Martin Brodeur never asked to be one of the divisive players in the “stats vs. observation” argument, but we ended up here. Before I even knew what a “blog” was, Alan Ryder wrote that “Brodeur was basically the same goaltender in 2004. He still plays behind the league’s
best defense and he still doesn’t deserve the Vezina.”

Ryder’s research is based around a ‘shot quality’ premise that a lot of modern hockey statisticians are skeptical about. Most shot quality studies involve shot locations provided by NHL.com, but two identical shots from the same location will be rung up officially differently in New Jersey than, say, at Madison Square Garden. With that in mind, however, we can all kind of agree that Brodeur’s success has largely to do with playing behind one of the best defence’s in the NHL for a long time: they prevented shots, it wasn’t just keeping the quality of shots low.

Put it this way: between the two lockouts in 1994-95 and 2004-05, 27 NHL goaltenders started at least 300 games. Brodeur was 1st on the list in wins, shutouts and goals against average, by considerable margins. His save percentage ranked 4th, however, behind Dominik Hasek, Patrick Roy, and Jose Theodore. [Hockey Reference]

My belief is generally that a goalie controls his save percentage more than anything else. The team in front of the goaltender controls allowed shots and scored goals. The rest is up to the goalkeeper. If there is a team that was good at doing both, such as this season’s Pittsburgh Penguins or the New Jersey Devils circa 1994-2004, the goalie has a high chance of racking up a low goals against average and a gaudy wins total.

Brodeur’s name has been incorrectly linked to some of the game’s greats, such as Terry Sawchuk. Many of Brodeur’s records: wins, Stanley Cups, shutouts, are counting statistics, present because Brodeur has simply been in the game of hockey for such a long time. When you stack his even strength numbers up next to Patrick Roy or Dominik Hasek for what we have available since 1998, they fail in comparison:

Career EV SV%
Dominik Hasek .933
Patrick Roy .928
Martin Brodeur .921

Then you have to take into account the fact that the average save percentage in the NHL actually increased post-lockout. The league combined save percentage weighted by overall shots signifies that an average goalie in Hasek and Roy’s careers would have a .915 EV SV%. The average goalie in Martin Brodeur’s world has a .917 thanks to a higher overall level of goalkeeping after the lockout.

In discussing Roberto Luongo at Canucks Army, I created a measure for comparing goalies historically called SV%+. By taking league average goaltending, you can find whether a goalie performed significantly above or below average when looking at the two. I’ve created it so that a SV%+ of .900 is exactly average:

EV SV%+
Dominik Hasek .920
Patrick Roy .914
Martin Brodeur .904

Here you see that Brodeur was only slightly better than league average through the 14 available seasons. Remove his last two, where he has put up below average numbers, and you’re left with an EV SV%+ of .906. It’s not terrific.

But here’s the thing: Martin Brodeur has done it for 14 seasons. For no other goaltender do we have such a blend of available statistics to judge and infer based on the opinion of hockey scorekeepers from 1993 to 2012. Roy never played post-lockout, and Hasek’s best years are obscured by a lack of even strength shot statistics.

Take the best seasons of SV%+ and line them up on a spreadsheet, and Brodeur’s name first appears 40th on the list, thanks to a very good 1998 season where he had an EV SV% of .928 with a league average of .915. This makes his SV%+ worth .914, tied with Byron Dafoe in 1999, Ron Tugnutt in 2001 and Ray Emery in 2007. But these are all blip seasons. Brodeur is a model of consistency. Slightly above-average consistency, and in the past two seasons he has been below that. He is getting old, but the counting statistics, the wins and Stanley Cup appearances, continue to grow.

While Brodeur is first in career wins, he is also first in career losses. Being the active leader in goals against average also means he is the active leader in goals against. Mike Smith and Brian Elliott both put up fantastic years this year, but they aren’t good enough as Brodeur to continually put up respectable numbers for years.

Brodeur will end his career a little more overrated than he ought to be, but nothing is obscured for him. Brodeur broke Sawchuk’s records, but we don’t know under what circumstances Sawchuk initially set them. No numerals, just stories, can help explain exactly what Terry Sawchuk exactly was.

What we know about Brodeur is that he is the only goalie to have been a little above average nearly every year of his career since the beginning of the 1990s. The other thing to know is that whatever it is Brodeur has been doing, he’s been in decline the last two seasons. He was 27th among starting goalies in EV SV%. James Mirtle even described the Devils’ goaltending as a weakness heading into the Stanley Cup Finals.

Overrated, perhaps. He was never great, but he was consistent. The greatness of a goaltender like Jonathan Quick has been replicated many times before by goaltenders, but never for 12 consecutive seasons. For this, Brodeur deserves credit, even if it means that the Devils have a clear disadvantage on the goaltending front going into these Cup Finals. Because it’s his career as a whole, not a single series or game, that will shape him.

Comments (87)

  1. If you’re going to use EV SV%, there should be some acknowledgement that Brodeur plays in New Jersey, where shots are undercounted perhaps more than anywhere else in the league. For instance, Brodeur’s road SV% overall is .004 higher than his home SV%. Luongo’s home SV% is .004 greater than his home SV% – he played for several years in Florida, which inflates shot totals.

    Brodeur still doesn’t hold up to Roy or Hasek, but he looks better.

  2. I have been saying this for years, nothing against Marty…a good consistent goaltender, but not great as many people feel.

  3. All of this analysis goes to show one other thing that modern statistical analysis cannot explain (or, at the very least, I’ve yet to discover)… It’s not about the “Best” goalie, it’s about the right one.

    Brodeur absolutely fit what New Jersey has tried to do what New Jersey has done since they drafted him. There’s a reason Lou traded DOWN to get Broduer instead of taking Trevor Kidd.

    This is a popular chicken / egg argument: Was New Jersey better for having Brodeur, or was Brodeur better for being with New Jersey.

    It’s a little of both, but considering how good the Devils have been for so long, with everyone anticipating their demise… the one constant is still Broduer. So I tend to believe the Devils are/were better for having Marty in net vs. someone else.

  4. I think you hit the nail on the head when looking at the stats you have presented. I believe Marty is really just an above average shot blocker. It is the other areas of the game that makes him great and enabled him and his team to win so many games.

    His stick handling ability is second to none and it is hard to argue that the league implemented the trapezoid without thinking about how it would affect his game. The Philadelphia-NJ series this year epitomized the way his playing the puck can simply shut down a high powered offense, without needing to make saves.

    He also has an unbelievable ability to read and change the pace of the game. He has a basketball coaches ability to stop play by freezing pucks and getting rests and timeouts for his team. He will also try to speed up the game when his team seems to be rolling.

    As you summed up shot blocking is not the strongest part of his game, throughout his career he has faced goalies that are much better than him (i.e. Hasek, Roy ). That statistic is not what makes Marty, Marty and hopefully someone will be able to come up with a measure to track many more of those current intangibles.

  5. First off that is a picture of Sean Burke.

    Secondly, a huge reason why NJ has traditionally given up less shots is his puck handling ability.

  6. I am admittedly bias since Brodeur is my favourite hockey player of all-time, but I do believe that advanced stats like EV SV% fail to take into account Brodeur’s true value. They measure his abilities as a puck-stopper, in which case I admit he is only slightly above-average, and not as great as people think. But they fail to take into account the following points:

    - His puckhandling easily cuts down on at least 4-5 shots a game. That’s a conservative estimate. It also changes the way that the Devils’ defencemen play around him, and helps launch the Devils’ counter-attack. The Devils have built their style of play around Brodeur for 19 years now. That’s no accident.

    - As one of the above posters noted, he has an intrinsic feel for the game that is difficult to measure. If he senses his team could develop an odd-man rush or has a speed advantage on the ice, he’ll kick out a big rebound to the center of the ice to lead the forwards up the ice. Sometimes this backfires and turns into a giveaway and a goal, but for the most part it’s a net plus.

