Posted by Shaun from Toronto on 08/04/11 at 03:01 PM ET

awould's avatar

Roy didn’t suddenly get much faster and far more coordinated at 37 than he was at 27 (when he dragged a crap Mtl team to a Stanley Cup).

Let’s not overstate it. MTL wasn’t some 8th seed beating the goliaths. They were 2nd in their division, something like 8th in the league, and had the luxury of watching the league’s heavyweights get beaten by other teams. Two of the teams they beat, including in the Finals, were worse in the standings that year and the other two were only slightly better. They made it past the first round the prior two years and the next six years. And like any Cup winning team, they had their share of future Hall of Famers.

And my point is, I’m sure if it were Osgood’s Cup win that year, those facts would have been pointed out by you. Again, Roy was one of the main reasons they won that year, but he didn’t do it alone. Just like Osgood was one of the main reasons Detroit won their last Cup, and he didn’t do it alone. The difference is, for whatever reason, Osgood is given no credit.

As for different eras and style of play, that line of reasoning is custom made for cherry picking. Basically, you just offered a lengthy opinion and presented it as fact. Yes, the game has changed but exactly how it has and the ramifications are debatable making it a useless exercise where anyone can argue support for whatever they want to believe.

The only true basis for comparison we have are the statistics, and he measures up well. Clearly the hockey experts who vote for the HHOF will debate a lot of these same points, but it seems here on the interwebs that the Ozzie haters do not argue fairly. Osgood accomplishments are suspect because his team was great? Stretching this logic a bit, Peter Forsberg shouldn’t be considered because he had Roy backstopping his team.

Posted by awould on 08/04/11 at 03:06 PM ET

UMFan's avatar

The same arguements about Ozzy were being said of Yzerman in the early to mid 90s. Just because you are not percieved as being in the same class as two generational players in Mario and Wayne doesn’t mean you are not hall of fame material. Ozzy had the unfortunate luck in playing at the same time as three of the greatest goaltenders to ever play the game in Roy, Hask and Brodeur. Was he in their league, no. He was a step down from them, equal in my opinion to Belfour. But because of that, does that mean only Roy, Hasek and Brodeur are the only ones that deserve hall of fame recognition? Ozzy was no imortal. His career though points in the direction of better than a good goalie. Thats why he deserves to get in.

Posted by UMFan from Denver, Colorado on 08/04/11 at 03:33 PM ET

Avatar

And my point is, I’m sure if it were Osgood’s Cup win that year, those facts would have been pointed out by you. Again, Roy was one of the main reasons they won that year, but he didn’t do it alone. Just like Osgood was one of the main reasons Detroit won their last Cup, and he didn’t do it alone. The difference is, for whatever reason, Osgood is given no credit.

It couldn’t have been Osgood’s cup win that year. Osgood never won with an 8th seed or a lineup anywhere near that weak. Look at that 93 lineup for a second. A few good defensive forwards, two good 2-way forwards, a bunch of pylons with hard shots on D and Brian friggin Bellows as the most-deadly scoring threat on the offense. Roy was not a major reason the Canadians won a stanley cup in 93; he was why they won it, hence the Conn Smythe. This finals gets replayed on NHLN from time to time: Roy was under siege and shut the damn door in Gretzky’s face.


As for different eras and style of play, that line of reasoning is custom made for cherry picking. Basically, you just offered a lengthy opinion and presented it as fact. Yes, the game has changed but exactly how it has and the ramifications are debatable making it a useless exercise where anyone can argue support for whatever they want to believe.

This is a pot calling an onion black and you have missed the point entirely. There’s no way to make a meaningful comparison between what Gary Cheevers did to what Tim Thomas did because too much has changed.

There are, however, three ways to eliminate situational bias.

You can take a guy and compare him to another guy when there’s career overlap, for only the duration of that overlap. Doesn’t adjust for age or team, but it does adjust for how they both performed in the same leaguewide conditions, if you’re looking to do a 1-1.

You can compare a guy’s numbers to what his peers do, on aggregate, during the same years he’s in the league. This subjects him to the same shooters, running the same offenses, scoring at the same rate as everyone else, of every age, of every team, at the time, but doesn’t do anything to address team biases.