    - This last point is somewhat intangible, but as someone who’s watched Devils’ games for the better part of 15 years, I feel confident in its validity: Brodeur’s old-fashioned, stand-up style that demonstrates athleticism and tremendous heart actually INSPIRES his teammate, as a great leader would. I’ve seen momentum switch in Devils’ games after a stack-pad save by Brodeur that got the crowd on its feet, whereas a statistically superior puck-stopper like Roberto Luongo simply doesn’t inspire his teammates when he crouches down in the butterfly position for the umpteenth time and the puck hits his oversized equipment.

    • Like Geoff said, cutting down “at least 4-5 shots per game” is certainly not a conservative estimate.

      Take a look at http://nhlnumbers.com/2012/5/27/the-impact-of-a-goaltenders-puckhandling-skills — over the last few years, Brodeur has faced about one shot less per 60 minutes than his backups have.

      If anything one shot per 60 minutes is probably a high estimate — other top puck-handlers like Turco, Belfour, and Smith see an effect less than half that large.

      Brodeur’s puck handling might be worth a couple of tenths of a percent on his save percentage, but is nowhere near the magnitude effect you have suggested. Maybe it helps the offense get more shots at the other end, but it’s a minimal effect at the defensive end.

      • Fair enough…good to have metrics on it. But it doesn’t alter my other two points or my overall contention that there are good “good-team” goalies and good “bad-team” goalies, and Brodeur is in the former category. Save percentage favours the latter category.

        Here’s why I don’t buy into the argument of SV% or even EV SV% or Corsi or anything else: as a goaltender, it basically means that in order for you to feel as if you are the best at your position, you have to hope that your team isn’t that great.

        Instead of doing what Brodeur does, which is to help them be great and help them win. But the doubters will keep popping up.

        • “Save percentage favours the latter category [bad-team goalies].”

          Dang, that makes Tim Thomas’s run last year even more impressive. And Lundvqist this year. And Luongo’s career. How did Patrick Roy manage such a consistently high save percentage on winning teams, anyway?

          You know, come to think of it, it seems like there are lots of goalies with high save percentages on good teams. What exactly makes you assert that it’s easier to have a high save percentage on a bad team?

        • I don’t exactly follow how you reached that conclusion in your second paragraph, but if your assertion is along the lines of “SV% means you have to be on a not-great team to be a great goaltender”, I don’t agree with that at all.

          Look at the Rangers this year. They received a ton of praise for being this scrappy, hard-nosed, defensive, shot-blocking team, and they were the best team in the Eastern conference. That didn’t stop Henrik Lundqvist–he of a .930 save percentage, tied for tops among regular starters–from getting universal praise for his incredible season. He’s probably going to get the Vezina. All behind a team that was a point away from the President’s Trophy.

          Look at Jon Quick, even just in this postseason. The Kings are destroying everything in sight but Quick is receiving praise…and he’s got a .946 save percentage, best of any goalie in the playoffs.

          Those are just two examples, but I think I made my point.

          • You are of course rationally right that save percentage is a better indicator of which goaltenders are playing well. But they are not necessarily the be-all-and-end-all.

            Look at Tomas Vokoun’s career save percentage. It has consistently been higher than Brodeur’s, which I suppose makes him a better goaltender. But good luck trying to convince anyone, even Vokoun, that that is the case.

            Vokoun thrived in Florida where he was being peppered by 40 shots a night, could allow 3 or 4 goals, lose 3-1, and go home as the team’s best player. When he went to Washington, a Cup contender that was supposedly only a premier goaltender away from greatness, he didn’t do so well…even though his save percentage was actually quite respectable this season (and better than Brodeur’s).

      • The problem with using his backups as a comparison point is the quality of competition argument. From 1997-2007, he played at least 70 games a season. Without looking up the actual stats, I can say quite confidently that his backups were not playing the better opponents.

        • Yes, and that’s why the linked article looked at the last four years, when he played considerably fewer games and quality of competition was much less likely to be a factor.

          Incidentally, the magnitude effect over those four years was exactly the same as what an article at brodeurisafraud found from looking at the prime of his career.

  7. Maybe now Pizzo can stop definitively calling Marty the Best Goalie Ever on every podcast.

  8. So… If a player is consistently good at his job and has done the same thing that is expected of him over and over again for 14 seasons….

    - Has the Cups, trophy’s, All-Star nods, Olympic accolades that people swear make you a top-quality superstar…

    - Plays the vast majority of his games for his team year in and year out…

    - Is so adept at one particularly underutilized aspect of his game that the league changes the rules and alters the playing surface design to curtail this aspect…

    - Is 4 wins away from winning a Cup in his 3rd DECADE AS A PLAYER…

    - And is unquestionably going to be in the Hall-Of-Fame…

    Everyone’s right… he’s not all-time great. Dude has just been lucky to have “simply been in the game of hockey for such a long time”. Especially on a super-stacked team like the Devils have had for all these years.

    • No one said he’s not an all-time great. But he was rarely top-three, let’s say, in any one season. Hasek was by far the best in several seasons.

      Check out Ray Whitney. Over 1000 points, but never top-ten in scoring. He did it with games played and consistency, sorta like a Messier-lite (Messier if memory serves has the most games played of any player and that’s probably what pushed him to second in career points).

      • Notify me when Whitney passes Gretzky in nearly every statistical category and then you can use him as a point of comparison.

        • Brodeur played many more games than Sawchuk, which is where the comparison comes in. Sawchuk’s teams played 70 game schedules for the most part; then in the final ten seasons of his career, Sawchuk played 50 or more games only once. Also, Brodeur usually faces fewer shots per game. You wear down a lot less over the course of a season facing 22 shots per night, than you do facing 32. It’s a wider margin of error: if you have a poor night and only stop 19 shots, your team can still scrape a win. That same save % over 32 shots is four, maybe five goals. So a large amount of Marty’s success is in having a defense to limit opposing opportunities.

          Sure, he can only stop what he sees; he’s done that admirably for a long time, and should get credit for it. Maybe if he had more seasons like 2006-07, where he proved he could handle a heavier workload, he would get the “proper” credit. He’s a great goalie, but he’s not the best over. He’s the most accomplished, which is not quite the same thing.

          • Sawchuk averaged 2.5 to 3+ GAA for the majority of his career. He had 5 great years with the Wings, but he also had 3 sub .900 sv pct years. I have a little trouble with the argument that seeing 10 more shots a game when you play ~50 games a year means you’d be more worn down than a goalie that played 70+ games per year for most of his career. Post-Stevens, it was extremely rare for Brodeur to see only 22 shots a game. Also, for a significant portion of Sawchuk’s career, shots against weren’t tallied, so we do not actually know how many he faced.

          • How are there any numbers at all for Sawchuk? The league only started tracking shots in the early 80′s. I know the hockey summary project has been doing Bossy’s work in gathering up all the boxscores in league history, but I have yet to see the numbers compiled into full statistics… unless you’re doing all of this by hand.

            Even if Sawchuk was sub-.900 a few times, how does that measure against the rest of the league?
            Roy once led the NHL at exactly. 900. It’s impossible to say, absent more complete data, that Sawchuk was really subpar just because his save % looks low to eyes spoiled by 20 years of .920+ league-leaders.

            For the GAA, Sawchuk’s numbers did jump. League scoring as a whole went way up, in fact: from 2.4 goals per team, per game in 1952-53, to 2.8 GPT/PG in 1957-58, and onward to hover right around three for the remainder of his career. Part of that was that he himself was allowing more goals – one goalie skews a six-team sample a lot more, obviously – but it was also the beginning of a time when goalkeeping was on a down trend. Expansion would put the keepers further behind in 1967.

            Again – it’s not that Marty’s terrible. He’s excellent. He’s just not the final word in goalies.

  9. The question that needs to be asked is what makes a goalie great?

    is longevity not a trait of a “great” goalie?
    is consistency not a trait of a “great” goalie?
    what about poise?
    what about timeliness?
    How about importance to your style of team?

    are stats really all that goaltenders should be judged by?

    I have played goal my whole life including 6 years in Juniors and College and there are three things people should know:

    1 – Unless you have played goal at a high level you really don’t understand the position at all (you can push stats all you like but they don’t tell the real story of being a goalie)
    2 – Being a valuable goaltender is as much about your intangibles as it is your stats.
    3 – its a whole lot easier to play on a crappier team.

    You can quote as many stats as you want but at the end of the day the goaltender is an entity unto themselves.

    Obviously goaltenders like Sawchuk and Roy are Great goalies but so too is someone like Chris Osgood.

    If you are to break Osgood down by stats then he is simply not a very good goalie.
    Career RS .905 SV and 2.49 GAA and PO .916 SV and 2.09 GAA

    But what did Osgood mean to the Red Wings? Everything. His stats were sub par but his intangibles were almost unmatched. 3 Stanley Cups including his 2008 Stanley Cup run is EXACTLY what makes him great. Playing as backup to Hasek (who you might note tops all your wonderful stats) he comes in and is an absolute rock and it was he who provided the Wings with the defensive confidence to play the the way they needed to to win. After all any hockey player will tell you that you take care of your own end first and when you trust your man between the pipes you play a hell of a lot better upfront.

    Take my list of qualifications for being great

    Longevity – 18 NHL seasons

    Consistency – He was never fantastic but Osgood was always pretty good

    Timeliness – 2008 Game 6 save to basically ensure the cup victory is what being a great goaltender is all about. Getting the Big save at the right time.

    Poise – Osgood is famous for being mentally unflappable and his cup rings prove it

    Importance to team style – Osgood typifies what this quality is all about. Was he great? no. was he EXACTLY what the Red Wings needed? YES! The Red Wings are the epitome of how Consistent Mediocre Goal tending that comes up big at the right times wins championships.

    At the other end of the spectrum you have someone like Luongo

    Great stats and a Gold Medal but far from great.

    Longevity – 12 seasons which is good

    Consistency – just not there. Luongo can be the best goalie in the NHL on any given night but he can also be one of the worst on any given night. players don’t want this. Players would rather have a consistent 8 than an occasional 10 who frequently is a 6

    Timeliness – Luongo has so many highlight reel saves but how many of them have been outright game savers? they usually come in a 5-2 game with 4 mins left in the 3rd.

    Posie – Luongos playoff lapses are legendary and his performance in the goal medal game was anything but confidence inspiring. Luongo almost never comes up with the big save at the right time because when the pressure is on you can tell that it is getting to him.

    Importance to team style – With the Panthers Luongo was everything they needed. A fantastic goalie who would steal them games and be counted on to make 40+ saves a night. But the Canucks? they didn’t need a game stealing goalie. They needed consistency and Timeliness which Luongo did not deliver and Cory Shnieder does.

    after all is Shnieder as good as Luongo? No absolutely not but he is EXACTLY what they Canucks need right now.

    Also,

    it is FAR easier to play for a weaker team that has less expectations and fewer mentally challenging situations because if you stop 42 out of 45 like Luongo did frequently in Florida then you are a hero for giving your team a chance to win. but apparently when you make 20 out of 24 saves like Brodeur then you are “a little above average”

    The discrepancy is that Luongo making 42/45 was giving his team a CHANCE to win
    While Broduer stopping 20/24 was helping his team TO win.

    Not to mention that when you are being shelled by 35-40 shots a game it is FAR easier to stay in the game and make 10 huge saves to keep your team close.

    When you face 20-25 shots a game its hard to keep yourself in the zone to make the 1 or 2 big saves needed to win the game.

    Bottom feeders love goalies like Luongo who steal games
    Cup Contenders with always do better with consistent good goaltending because quite simply they shouldn’t need their goalies to steal them games.

    Poise and Consistency and ESPECIALLY timeliness matter more to great teams than any stat line. That is why Broduer is the best of all time. because quite simply at the end of the day you need a goalie who will get it done when it matters most and Broduer has always done that better than anyone else.

    • Graham, Very well put. I agree 100%. Stats are not everything.

    • Ha, wow. This comment was a thousand times better than the article. A million thumbs up.

    • This comment is long and complicated, and I disagree with almost all of it. I’ll pick a few specific examples:

      “Unless you have played goal at a high level you really don’t understand the position at all (you can push stats all you like but they don’t tell the real story of being a goalie)” — It’s nice to start your comment off with a declaration that nobody else in the conversation has any authority for disagreeing with you. You should probably write a letter to several NHL teams (the Blackhawks, Blue Jackets, Canadiens, Predators, etc) telling them that since their goaltending coaches never played goal at a high level, they don’t understand the position.

      “If you are to break Osgood down by stats then he is simply not a very good goalie.
      Career RS .905 SV and 2.49 GAA and PO .916 SV and 2.09 GAA” You don’t have to rely on stats to come to this conclusion. The NHL teams themselves made it pretty clear that they did not think he was anything special. The Red Wings could not get anyone to give up anything for him in trade and eventually left him unprotected in the expansion draft. Then the Islanders traded him with a 3rd round pick for Justin Papineau and a 2nd round pick. Then the Blues opted not to re-sign him, and Detroit was able to sign him for a measly $800k per year. His next contract with them was for $900k per year. And a year or two later he was suddenly considered a clutch Hall of Fame goalie for some reason — it’s strange that no team in the NHL could see what is apparently so obvious to someone with your played-at-a-high-level expert vantage point.

      “3 Stanley Cups including his 2008 Stanley Cup run is EXACTLY what makes him great.” — I’ll cite Brodeur Is A Fraud to emphasize how much of a role his incredibly stacked teams played in his playoff career: in playoff series where his team had a 20+ point advantage over their opponent, he was 7-1; in series where his team had less than a 20-point advantage he was 8-8. Oh, and they were 9-4 in series where he was on the bench as the backup goalie.

      “Consistency – He was never fantastic but Osgood was always pretty good” — Again, we’re talking about a goalie who was waived, traded for peanuts, and signed for under a million bucks twice, all after his second Cup win. Nobody ever thought he was any good until that third Cup run when suddenly the history books got completely re-written.

      “it was he who provided the Wings with the defensive confidence to play the the way they needed to to win” — I had no idea that Lidstrom and Datsyuk and company owed so much of their defensive prowess to Osgood providing them with confidence.

      “When you face 20-25 shots a game its hard to keep yourself in the zone to make the 1 or 2 big saves needed to win the game.” — And yet we don’t see save percentage dip when goalies go a long time between shots (see: http://www.arcticicehockey.com/2009/12/3/1110764/do-goalies-need-to-face-shots-in).

      “Cup Contenders with always do better with consistent good goaltending because quite simply they shouldn’t need their goalies to steal them games.” — And yet no goalie has shown the ability to play more consistently than any other, and they probably aren’t any more or less consistent than random chance (see: http://nhlnumbers.com/2012/5/11/the-myth-of-the-hot-goalie-consistent-goaltenders-vs-inconsistent-goaltenders).

      • Eric’s comment is much more accurate then Graham’s and the difference is between the two is at the heart of ‘stats v. observation’ debate.

        The ‘observation’ people can not overcome two problems. The first is when they make an assertion like this:

        “Luongo has so many highlight reel saves but how many of them have been outright game savers? they usually come in a 5-2 game with 4 mins left in the 3rd.”

        Well, prove it. Prove that Luongo only makes saves when there is no pressure on him. Because frankly, I’m not going to just take your word for it. Now your stuck, because the only way to prove or disprove this statement is with stats.

        The second problem the observation people have is that stats, properly deployed, have preditive aspect to them. Observation analysis are all post hoc.

    • Loved this response. This is exactly why Brodeur is great. Thank you Graham for saying what I tried to express but could not. And if Brodeur pulls it out and wins again this year his legend will be forever cemented in hockey lore.

    • I stopped reading after Osgood is great. Chock full of some great clichés though.