The last way to eliminate situational bias is to look at what the other guy on the same team with the same job did. This eliminates team factors.

Osgood flunks all three of these tests, getting dusted by Roy and Belfour during career overlap, performing below league mean in the second one and performing below his teammate in the other one.

There’s no opinion here. Data says he flunks.


The only true basis for comparison we have are the statistics, and he measures up well. Clearly the hockey experts who vote for the HHOF will debate a lot of these same points, but it seems here on the interwebs that the Ozzie haters do not argue fairly. Osgood accomplishments are suspect because his team was great? Stretching this logic a bit, Peter Forsberg shouldn’t be considered because he had Roy backstopping his team.

This is totally beside the point, but I’ll address it because this is so stupid. Please explain how Peter Forsberg’s totals were inflated by Roy’s ~20 assists in the ~500 games he played in Colorado.

Posted by steviesteve on 08/04/11 at 03:48 PM ET

Alanah McGinley's avatar

This is a bit random, but a piece popped up today on Light House Hockey about how Osgood’s (and DiPietro’s) cage is made. (Those cages are no longer professionally manufactured.) You can check it out here, if anyone’s interested.

Posted by Alanah McGinley from British Columbia on 08/04/11 at 03:53 PM ET

Avatar

The same arguements about Ozzy were being said of Yzerman in the early to mid 90s.

I would love a link to any article from then that says Yzerman’s production was below the statistical mean for a forward in the late 80s. Or one that says the second-best center had greater ppg from 83-95.

Posted by steviesteve on 08/04/11 at 03:56 PM ET

Avatar

Second-best red-wing center, that is.

Posted by steviesteve on 08/04/11 at 03:58 PM ET

awould's avatar

There are, however, three ways to eliminate situational bias.

There is no way to eliminate situational bias in sports without creating some other bias. That’s why it is a fruitless argument, it is a tool for people who need to reframe an argument to ‘prove’ their point. You jump through hoop after hoop in an effort to disprove numbers that sit right on top. Saying you can only compare a period of a career against one another to compare players is also ridiculous - the HHOF recognizes a career, not some span of time within a career. So career stats are what we got. And what counts more than winning?

But as far as your theories on eliminating this bias goes, nobody is saying Osgood is better than or as good as Roy. But Roy, arguably greatest goalie who ever played, is not the measuring bar for the HHOF.

But for the record, during their overlapping careers (basically 95-03), Roy had a save% of 0.918 compared to Osgood’s 0.908; GAA of 2.24 and 2.33. If you want to ding Osgood for playing on such great teams, you must factor in his two years on NYI while Roy was in Colorado this entire time to make a fair comparison. So let’s get rid of those years. Let’s take this useless exercise a step further to make it more comparable.

I believe that Detroit/Colorado from 95-01 were very similar, excellent teams. The Stanley Cups they won and their playoff success show this. Colorado had 273 wins and 2 cups, Detroit had 284 wins and 2 cups. The rivalry existed for a reason. During these years, here is how they stack up:

Games Played
Osgood: 329
Roy: 352

GAA
Osgood: 2.29
Roy: 2.30

Save%
Osgood: 0.909
Roy: 0.916

Win%
Osgood: 0.56
Roy: 0.55

These stats amount to a difference of 4 goals over the entire duration if you equalize for games. Wow. I bet if Roy played for Detroit during these years, they wouldn’t have lost a single game since they wouldn’t have had to cover for their crap goalie.

When it comes down to it, the only common denominator is they all play in the NHL. During his NHL career, Osgood amassed the 10th most wins ever, and a bunch of other solid career statistics that apparenlty only count if you’re not Chris Osgood.

Posted by awould on 08/04/11 at 06:13 PM ET

Avatar

There is no way to eliminate situational bias in sports without creating some other bias.  So career stats are what we got.

Posted by awould on 08/04/11 at 05:13 PM ET

Curious then, how Datsyuk isn’t about to flunk out of the Hall of Fame. The chance of him matching Brian Bellows’ point totals is nonexistant. If you want to use a guy who was a decent two-way player (Bellows was not), Datsyuk’s chances of matching Damphousse’s totals is also nonexistant. Those guys didn’t make the cut.