    • Precisely what I’ve always thought. Ken Dryden, who should know a thing or two about the subject, puts it perfectly in “The Game”: there are good “good-team” goalies and good “bad-team” goalies, and they aren’t the same thing. The good “bad-team” goalies will always get the love, but they rarely actually get the job done.

      I remember the same statheads used these arguments to suggest Curtis Joseph should be Team Canada’s 2002 starting goaltender over Brodeur. We all saw how that turned out. And then when Joseph went to the Red Wings, who were just coming off a Stanley Cup win, it seemed self-evident that they would win a Cup.

      Joseph + Cup Powerhouse = Sure Cup

      Didn’t happen.

      • Didn’t Detroit score all of 6 playoff goals in the 2 years Joseph was in net? I think the blame should be shifted to Detroit’s inability to score instead of goaltending (which was well above average)

        • They had six goals in four games his first year, when Joseph had a 2.08 GAA as the team got swept because he was a LOSER BUM who lost 2-1 in triple overtime in game one because he wasn’t clutch enough to make the one save they really needed.

          The next year, they scored a whopping 18 goals in the nine games he played, as he posted a 4-4 record despite a playoff-leading .939 save percentage and 1.39 GAA. Once again, he was incredibly unclutch and lost overtime games 2-1 and 1-0. Why can’t you stat nerds see that it’s not about how many dozens of shots he did stop; it’s about the fact that he didn’t make that one save at the key moment in overtime?

          [Sadly, I feel compelled to point out that the above is heavily laden with sarcasm.]

          • I knew someone would bring up the lack of goal-scoring point. But I watched the 2003 Ana/Det series, and watched the analysis afterwards. Even if he was losing 2-1, there were some BAD goals in the mix. Joseph was getting heavily criticized…that’s a fact.

            For x, y, or z, Joseph couldn’t get the job done when it mattered most. In 1996 at the Team Canada World Cup, he gave up three goals in the final eight minutes. In 2001 in the semifinals against Brodeur and the Devils, he outplayed Brodeur in the first five games, yes, but then gave up a combined 9 goals in Games 6 and 7.

            I loved Joseph, he was a great goaltender. But like Luongo, he just couldn’t quite get the job done when it mattered most. Is it fair that we judge athletes this way? Absolutely not. If I was a general manager, I would use all of the advanced metrics I could to come to accurate conclusions about players.

            But as a fan, it’s fair to assess careers this way. Chamberlain was a statistically superior player than Russell…but ask anyone who saw them both play for a decade who was better, and it’ll come out about 70-30 Russell. Marino was a better statistical quarterback than Montana. I could go on…

          • Oh by the way, just as one last point, despite all my Brodeur defense and the fact that he’s my favourite hockey player ever, I actually DO NOT consider him the greatest goalie ever. That honour would still go to Patrick Roy, whose three Conn Smythe Awards give him the nudge over Brodeur’s longevity.

    • Wanna know another goalie who has average/mediocre career numbers, but who has a ring and who, just this past April, people claimed gave his team confidence and an edge in the postseason?

      Marc-Andre Fleury.

      That sure was prescient, wasn’t it?

      It’s always amusing how people talk about intangibles as something that are quantifiable and, y’know, tangible. It’s also always funny to see which guys allegedly have intangibles/instill confidence in their teams and which one’s don’t. Cam Ward has a ring. Antti Niemi has a ring. Why doesn’t anyone talk about how confident they made their teams on their way to a Cup?

      • The problem with attributing cup rings to goalteners is that they are just one position.

        Brodeur has 3 cups so he is great.

        Kirk Maltby has 3 cups yet he’s not listed as an all time great forward.

        Winning the cup involves a few things, one of which is getting good goaltending for a stretch of about 20 games. There’s very few teams who have won in the playoffs while getting subpar goaltending for the playoffs.

        Look at Cam Ward back in 05-06. He was rather pedestrian during the regular season, came in for Gerber in the playoffs and went on a run. Next season, he was pedestrian again.

        So yes, Brodeur has 3 cups, but he’s also played on a fantastic team for a long time, which afforded him the ability to string together some great playoff runs (much like Osgood). Pekka Rinne and Thomas Vokoun never got those opportunities.

        It’s like scoring clutch goals. Jordan Eberle has a reputation for scoring clutch goals. Why? Because he’s a good player who happened to score a bunch of goals in highly publicized games. They make mention of his goals vs Russia and the US, but how many times in Regina did his shot go wide in the dying seconds of a game with his team down by 1?

        Opportunity is a greater factor than ability when it comes to winning. Many great players never win the big event simply because the opportunity hasn’t been there.

  10. Pretty sure these stats were not be hind the devils Defensive style.

    Brodeur was selected as Team Canada’s back-up goalie to Patrick Roy for the 1998 Winter Olympics in Nagano, Japan; but did not get to play. Canada failed to win a medal after losing the bronze-medal match to Finland, a game in which many people thought Brodeur should have played.[48]
    In the 2002 Olympics at Salt Lake City, Utah, Brodeur was initially named the backup behind Curtis Joseph, but following Joseph’s losing the tournament opener against Sweden, Brodeur was named the starting goaltender the rest of the way, and won gold for Canada. He had the best GAA in the tournament and went undefeated, stopping 31 of 33 shots in the gold-medal victory over Team USA.
    Brodeur then led Team Canada to a World Cup of Hockey championship in 2004, allowing only 5 goals in five games. He led all goalies in GAA and save percentage while going undefeated. He had another impressive performance for the team at the world hockey championships in the following year. After this, The Sports Forecaster 2005–06 said the following:[4]

    • Ooooh! We’re posting Wikipedia paragraphs! Can I post the next two? I figured you would but I don’t see them there.

      Brodeur was selected as Team Canada’s starter in the 2006 Winter Olympics in Turin, Italy. He started in 4 of 6 games, but Canada failed to win a medal after losing to Russia in the quarterfinals.
      He was one of the three goalies on Team Canada for the 2010 Winter Olympics in Vancouver, Canada. He registered a shootout win against Switzerland and a loss to the United States. After the loss to the US, he was benched for the remainder of the 2010 Games in favor of Roberto Luongo.

  11. Here’s another thing Cam!

    The Vezina Trophy is awarded annually to the National Hockey League’s goaltender who is “adjudged to be the best at this position”.[1] At the end of each season, the thirty NHL general managers vote to determine the winner.

    Please note, best at the position. Not best Sav % Considering the GM’s vote I think they may have an idea of what there talking about. AT least a heck of a lot more than you and I. I think there’s more to goaltending than just sav%,

    Every player that’s ever played with Brodeur always says the same thing. It’s great knowing he’s back there because we feel confident doing are thing up here because we can count on him to make “THAT SAVE” That is what makes him the greatest. Hey, I’ve seen him make some serious blunders and let in horrible goals, but again what makes him great is he doesn’t let it bother him. He comes right back calm and relaxed and makes a great save.

  12. As a Devils fan, I’ve heard a lot of reasons over the years as to why Marty isn’t as “great” as people proclaim him to be.

    First of all, he was a different player pre and post-lockout. Before the lockout, the Devils allowed so few shots against that he didn’t have to make 40 saves per night. Save percentage, then, becomes a flawed stat, as teams take different approaches to score against the Devils. Save percentage also doesn’t consider that one of the main reasons Brodeur didn’t face as many shots against is his ability to handle the puck.

    Also in contrast to your argument is that Marty had some of best seasons, in terms of saves overall and save percentage after the lockout. Statistically speaking, he was a better goaltender from 2005-2010.

    • You’ve hit the nail on the head. Everyone talks about how Brodeur was lucky to play with Stevens, Niedermayer, and Daneyko – but many of his best statistical seasons were after those guys were gone.

      How many rules were changed because Roy and Hasek gave their teams an unfair advantage? None? Ok.