Without situational adjustments to account for things like ‘how hard was it to score in those years’, Datsyuk’s totals are pedestrian. Forsberg’s totals are pedestrian (still waiting for an explanation for how Patrick Roy inflated his numbers, by the way).

That’s why you look at what a players’ peers did at the same time. What his teammates did put in the same position (look at all the guys who’ve had career years on Thornton’s wing, been taken out of that role and done jack squat after: Michalek, Cheechoo, Ekman, Murray, etc.). Looking at peers is the only way to avoid these apples to oranges type comparisons that require some warped logic pretending Billy Smith wore JS Giguere’s chest protector. Or that Chris Osgood’s career save percentage tells me anything about Grant Fuhr.

Adjustments for current circumstances are not unique to Osgood, in any way. Ovechkin’s 65 in 2008 is far more impressive than Robitaille’s 63 in ‘93 because of what everyone else in the league was capable or not capable of doing against similar opposition in those particular years. Iggy’s 96 in ‘02 is, likewise, much more significant than Rob Brown’s 115 in ‘89 for the same reason.


But Roy, arguably greatest goalie who ever played, is not the measuring bar for the HHOF.

You are, however, right about this. The measuring bar is not best-ever, it’s best guy not in now.

That guy is not Chris Osgood.

It’s Tom Barrasso

vs. Osgood
-the same number of rings as Osgood as a starter,
-4 more Vezina nominations, (indicating GMs thought he was top 3 in the world at his job)
-one more Vezina win, (indicating GMs thought he was the best in the world at his job)
-two second team all star selections and a first
-a calder,
-retired as the winning-est American in history (10th overall at the time, same as Osgood now)
-currently has the most assists of any goaltender ever, including Brodeur.

Do I think Barrasso should go in? Nope. I didn’t think so when he retired and I don’t think so now. He won a lot, but he was also on some powerhouses. Half the time either the backup or half the guys in the league had better numbers. Sound familiar?

Posted by steviesteve on 08/04/11 at 07:26 PM ET

awould's avatar

Barasso,huh? One more hoop to complicate a simple argument. Those four or five years he had a GAA under 3 were really great. He even topped 0.91 save percentage a couple years. A Vezina is nice, to be sure, but that takes one good year. Barasso’s career stats: 3.24 GAA, 0.892 Save%. That might have something to do with why he’s on the outs. Thanks for making me dig that out, guess since it didn’t fit your argument it wasn’t worth mentioning.

But I do like how you prove your opinion that Osgood isn’t worthy by using your opinion that Barasso is the best guy not in now and then providing further support by stating your opinion that even the vaunted Barasso isn’t worthy, effectively closing the door on any consideration of Osgood. In your opinion anyways. I find that argument lacking for some reason.

The Roy/Osgood numbers for their 95-01 seasons, the ones you ignored, support the idea, based on your own argument of eliminating the bias, that Osgood is worthy. Why don’t you argue against that without contradicting your previous post and theories.

Osgood’s case for the NHL is simple to see – his stats. They stand on their own and cannot be ignored. His detractors all rely on a massive presentation of cherry-picked comparisons and tortured logic to support their belief. Usually the simple argument wins because facts are easy and don’t need caveats and assumptions. Barasso?

As for the Forsberg thing. You missed my point so badly it feels intentional. My point is that by your logic, Osgood’s accomplishments are due to the great team he played on. So any player on a great team, surrounded by other great players, should also be held to that standard. I don’t agree with your logic as it fails to account for a lot of things, starting with the fact that as a member of those great teams, he contributed to their success. You’ll turn around and say that the Forsberg analogy, which I agree is stupid, doesn’t work because a goalie doesn’t influence a player’s success like a player can influence a goalie’s. That is true. A goalie’s success is greatly influenced by the team in front of him. And by the coaches, the style of offense/defense they play and a lot of other things that I’m sure his detractors will latch on to in due time if it supports their argument. (Bowman coached Osgood, for crying out loud, so how could he NOT succeed!) But why is Osgood the only goalie who is derided for winning on a good team? Every HHOF goalie won on good teams. This argument is weak and should have little weight in the discussion.