  13. Isn’t the fact that Broduer has been so consistent for so long a reason to consider him great? He is different then the all time greats – Broduer handles the puck like a third defenseman – he is always calm and poised – though he’ll let in a weak goal every now and then – he is always clutch – he always makes the big save. I think this argument is cute, but also stupid. Haters gonna hate.

    • Yep, there are still a lot of people out there who hated the “dead puck era” and are looking for reasons to shoot Brodeur down. Fact is that Brodeur holds pretty much every record in goaltending, was incredibly consistent, and has had a rule made to limit his advantages. Can you make an argument that Roy, Hasek or Sawchuk were better? Yeah, probably…but it’s a harder and harder sell as more and more of their records get handed over to Brodeur – who is 4 wins away from his 4th cup.

  14. I find it somewhat ironic that in hockey this ‘Stats v. Observation’ debate has very often centered around Broduer, who plays for arguably the best ‘sabermetric’ organization in sports.

  15. You massage all the stats you’d like, but I’d like to see you make saves like this when you are 40 years old – http://www.youtube.com/watch?v=9flyzH9jImQ

    Honestly, he has more points in the playoffs than most players and he’s the _goalie_. You completely throw away his significant puck handling abilities and poise under pressure. I’m sure conservative goalies like Lundqvist and Bryz could have a word or two about how difficult it is to be so good at that.

    You don’t accomplish all he has by merely being “above average” for a long time. Nothing takes away from the success of Roy or Hasek, but give Marty his due. He’s absolutely one of the greatest of all time.

    • What poise under pressure? I’ve seen Marty melt down any number of times. I’ve seen him give up howlers. Heck, Brodeur’s teams have gone out in the first round seven times – including twice as the East’s first seed, and once as its second seed. Nobody’s perfect. And a “clutch save” is just a save before your team scores a goal. If your team doesn’t score, and if they give up an empty-netter or something, then that exact same “clutch save” is “stat padding.”

      IOW, a “clutch save” happens a lot more often with a team good enough to make it into a clutch save. The Devils have had quite a good team for a long while now. Give him credit for being an important part of that, but people talk as if the Devils would be Columbus or something without him.

      • A lot of people should read The Game, specifically the chapter on the mental side of being a goaltender, which touches on a lot of what we’re talking about here. I’m sure the stats-heads would say that Dryden is only saying what he’s saying out of self-interest…he was a good goalie on a fantastic team. But believe me, he’s more than modest about his own contributions…he just recognizes that there are moments where the goalie’s save can mean more than it appears.

        • I have read The Game, and it’s superb. I find that the psychology is accurate. And it really really annoys me to play a team even for 55 minutes and be on the verge of stealing the thing, and then get beat at the very end to lose. Nobody wants to be that guy. So I actually do agree with all of that – from an interior perspective. Being a goalie almost forces you into that perspective, because otherwise how could you keep on? A goalie can never win the game, ever. Any goal he contributes to is likely a fluke. He can only lose the game by giving up goals. Mentally, it is a huge help to rise to whatever occasion you get – to know that you’re not messing up your team’s great work in front of you.

          But as an objective evaluation, no matter how I personally *feel* about the game as I’m playing, I *know* that those “big saves” that boost my team, or those dagger goals that lose the game, only become that in retrospect. For a guy like Brodeur or Dryden, there is probably some merit to the narrative: otherwise it’s easy to go too far the other way and say that they just coasted to wins because they played on powerhouse teams. Not everyone could do what they did for Montreal or New Jersey. Fair’s fair.

          • Well put. For someone who in this post has been so much a defender of Brodeur’s, I actually do think that advanced statistics should be used by teams when they are making personnel decisions.

            However, as a fan, it’s becoming more and more ludicrous to argue that Brodeur has not had a fantastic career. And given that some goaltenders, like poker players, just seem to get back to that final table no matter what variables changed, while others who may technically be better seem to lack that quality, one can’t help but make grand proclamations about “clutch abilities”, etc., even if they’re objectively inaccurate or misleading.

            Hockey is not baseball. It’s not a series of individual confrontations masquerading as a team game. Like basketball and football, it is far more difficult to measure. A goalie in hockey is like a quarterback in football…it’s entirely possible that Dan Marino was, in every measurable aspect, a better quarterback than Joe Montana. And it’s quite possible that Roberto Luongo has been for the last ten years a “better” goaltender than Martin Brodeur.

            But fairly or unfairly, that simply isn’t the perception, and I think one of the failings of the “stats guys” is that they forget that the appeal of sports often lies in that which can’t be measured. Brodeur has been lucky to be with New Jersey, just as Dryden was lucky to be Montreal and Fuhr was lucky to be with Edmonton. It doesn’t make their careers any less great.

  16. How do you quantify his stick-handling ability, especially in terms of how it affects the number of shots he faced?

    If a goalie gets to the corner and sends the puck up ice before the forecheckers can arrive, is he less of a goalie because he ended up not facing a shot on that shift? Or is he a better goalie because he’s keeping the puck from spending more time in his zone?

    Only guess I have is to look at assists–4th all time regular season (40, behind Barrasso, Fuhr, and Roy) and 2nd in the playoffs (12, behind Fuhr). Not bad for a guy who never played with Gretzky or (Mario) Lemieux.

    Also, is shutout really a counting stat, or just dismissed summarily because it doesn’t fit the “Brodeur is overrated” argument? Is playing 60 minutes without allowing a goal not indicative of good goaltending? Why not convert shutouts into a rate stat? It actually helps your argument (Hasek had shutouts in 11% of GP; Brodeur 10%). Of course, in the playoffs it jumps to 12%, edging Hasek and leaving Roy far behind.

    • Shutouts are affected by the number of shots a goalie sees. It’s harder to stop forty shots than twenty. Just as a for-instance; Brodeur has 22 RS SO out of 119 where he faced fewer than 20 shots. Hasek has 12 of 81. Luongo has 3 of 60. Brodeur has only two career shutouts of 40 or more shots (RS). Hasek only has one, so there’s a point for Marty. Luongo has three. Over 82% of Marty’s RS shutouts (97 of 119) were fewer than 30 shots. Now that’s still second all-time by itself, so again, point for Marty. Also, point for the wonderful defenses he played behind. Hasek’s percentage is 60.4 (49 of 81); Luongo’s is 55.0 (33 of 60).

      Including the playoffs, Marty’s ride gets easier. Brodeur has three career playoff shutouts of 30 or more shots, same as Hasek (with ten more shutouts) and one fewer than Luongo (who only has one playoff shutout under 30 shots). 17 out of Brodeur’s 24 (and counting) PO SO were fewer than 25 shots.

      It’s hard slogging through the game logs, so I hardly have complete records, but I’ll keep checking. Again, we’re not making Brodeur out to be a four-flusher or anything, but his teams really take good care of him, at least as good as he takes care of them.

      • “It’s harder to stop forty shots than twenty”

        And yet goalies always say the opposite. That it’s very hard to stay in a game when you aren’t consistently being challenged.

        • I think that while “clutch” play is extremely hard to quantify, it can be agreed that it means playing your best in the biggest moments.

          In this category you can’t argue with Brodeur’s statistical record during his team’s cup victories.

          12-5 GAA: 1.65 SV%: .927 SO: 3 (including game 7 in 2003 in one of the best goalie battles of all time vs. Giguere)

          Yes Marty has had some poor performances, every goalie does (Carolina game 7, 2001 cup final), but the long record of success and rising to the moment outweighs the bad moments. Is he the greatest ever? Maybe not, but he is certainly in the conversation. If you question his ability as an exceptional puck-stopper, all you need to do is look at his highlights on youtube. They are second to none (with the exception of perhaps Hasek) in degree of difficulty. He has made 2 or 3 alone in these playoffs.

          • Even watch the top ten plays of the last round posting on this very sight. I believe Brodeur had 3 of the 10.

          • Your argument in favor of Brodeur is:

            Look at how well he played when he had the best team in the NHL in front of him. That shows he’s clutch, because when his team was the best in the NHL, he had really good numbers.