Posted by awould on 08/04/11 at 08:35 PM ET

Avatar

Barasso,huh? One more hoop to complicate a simple argument. Those four or five years he had a GAA under 3 were really great. He even topped 0.91 save percentage a couple years. A Vezina is nice, to be sure, but that takes one good year. Barasso’s career stats: 3.24 GAA, 0.892 Save%. That might have something to do with why he’s on the outs. Thanks for making me dig that out, guess since it didn’t fit your argument it wasn’t worth mentioning.

Era adjustments, team adjustments, etc. Doesn’t matter, though. I’m not arguing for Barrasso. He doesn’t belong. For what it’s worth, though, Bowman thought more of Barrasso, considering he never exposed him to waivers when he was DPP in Pitt, something that cannot be said of his treatment toward Osgood.


The Roy/Osgood numbers for their 95-01 seasons, the ones you ignored, support the idea, based on your own argument of eliminating the bias, that Osgood is worthy. Why don’t you argue against that without contradicting your previous post and theories.

I ignored them because your criteria make no sense. You’re trying to force an adjustment that can’t be made.

By tossing out 3 years a piece for the reasons of ‘I think it’s about the same to play for team A and B’ you’re making an mathematically unsupportable adjustment (not at all suspicious in that the years you’re throwing out favor Roy, greatly) to a valid (but incomplete, since team factors aren’t accounted for by looking at career overlap) one.

What I do know for certain
-Are both these teams good? Yep.
-How did Patrick Roy do against 98’s shooters behind the Avalanche D? Better than average GAA and svp.
-How did Billington do against 98’s shooters behind the same Avalanche D?  Slightly better than Roy, actually.
-How did Patrick Roy do against 99’s shooters behind the Avalanche D? Way better than average
-How did Billington do against 99’s shooters behind the same Avalanche D? Not nearly as well as Roy.
-How did Chris Osgood do against ‘98’s shooters behind the Red Wing’s D? Better than average GAA and svp.
-How did Hodgson do against ‘98’s shooters behind the same Red Wing’s D? Not well. Maracle did better but the sample size is too small
-How did Chris Osgood do against ‘99’s shooters behind the Red Wing’s D? Below league average GAA and svp.
-How did Maracle do against ‘99’s shooters behind the same Red Wing’s D? Way better than Osgood (as did Ranford, in limited appearances).
etc.

What you’re telling me I know but don’t
-Is the task of goaltending behind these two teams, in either of these seasons, equivalent?
or
-If I swap these guys will their numbers change?

I have no idea based on this unknowable variable you introduced for reasons of ‘if I truncate a few seasons of overlap Osgood will look better’. Neither do you.

Era factors can be adjusted for if there is overlap (age and team factors can’t be, though). Team factors can be adjusted for -within the team- (though not while considering era). It would be nice if these could be combined, but, as far as I know, they can’t, at least, not in the way you’re trying to.

Here’s the actual numbers for career overlap

Roy
.916 SA%
2.36 GAA
16974 ShA
36010 min

Osgood
.906 SA%
2.47 GAA
12658 ShA
28743 min

This isn’t particularly close and that’s ignoring that Roy’s career was winding down and Osgood was at peak health during this period (I ignore this because I can’t know what 24 year old Roy would be able to do in circumstances he was never in).


Osgood’s case for the NHL is simple to see – his stats. They stand on their own and cannot be ignored. His detractors all rely on a massive presentation of cherry-picked comparisons and tortured logic to support their belief. Usually the simple argument wins because facts are easy and don’t need caveats and assumptions. Barasso?

There’s no cherry-picking. Who the GMs thought was one of the best 3 goalies auto-compensates for relative performance across eras. Save percentage doesn’t. Raw point totals don’t.

I’m not the one pretending Warren Young was deadlier than Marcus Naslund because he scored more goals than Naslund ever did in one season when goals were a lot easier to come by. Or failing to grasp that a GAA in the 3s was good in 1989, when there were 4 150 point players and no shotblockers. Or that a 100 point season in those circumstances meant less.

Three times now you’ve failed to account for the fact that your own reasoning: that raw stats mean the same things across eras (whether they’re from 1988 or 1898) would preclude Datsyuk from consideration for the Hall (based on the much more-impressively numbered players who are not there), let alone induction. It would be a travesty if that guy got sacrificed on the altar of Osgood, who couldn’t even do better than his backups more than half the time and couldn’t even hit the statistical means across his career.