            Oh, and look at his fantastic saves on YouTube. Nevermind all those goals he gave up which lower his save percentage, look at how spectacular the ones he did stop looked!

          • Geoff Detweiler says “look at how well he played when he had the best team in the NHL”. Hmm…so he didn’t contribute to that team at all? How many people picked the Devils to win the Cup in 1995? How about in 2000? (2003 was probably the closest they’ve been to a Cup favourite).

        • Goalies always say that. It’s how goalies are built. They wouldn’t be goalies if they didn’t look forward to being difference-makers and stopping a million shots and winning. But statistically speaking, yes, it’s easier to stop twenty than forty. Being mentally aware and “in the game” is just fine – but facing twice the shots means twice the opportunity of being screened, having a shot tipped or bang off a leg, or just getting sniped or beaten on a great move. That’s just how it is. That’s why no matter how much the goalie says the opposite, coaches still tell their defense to take the puck away from the other team. No coach in history ever would say “Let the keeper face about ten shots on the first shift to make sure he’s in the game, so he’ll play better.”

          And honestly, it’s a little bit of selection bias in the “better with more work” perception. A goalie isn’t going to face 40 or more shots unless he’s really playing well, or if his team has no other good option. If he’s having an off night, he’s either getting by because he only gets 15 or so shots – thus leading to the “had trouble staying in the game” rationale – or he’s getting yanked before he racks up a lot of shots.

  17. that was meant to read “site”

  18. This article is inherently flawed because its primary assumption is FAILED. The article assumes that a goalie’s greatness should be judged by his save percentage and “[t]he team in front of the goaltender controls allowed shots and scored goals.” In Brodeur’s case, this is WRONG. One of the reasons Brodeur is arguably the greatest goaltender of all time is because his stickhandling skills allow him to come out and play the puck on dump-ins and drastically reduce the opposing team’s total offensive possessions per game. Having MB in net is like having another defensemen. No other goalie in hockey history has or had the ability to do this as well as MB. If you limit the other team’s possessions in your zone, then you can drastically shift the game in your favor, which is a HUGE reason why the Devils have been so successful.

  19. Can we not read? Is there any place in the article that indicates Brodeur is bad? To the point where the contra-argument turns into “Well I once saw … intact, her fit is on youtube…” and the like. The argument, when you actually read the article, states that in EVSV% Brodeur was a little above average. Let’s look at a bit more:

    Overrated, perhaps. He was never great, but he was consistent. The greatness of a goaltender like Jonathan Quick has been replicated many times before by goaltenders, but never for 12 consecutive seasons. For this, Brodeur deserves credit, even if it means that the Devils have a clear disadvantage on the goaltending front going into these Cup Finals. Because it’s his career as a whole, not a single series or game, that will shape him.

    In fact, this statement is telling us that Brodeur is /was /has been pretty damn good. Cam Ward can have a great playoff, it doesn’t make him great. Brodeur has had years and years of pretty good. That has value, and it incorporates all the intangibles that some people want to tangible-ize. What I think the article’s author is really getting at is what Bill James used to describe as the difference between peak and career performance. If you’re going to judge the greatness of a player, do you mean how good was he for one stop on youtube, how good was he for his five best seasons, or how good was he for 15-20 years. Think Mark Fidrych vs. Don Sutton. Do you want a guy who scores 50-60 goals five times in a nine year career and then pretty much disappears, or a guys who lasts 19 years, and averages 19 goals a year (Whitney)? That’s the Brodeur argument – pretty damn good over a very long period of time – in his case good enough to win Stanley Cups if everything else is in place – or someone who can possibly lift you to awesome paces for a short period, then is gone, and has to be replaced with something much less – maybe less than the 19 goal scorer. I won’t presume to try and make the James argument, its strong enough on its own. The point being,Brodeur – good goaltender. Can win Cups with him. If you measure greatness by quality over long periods, he’s great. If you measure it by leading the league in stats that no one knew about when Terry Sawchuk played, then he’s only pretty good.

    Which isn’t bad.

    • The article describes one of Brodeur’s seasons as “very good” and at no time used the adjective “great” except to describe Terry Sawchuk and Jonathan Quick. Charron wrote that Brodeur has been incorrectly associated with greats like Sawchuk.

      So, whatever argument Charron is trying to make, it’s not that Brodeur is great or “pretty damn good” or even “good.”

      Charron admits only that Brodeur has been consistent, but tends to attribute that to the defense even though it changes 50 times per game, more than 50 games per year, over many seasons.

      Charron professes to be a journalist and journalists chose words to convey the desired effect. At no point did he use the adjectives great, pretty good, or good to describe Brodeur across his career, except to differentiate him from greatness.

      So – with apologies to your correct observation that Brodeur has been pretty damn good across his entire career – Cam Charron has not made that point. The astute reader, like you, has had to make it for him because he refuses to.

    • I would also say that even if the point of the article is that Brodeur was “never great, but consistent”, that in itself is flawed. In 2006-2007, Brodeur was in serious discussion for not only the Vezina trophy (which he won), but the MVP award. So either everyone watching the games night-in, night-out doesn’t understand the game, or he’s great.

      4 Vezina Trophies, 4 First-Team All-Star teams, 4 Second-Team All-Star teams playing in inarguably the strongest stretch for goaltenders (1994-2004) we’ve seen since the days of Plante, Sawchuk, Hall and Bower. If that isn’t great, I don’t know what is. Unless again, the author of this article is claiming that the 30 general managers and countless reporters who vote on these awards don’t know what they’re talking about (which given how some GMs operate is more likely than one might think at first).

      Perhaps a better (and less controversial) track for this article would have been to claim that just because Brodeur holds many of the records that once belonged to Patrick Roy, that doesn’t mean Brodeur was a better goalie than Roy (I’d agree with that). And just because Brodeur played for better teams than Hasek, that doesn’t mean that Hasek wasn’t the more talented goaltender (I’d agree with that also).

      But to claim that Brodeur “was never great, just consistent”? Ummm…no.

  20. Cam Charron is from Vancouver – home of the very great Roberto Luongo, that not-ever-porous escape goat. BSH EricT is a Flyers fan. You know the Philadelphia Flyers and their long legacy of competent goaltending.

    So, can either Charron or EricT – the two who are most earnest in their argument that Brodeur is not a great goaltender but merely “a little above average for a very long time” – attest that they would prefer their netminders rather than Brodeur over the past 18 years? Based on his stats, that is. (Not the rings and Vezinas and records and all that).

    If you could have seen Martin Brodeur lace up as a rookie in orange-and-black (the thought disgusts me but since both Vancouver and Philly were Halloweenies in 1994 it works) for 18 years and collect his numbers in that time frame, would you refuse that marginally above-average goaltender over “a very long time”?

    No. I doubt you would, and the reason is simple: maintaining above average stats over a long career with many games played and lots of consistency/ predictability is just the FIRST mark of greatness.

    Have no illusions that greatness is achieved with a .990 SV% and a 0.5 GAA while literally standing on your head and delivering a baby.

    Greatness is taken in small steps and Martin Brodeur has taken them. First, with consistency. Second, by breaking records through his consistency. Third, by winning several trophies with it. Fourth, with puck-handling which disrupts the attack. (Eric argues it’s only by a few % points though his article does not account for pre and post-lockout, stand-up or butterfly differences, home and away, etc). Fifth, by forcing Philadelphia Flyers GM Bobby Clarke to conspire with the Board of Governors to put the trapezoid in place. Sixth, by playing past age 40 into another SCF…

    Cam: you even acknowledge this in your post, but don’t realize it:

    “Here you see that Brodeur was only slightly better than league average through the 14 available seasons… Martin Brodeur has done it for 14 seasons… Brodeur is a model of consistency. Slightly above-average consistency… Brodeur [has] continually put up respectable numbers for years… What we know about Brodeur is that he is the only goalie to have been a little above average nearly every year of his career since the beginning of the 1990s.”