While we’re at it, may as well tell Bure to go to Hell. Wouldn’t want him blocking Brian Propp’s induction.

Posted by steviesteve on 08/05/11 at 12:09 AM ET

awould's avatar

There’s no cherry-picking.
Your entire theory is cherry-picked. That is actually my overall point. It is so convoluted as to be pointless since anyone with a hockey database and a working knowledge of excel could turn it around.

How did Patrick Roy do against 98’s shooters behind the Avalanche D?
He wasn’t just behind the Avalanche D. He was also behind the Avalanche O. Just like Osgood in Detroit. So now please factor in the goals scored and place every situational statistic into a database and get a stats wizard to program a kick ass algorithm to model it all and tell us who is HHOF worthy. Because that’s where your argument must go unless it is arbitrary.

raw stats mean the same things across eras
My only point here is that if a player reaches Top 10 in Wins All Time, I don’t care what era they played in. The NHL has been going on a very long time, top 10 in any stat is impressive but top 10 wins for a goalie is more so. Do I think Datsyuk won’t get in because some other guy scored more goals than him 5 or 20 years ago? Talk about cherry picking. The criteria is vastly different for a skater than a goalie anyways.

Also, it is a misleading exercise to argue about what player should or shouldn’t get in based on who isn’t in. Barasso, by many accounts, isn’t in the HHOF because he’s apparently a total jerk who nobody likes. Each player is different and each case to be made for the HHOF is different. To argue no Barasso=no Osgood assumes the criteria never changes and is applied evenly.

Era factors can be adjusted for if there is overlap (age and team factors can’t be, though). Team factors can be adjusted for -within the team- (though not while considering era). It would be nice if these could be combined, but, as far as I know, they can’t, at least, not in the way you’re trying to.

You’re right, they can’t. If they could, the basis for your argument might be useful. Also, I’m not trying to combine them – at least not to prove anything other than it is a dumb idea. I think the entire premise is pointless. Any 8th grade science teacher, or student for that matter, would tell you that you can’t draw conclusions from an experiment when you have more than one variable out of your control. Your exercise attempts to set up 3 scenarios where you actually control only one variable, instead of the other way around. Any conclusion is basically worthless and it is very easy to craft an argument that proves the other thing. Such as comparing Roy/Osgood with career overlap on quality teams – that controls for career overlap and ‘within a team’, at least it comes close. It doesn’t control for age. But it hits two of your three things and it makes numbers, so, hey, I proved something. You disagree because you think that the criteria you defined is the only right one to consider, which is just part of the cherry-picking.

You don’t feel Osgood belongs in the HHOF and you have an argument that you believe supports that conclusion. I believe your argument is inherently flawed. Primarily, I think any argument that delves into slicing and dicing numbers, making assumptions on top of assumptions and drawing conclusions from such a mess is a bad idea. At some point, it really just isn’t that complicated. Certain guys you look at their stats and it is obvious. Certain guys have good but not great stats but always excelled in the playoffs, scored crucial goals or made clutch saves and their peers all support their inclusion. I think Osgood falls a little into both categories, which is why it is controversial.

Anyways, we’re both just here making an argument that is going to impact nothing in the world. Clearly we both like to argue. But I bet in the end, Osgood gets in. Probably not the first time around, but within a reasonable period where he can still walk up and deliver an acceptance speech without needing an oxygen tank.

Posted by awould on 08/05/11 at 02:16 PM ET

SnLO's avatar

I’m a little late to this discussion, but, just for arguments sake, to add a little more perspective that was not explored in the original post:

 


              GP       W       W%
Joseph   943   403   42.7%
Smith     680   305   44.9%
Fuhr       868   403   46.4%
Belfour   963   484   50.3%
Hasek     735   389   52.9%
Roy       1029   551   53.5%
Osgood   744   401   53.9%

This is a rather simple analysis of regular season games played and wins to arrive at a career win percentage. I arranged it in ascending order. Only four of the seven goalies discussed won more than half the games they played.