    That’s the story. Consistency. Longevity. Lou Gehrig and Cal Ripken come to mind. Their slugging percentages, OBP, and home runs were not at the top every season, but they were men of iron and their stats were consistently strong.

    They didn’t simply rack up the numbers due to mere longevity. Their consistency and longevity led them to rack up high numbers.

    It’s of no moment to consider the piddling pseudo-stat differences between Brodeur and the goaltenders with better GAA and SV% who are therefore “superior” to Brodeur. Aside from Hasek and Roy… those superlative netminders are 100 miles (161 kilometers) behind.

    The notion that greatness is measured by SV% is also fundamentally flawed because of Cam’s other premise: “My belief is generally that a goalie controls his save percentage more than anything else. The team in front of the goaltender controls allowed shots and scored goals. The rest is up to the goalkeeper…”

    A goalie controls not just his save percentage, but by style, reputation (freezing the puck, showing glove, making acrobatic saves, etc), rebounds, stickhandling/ passing, communication and many other things which may well have an impact on GAA and SV%. Moreover, a goalie cannot alone control his SV%. It, too, is dependent upon his team because just as not all shots are quality (bad angles, etc) not all shots are saveable (one-timers, breakaways, deflections, etc).

    There is another quote which, I think, is worth considering:

    “With that in mind, however, we can all kind of agree that Brodeur’s success has largely to do with playing behind one of the best defence’s in the NHL for a long time: they prevented shots, it wasn’t just keeping the quality of shots low.”

    I agree that the Devils reputation is for defensive toughness, but what’s your metric which shows Roy and Hasek were not nestled behind even stodgier defenses? Echoing the NZT takes us nowhere.

    As Brodeur is a Fraud put it: Roy was a fraud, too, because the Habs and Avs were very good.
    http://brodeurisafraud.blogspot.com/2008/04/patrick-roy-never-singlehandedly-won.html

    I assume Hasek was aided more in Detroit than in Buffalo.

    Since BiaF’s premise is that Brodeur is a fraud and that Hasek/ Roy are superior, he goes to great lengths to demonstrate that point through lies, damned lies, and statistics which, however, do not involve shot quality. I.e., there’s no known comparison between how the Sabres/ Red Wings, Habs, and Devils defenses allowed low-percentage shots from bad angles on net to reach Hasek, Roy, and Brodeur. Without an appreciation of shot quality, it’s difficult to weight Hasek’s much higher career shots-against and relatively higher career GAA and SV%. BiaF uses GAA and SV% because shot quality is an unknown.

    But, at least BiaF has the dignity to admit that Brodeur is, in fact, “an all-time great goalie, not a fraud.” The blog title is, it seems, tongue in cheek.

    Both Cam and EricT have cherry-picked a plausible set of lies, damned lies, and statistics to first tell and then reinforce their narrative that Marty Brodeur is gound beef, not a choice cut. They’re wrong. If Hasek is tenderloin, Brodeur is porterhouse, and Roy is prime rib.

    • I would rather have Roberto Luongo on the Canucks than Martin Brodeur, at least at the point of their careers when Luongo came to Vancouver. Oddly enough, Luongo has been one of the most consistently great goaltenders of the last decade. So yeah, I’ll take Luongo.

      Also, you’re misusing Mark Twain’s famous “lies, damned lies, and statistics” quote. http://canucksarmy.com/2011/11/17/lies-damned-lies-and-statistics

      • Is this serious? I think I’ve just read the stupidest thing I’ve ever read on the internet…

      • Haha, enjoy that Cup-less streak Vancouverite.

      • Luongo over Brodeur?

        I’m not misusing the quote. The quote reveals that statistics are useful only if they’re presented honestly and analyzed properly. In the hands of a deft statistician, they can be misused to prove whichever point the arguer seeks to make.

        The author of the argument you’ve linked to is wrong: very few people, and least of all me, use that quote “as a refutation of all statistics.” I used it to refute the over-reliance on certain statistics which are employed to demonstrate what Charron and BiaF already believed.

        To do so is like using the Bible to “prove” to others the truth of one’s belief in God and the Bible. Other people must accept the Bible as proof in order to believe the arguer. The difference is that while the Bible arises from ancient mythology, statistics arise from modern record-keeping and logic.

        Both, however, can be misused to serve the faith of their believers.

    • “Since BiaF’s premise is that Brodeur is a fraud and that Hasek/ Roy are superior, he goes to great lengths to demonstrate that point through lies, damned lies, and statistics which, however, do not involve shot quality. I.e., there’s no known comparison between how the Sabres/ Red Wings, Habs, and Devils defenses allowed low-percentage shots from bad angles on net to reach Hasek, Roy, and Brodeur.”

      I’ll leave the rest of your comment alone for now and reply to just this part.

      You have it backwards. It’s not that the premise was that Brodeur is a fraud and he went searching for statistics to back that up; it’s that the statistics that are most useful for evaluating goalies generally make Brodeur look less dominant than he is often thought to be.

      He also did try to factor in shot quality at one point — see, for example http://brodeurisafraud.blogspot.com/2009/03/brodeur-roy-and-hasek-debate-rages-on.html — and it generally makes Brodeur look worse, not better (nobody ever said New Jersey’s defense was weak in those years). However, shot quality arguments are largely discredited these days because shot location data shows very slim differences between teams.

      • It’s not altogether clear whether it’s BiaF’s premise, his conclusion, or his secret desire that Brodeur is a fraud. Despite his disclaimer, he named his website “BiaF” and has spent years of his life dedicated to knocking Marty Brodeur off his perch and attributing Brodeur’s success to the Devils defense, the trap, falling asleep watching the Devils, blah blah blah.

        He’s become more moderate and more precise over the years, but the overall trajectory has not changed: Hasek above all, Roy, Brodeur, Belfour, Luongo. That was his conclusions before he began the website and crunched all the numbers to support it. As I said, lies, damned lies, and statistics.

        Nobody takes on the task of BiaF from passion alone. If he did, the website would be called the less provocative “Hockey goalie analytics” or something like that. Whether it’s personal (doubtful that it’s Hasek or Sawchuk’s nephew?), professional jealousy (a fellow goaltender with much less success), a superfan with a preference (Rangers fan, Flyers fan, etc), or simply a statistician-contrarian with a little too much hate in his heart, it’s pretty clear he has an axe to grind with Brodeur and the Devils.

        (I wonder if he’d harbor the same animosity If the Roy and the Habs ran the boring trap? His revisions which account for the effect of Roy playing before a good team have been rather muted).

        Even so, the overall point is that Hasek, and Roy (and Belfour and etc) have comparable stats to Brodeur in some areas and superior stats in others. You are right that BiaF uses “the statistics that are most useful for evaluating goalies,” but there are several problems with that. First, GAA and SV% are imperfect. Second, therefore, they may not be the most useful, but rather they’re the only “the most useful” measure which now exists and for which their is widespread data. Relying heavily on stats is one thing, but leaning upon imperfect stats at the exclusion of all other data is bollocks.

        I, too, reject the validity of mere observation and the reliance on “clutch” playmaking. I also reject statistician diehards. Both are extremists and prove themselves laughing-stocks long-term. I have the stats to prove it!

        At least BiaF has the honesty to admit Brodeur does have consistency, even if he hides the ball about his real agenda.

        • And by the way, if BiaF didn’t have a thinly-veiled “secret agenda” vaguely in the manner I’m describing, the blog would be “Hockey Analytics by Joe Smith.”

          It would not be “Brodeur is a Fraud” by “The Anonymous Contrarian.”

          This isn’t the Federalist Papers. (For you Canadian readers, that’s a foundational document in American history which preceded the Constitution). That he still maintains his anonymity is off-putting to say the least.

      • All I know is this: in 2006-2007, when Luongo was in his first year with the Canucks and Brodeur was still in his prime and having arguably his finest regular season, many pundits referred to Luongo as “arguably the best goaltender in the NHL”, and I took severe issue with that. They made the same arguments BSH Eric T (a Flyers fan…priceless) is making: Luongo did more with weaker teams, Brodeur was insulated, etc.

        And now look at Luongo: essentially run out of town by Vancouver Canucks’ fans (an overreaction, I’ll admit). Out-goaltended when it mattered most. Whereas Brodeur will finish his career playing its entirety for the Devils. Hmm.

        If Brodeur was indeed only “slightly above-average”, just for an admirable length of time, then I’ll put this test to you: from 1997 to 2008 (let’s say Brodeur’s prime), how many goaltenders would he have been traded for in a 1-for-1 deal in any given year? Two? One? None?

        If he was only slightly above average, the answer should be 12 or 13.

  21. By this argument, Cal Ripken Jr is just a decent player, and not one of the all-time greats.

    • Yeah, except you know what the difference is?

      Ripken had three years where he led the whole league — not just all shortstops, but the whole league — in WAR, a very widely-used value metric. Among shortstops, his annual OPS — a widely-used assessment of hitting prowess — ranked 2nd, 1st, 1st, 1st, 1st, 5th, 2nd, 3rd, 2nd, 1st, 8th, 6th, 3rd, and 5th among shortstops before he moved to third base at 36 years old.

      Brodeur, in contrast, finished 4th, 15th, 7th, 3rd, 5th, 21st, 16th, 20th, 19th, 14th, 14th, 11th, 3rd, 8th, 11th, 12th, 33rd, and 33rd in save percentage, the most widely-used assessment of goaltender prowess. Can you see the difference?

      There is an argument to be made in Brodeur’s defense. “Then I guess you think Ripken is merely decent” isn’t it.

      • If you’re going to use WAR, then I could counter with Goalie Point Shares, the formula developed on Hockey-Reference.com, and more of an equivalent than save-percentage (which I could argue is like OPS for hitters in baseball).

        Brodeur finished first in goalie point shares in 2006-2007, second in 2007-2008, and also had a 3rd, a 4th, a 5th and three 6ths. So he was the most valuable goaltender once, top five five times, and top six eight times. Again, if this isn’t great, I guess I need a new definition of the word.

        • In other words, I’m arguing save percentage should NOT be the most-widely used assessment of goaltender prowess.

          • Unfortunately, goalie statistics are always questionable and some reading between the lines must be done.

            For example:
            Two goalies get shutouts on the same night, the sv pct for their games is 1.000 and they each make 30 saves.

            -Goalie 1 faces tons of traffic, good shots from the slot, and odd-man rushes
            -Goalie 2 only faces soft perimiter shots because the opposing team can’t seem to get deep into the zone

            Statistically, these goalies are the same. Which goalie would you want?

            Brodeur may not have always stopped the highest percentage of shots of any goalie, but he’s certainly shown an ability to stop the “right” shots over the course of his career.

          • … What should be, then?

          • Perhaps an even more advanced metric, like Point Shares. (similar to WAR in baseball)

  22. I’m sorry, but you can’t win multiple vezinas and just be ‘slightly above average’. Whoever wrote this article is a tool

    • You can win multiple Cy Young awards and just be “slightly above average”, so why can’t you win multiple Vezina Trophies and be “slightly above average”?

      Look at Denny McLain. Two Cy Young awards, career 101 ERA+, with a negative 1.2 WAR over ten seasons. He led the league in losses once, earned runs against once, and home runs allowed three times.

      Slightly above average pitcher, multiple Cy Youngs. There are good arguments in favor of Brodeur, but the number of Vezinas he has is not one of them. Neither is resorting to childish name calling.

  23. .07 difference in save % means for every 100 shots Marty gave up 0.6 more goals. Give me a break. Quarterbacks are based on super bowls and wins, pitchers on wins and world series and basketball players on championships ( I.e why people slam lebron). Marty is one of the best of all time.

  24. Isn’t using stats post-1998 incredibly misleading?

    Patrick Roy only played until 2003. This only encompasses 6 seasons, which also happen to be 6 of Roy’s better seasons statistically.

    If you looked at the larger picture,

    Patrick Roy’s career SV% is .912
    Martin Brodeur’s career SV% is .913

    Utilizing SV% against Marty in the argument against Roy is actually a poor strategy if you wanted to take Roy’s side, and probably only works if you boil it down to picking and choosing what seasons and conditions you pick.

    If I were to try to take the Roy side in the Roy vs. Marty debate, I would probably try to zone in on the Playoff Success angle. Trying to pull any stunt with longevity and “success over time” metrics is going to favor Marty. Patrick sure looked like a much better big-game goalie, though, and that’s probably what i’d go for.

    I’ve read the Brodeur is a Fraud material a few years back, and I found it to be an interesting read. I don’t agree with it, but there’s a lot of it that isn’t wrong on the whole.

    It’s tough to argue against the fact that in point situations, Dominik Hasek was the best single-game goaltender, perhaps in history. Unfortunately, and this was touched upon in this article, dominance as a flash in the pan is something that isn’t entirely uncommon. I wouldn’t ever argue for Tim Thomas being a hall-of-famer (yet), though he has been a monstrous force the past few seasons. I would still like to see 5-6 seasons of this before dubbing it HoF material.

    If you look at the trio, the “big three” of Hasek, Roy, and Brodeur in the modern era, Roy has NEVER played a 70 game season. Hasek has played 72 games once, and then had a whole rash of injuries, issues, retirements, etc. which to me, ruin his candidacy as the greatest of all time. Martin Brodeur has done it 12 times, with a string of 10 in a row. Not just modest 70s, either. There are 78s and 77s in there.

    Consistent success is the greatest of all successes.

    If you ask Jaroslav Halak and Brian Elliot, playing 40 games in a season is sure a very nice way to make your stats look very good without actually being very good. Marty is on the very opposite side of the spectrum: his stats have almost zero shielding from being well rested.

    Anyway, if you’re curious of what my opinion is, it’s that if you could clone Hasek in his prime and stretch that over 70 games a season and playoffs, that’s the GOAT.

    Unfortunately, there was only one Hasek, and it was the Hasek we now know, so that’s why he falls 3rd in my mind, in this modern era. The 1st 2 are definitely Brodeur and Roy. Roy takes it for big-game power, Marty for consistency and influence. Both goalies had a profound impact on the game. People frequently mention the trapezoid for Marty, and I agree, but remember that Roy practically bred this generation of butterfly goalies.

    Are Marty and Patrick better than, say, Terry Sawchuk? If you time warped them back, bigger pads and all, there’s no doubt they’d be better. If you time warped Dwayne Roloson (random average goalie) back with his modern pads, he might even put up monster stats back in the day. The key is context. The shooters then could never shoot as hard as Chara and Weber do, or do any of the things a non-drunk Ovechkin could do. The game evolves, and you can’t transplant players. Greatness is completely relative. Sawchuk was dominant in his time, just like Roy and Marty are dominant in theirs. Players 20 years from now will undoubtedly be quicker, stronger, and faster than they are now. Warping Roy and Brodeur forward in time, even in their primes, would similarly produce disastrous results (for them).

    It’s no secret that every generation produces a better athlete. To use an example from a sport where physical power is more readily obvious, no professional basketball player from another era can match the physical attributes of a modern day LeBron James or Dwight Howard. This is part of the difficulty of comparing athletes from era to era.

  25. Hi there! I simply wish to give a huge thumbs up for the good info you

  26. Hello terrific blog! Does running a blog such
    as this require a great deal of work? I’ve no knowledge of computer
    programming but I was hoping to start my own
    blog soon. Anyhow, if you have any suggestions or techniques for
    new blog owners please share. I know this is off topic however
    I simply needed to ask. Appreciate it!

  27. Découvrez Hynsh Immobilier, le leader de la transaction immobilière à Luxembourg.

    Immobilier Haut de Gamme.

Leave a Reply

Your email address will not be published. Required fields are marked *