Posted by SnLO from beyond the M-1 on 08/05/11 at 03:38 PM ET

Avatar

You disagree because you think that the criteria you defined is the only right one to consider, which is just part of the cherry-picking.

Career overlap between players A and B is only subject to selection bias if I pick the players, which I didn’t. I can’t jury-rig that to produce a result I want, unless I re-select these players, which I also didn’t.

Vs teammate is not subject to selection bias. I can’t jury-rig that to produce a result I want.

Vs. league mean across career is not subject to selection bias. I can’t jury-rig that to produce a result I want.

Overlap on teams you personally happen to think are both equal for a goaltender to play behind is utterly arbitrary.

You’re right that it would be preferable to have a theorem that accounted for all modifiers, because yes, I’m only looking at one thing at a time, but nothing like that exists. So I have to work with what I do know. The questions I do know the answer to are how did guy x do in the conditions he was in vs the other guys who did his job in the league at the time in the conditions they were in? How did guy x do in comparison to teammates in identical conditions during his career?

If you don’t like imperfect (but impartial) attempts to normalize for league conditions across two careers that half overlap, take it up with lsefton, not me. The premise that Roy (and, uh, Billy Smith) played under the same league conditions and, somewhat ludicrously, wore the same size pads as Chris Osgood the whole time is the poster’s supposition, not mine.

If it was my choice, I would never put forth a numerical comparison between Patrick Roy and Chris Osgood (I merely responded to one) or, say, Sidney Crosby to Peter Forsberg (statistically) because you’re right—there are too many variables I can’t combine, overlap or not. But it’s not my premise.

I’d wait until the day was over and see who won more richards, art rosses and harts. Championships. How much (or if) they overperformed vs the other guys THEY played against. How long they were an impact player for? Basically, things that aren’t going to change whether I’m looking at Teemu Selanne or Guy Lafleur.

This isn’t to say I wouldn’t look at Crosby vs. Ovechkin by stats, though. They’re about as close comparables in age and league-conditions (probably team strength as well, but that’s a total guess) as two hall-track forwards on different teams are ever going to get right now.

In Osgood’s case, the Hall-bound guy whose career mirrors his own conditions (not results) the most (age, starting point) would be Brodeur, which is probably why four guys with numbers ballooned by playing in the 80s were selected for comparison instead, as if that meant anything.

But if you now want to throw out stats completely because they aren’t comparable and can’t be normalized, which I’m fine with, Osgood’s not a Hall of Fame goaltender simply because there is no NHL goaltender since the Vezina trophy’s existence to be inducted without winning at least one. His induction would be unprecedented. Nothing he’s done in his career is unprecedented (10th in wins means 9 other guys won more), let alone unprecedented enough to waive the apparent necessary condition of ‘perceived best guy in the world at his job at one point in time”.

Posted by steviesteve on 08/05/11 at 04:52 PM ET

awould's avatar

We will go in circles here. I’m not saying you’re jury-rigging the stats you select to produce results you want, I am saying the choice you make in the stats you deem important or relevant aren’t useful in this argument due to the massive amount of other variables. I’m also saying that you are wrong to assume that the choices you make, the stats/comparisons that you select as important, are unquestionably the best basis for comparison. I think a lot of people would agree that comparing Osgood with Roy while both were playing during that rivalry is a fair comparison. You call it cherry picking but it is no more cherry picking than just choosing career overlap.

As for the different eras, you keep tossing out this notion that nobody blocked shots back in the day and their equipment was tiny, as though that must have made life so hard on the goalies of yesteryear that they would surely excel in today’s NHL. But each goalie can easily be compared to the league during their career. Here’s the overall NHL avg. shot per game, total goals per game, overall save% and each player’s save% during each player’s career and the GAA. These stats come mostly from http://www.quanthockey.com; data only goes to 2010.

Player   Period   League Avg Shots Per Game   League Avg Goals Per Game
Smith   1971-1989   30.39   7.18
Fuhr   1981-2000   29.95   6.92
Barasso   1983-2002   30.16   6.62
Roy   1985-2003   29.28   6.41
Belfour   1988-2007   29.23   6.09
Osgood   1993-2010   28.86   5.64
Brodeur   1993-2010   28.90   6.09

Player   League Avg Save %  Player Save%  GAA   Difference
Smith   88.19%  88.20%  3.17   0.42
Fuhr   88.45%  88.70%  3.38   0.08
Barasso   88.19%  89.20%  3.24   0.07
Roy   89.05%  91.00%  2.54   0.67
Belfour   89.58%  90.60%  2.50   0.55
Osgood   90.23%  90.50%  2.49   0.33
Brodeur   89.58%  91.30%  2.22   0.83

The “Difference” shown is the players GAA compared to the league-wide GAA during that player’s career. Each of these great goalies had better than average GAA during their careers. As these numbers stack up, Fuhr and Barasso compare least favorably in every metric. They each only slightly out-performed the league in GAA and Save%. One is in the Hall and the other isn’t. The cut off appears to be somewhere in there, based on modern inductees and their stats. It is no surprise that Roy and Brodeur, who battle for title of greatest ever, have the best numbers. Belfour is not in the HHOF (yet) and Brodeur is a lock. Osgood is firmly in the middle of this pack. This is a comparison of each player against the league averages during the times they played. There are 30 teams in the league now and some of those teams are truly terrible while others are excellent. Given that each of these players has at least one Stanley Cup ring, it is safe to say they all benefitted from playing on good teams, just as those teams benefitted from having great goaltending.

As for the Vezina and the importance you place on it…. you are wrong that no goalie has gotten into the Hall without a Vezina. Cheevers never won a Vezina. So, in that sense, it would not be unprecedented. Which doesn’t matter anyways. From 1946 to 1981 that award was given to the goalie who started for the team that gave up the least goals that season – it was not voted on. Many years two goalies won it. As it is, there are 21 goalies in the HHOF that have a Vezina (not including Vezina himself) and ten of them won it because their team let in the fewest goals that season. In 2007-2008, that would have been Osgood. So if that award is so important, the criteria that gave it to those ten goalies puts Osgood in the mix. Again, I’m not saying it is as important as you claim, but since you claim it then its value in getting Plante, Sawchuck, Dryden, Esposito and others into the HHOF should be applied equally. I’m sure some of those ten would’ve won it even if it were voted on, but quite possibly not all of them. Also, a Vezina win in the modern voting system only awards a single great season. More than a handful have won it that have no chance of making the HHOF. The HHOF recognizes a great career. Sustained success shouldn’t be dismissed just because Ozzie doesn’t share a trophy with Olie the Goalie and Ron Hextall. Osgood is overshadowed by Roy and Brodeur, as is nearly every goalie who ever played. That shouldn’t diminish his place in history, just like Yzerman isn’t called a second tier player just because he shared the ice with Lemieux and Gretzky through most of his career.

10th in wins means 9 other guys won more
This is a dumb statement by any measure and highlights how entrenched you are against Osgood no matter the argument. You’re saying that he’s basically the 10th place loser behind 8 other losers. Go to the NHL site and where it lists career stats for NHL goalies, there are 680 of them. So Osgood won fewer games than 9 other guys but more than 670 other guys who played in the NHL. What a chump! Way to dismiss a great accomplishment out of hand.

So basically, your statistical argument is full of assumptions (holes), your Vezina argument is weakened greatly, and you’re shown to just be biased against anything pro-Osgood. I’ve never made the claim that Osgood is the best ever, or even as good as many HHOF goalies. Just that his accomplishments merit induction. He has better career stats compared to many of these goalies and has performed better against the league average than some as well.

Posted by awould on 08/05/11 at 07:34 PM ET

Add a Comment

Please limit embedded image or media size to 575 pixels wide.

Add your own avatar by joining Kukla's Korner, or logging in and uploading one in your member control panel.

Captchas bug you? Join KK or log in and you won't have to bother.

Smileys

Notify me of follow-up comments?

Feed

Most Recent Blog Posts

About Kukla's Korner Hockey

Paul Kukla founded Kukla’s Korner in 2005 and the site has since become the must-read site on the ‘net for all the latest happenings around the NHL.

From breaking news to in-depth stories around the league, KK Hockey is updated with fresh stories all day long and will bring you the latest news as quickly as possible.

Email Paul anytime at pk@kuklaskorner.com