The Weakest Link

             

 

 

I’ll confess a certain level of amusement over the handwringing going on in the comments here, when it’s blatantly obvious that the biggest impediment to the Flames reaching the playoffs last season was a player that has received no mention at all. It’s a bit odd, really, that people are willing to agonize over the mediocre play of middling types like Stajan while never, ever discussing the elephant in the room. 

The primary reason the Calgary Flames watched the playoffs from the comfort of their homes was the substandard play of their number one goaltender, full stop. Miikka Kiprusoff seems to get an astonishing amount of slack from the local media and fans, but even a cursory examination of his 5v5 work last year should have given people the hint. The Flames were one of the stingiest teams in the league 5v5, allowing shots at a rate of 27.5 per 60, 4th best overall, and yet somehow finished 16th in 5v5 goals against. 

I’ve collected a wide range of data from Hockey Analysis to show just how badly served the skaters on the Flames were by their alleged elite goalie last year. First, here are the 5v5 shots for/against percentages for the regular goalies in the league when they were in net. I’ve included everyone that played at least 1500 minutes at 5v5 in 2011/12, covering 32 goalies, with the major starters and 1A types captured.

 

    Shots %     Shots %
1 ELLIS, DAN 0.539 17 BRYZGALOV, ILYA 0.497
2 CRAWFORD, COREY 0.533 18 LUNDQVIST, HENRIK 0.493
3 BRODEUR, MARTIN 0.532 19 ELLIOTT, BRIAN 0.492
4 NIEMI, ANTTI 0.529 20 RINNE, PEKKA 0.490
5 FLEURY, MARC-ANDRE 0.528 21 ROLOSON, DWAYNE 0.488
6 KIPRUSOFF, MIIKKA 0.524 22 BUDAJ, PETER 0.485
7 HOWARD, JIMMY 0.519 23 PAVELEC, ONDREJ 0.484
8 LUONGO, ROBERTO 0.514 24 VOKOUN, TOMAS 0.484
9 PRICE, CAREY 0.514 25 ANDERSON, CRAIG 0.480
10 MILLER, RYAN 0.513 26 WARD, CAM 0.472
11 HALAK, JAROSLAV 0.512 27 LEHTONEN, KARI 0.466
12 NEUVIRTH, MICHAL 0.507 28 DUBNYK, DEVAN 0.462
13 QUICK, JONATHAN 0.506 29 REIMER, JAMES 0.460
14 BOBROVSKY, SERGEI 0.504 30 KHABIBULIN, NIKOLAI 0.459
15 THOMAS, TIM 0.500 31 HILLER, JONAS 0.452
16 MASON, STEVE 0.498 32 BACKSTROM, NIKLAS 0.433
 
 
When Kiprusoff was in net, the Flames were outshooting their opposition by a fairly healthy rate, which runs contrary to the myth of a goalie battling insumountable odds. In other words, Kipper wasn’t exactly bombarded compared to his brethren. With that sort of shots ratio, the club should have been in pretty good shape 5v5.
 
So, what went wrong? The answer to that might be found in our man’s 5v5 save percentage:
 
  

    5v5 SV%     5v5 SV%
1 THOMAS, TIM .949 17 MILLER, RYAN .926
2 RINNE, PEKKA .935 18 QUICK, JONATHAN .925
3 LUONGO, ROBERTO .934 19 ROLOSON, DWAYNE .924
4 REIMER, JAMES .933 20 ANDERSON, CRAIG .921
5 BRYZGALOV, ILYA .933 21 DUBNYK, DEVAN .921
6 NIEMI, ANTTI .931 22 VOKOUN, TOMAS .920
7 BACKSTROM, NIKLAS .931 23 NEUVIRTH, MICHAL .919
8 PRICE, CAREY .930 24 HALAK, JAROSLAV .918
9 HILLER, JONAS .930 25 KIPRUSOFF, MIIKKA .917
10 WARD, CAM .929 26 HOWARD, JIMMY .915
11 LUNDQVIST, HENRIK .929 27 BUDAJ, PETER .914
12 LEHTONEN, KARI .929 28 MASON, STEVE .912
13 PAVELEC, ONDREJ .928 29 BRODEUR, MARTIN .911
14 CRAWFORD, COREY .928 30 ELLIS, DAN .907
15 BOBROVSKY, SERGEI .927 31 KHABIBULIN, NIKOLAI .901
16 FLEURY, MARC-ANDRE .927 32 ELLIOTT, BRIAN .900
 

That really isn’t up to snuff for a goalie that’s carrying the 5th highest cap hit amongst goalies, at least in my view. As an aside, a drop of one percent is likely worth about 12-15 goals against 5v5 for a regular goalie, since most of the main guys face between 1200-1500 5v5 shots.

Now, one of the common refrains is that the Flames ran up shot totals when they were chasing the game. With that noted, here are the shots for/against % when the game was tied for the same approximate cohort of goalies. For these next two tables, I’m using the filter of 750 5v5 tied minutes played during 2011/12 on Hockey Analysis, which gives me 27 goalies in total:

 

    Shots %     Shots %
1 HOWARD, JIMMY 0.547 15 ELLIOTT, BRIAN 0.503
2 BRODEUR, MARTIN 0.544 16 QUICK, JONATHAN 0.501
3 CRAWFORD, COREY 0.544 17 BRYZGALOV, ILYA 0.499
4 BOBROVSKY, SERGEI 0.537 18 PAVELEC, ONDREJ 0.492
5 FLEURY, MARC-ANDRE 0.532 19 VOKOUN, TOMAS 0.491
6 KIPRUSOFF, MIIKKA 0.530 20 LUNDQVIST, HENRIK 0.490
7 NIEMI, ANTTI 0.529 21 ROLOSON, DWAYNE 0.489
8 THOMAS, TIM 0.526 22 NEUVIRTH, MICHAL 0.478
9 RINNE, PEKKA 0.525 23 ANDERSON, CRAIG 0.465
10 LUONGO, ROBERTO 0.518 24 HILLER, JONAS 0.462
11 MILLER, RYAN 0.513 25 WARD, CAM 0.460
12 MASON, STEVE 0.506 26 LEHTONEN, KARI 0.454
13 PRICE, CAREY 0.506 27 BACKSTROM, NIKLAS 0.424
14 HALAK, JAROSLAV 0.505      

 

Again, Miikka wasn’t all that hard done by. The Flames actually controlled more of the shots when the game was tied than overall, and Kipper had the 6th best support of his peer group. Honestly, if your team is sporting a 53% shots for/against ratio when the game is tied, average goaltending should do the trick unless you have a season like New Jersey where nothing was going in the other team’s net for the first two months. At any rate, here are the SV% numbers in that game state for the same crew of goalies:

 

    5v5 tiedSV %     5v5 tied SV %
1 NIEMI, ANTTI .944 15 QUICK, JONATHAN .930
2 RINNE, PEKKA .944 16 BACKSTROM, NIKLAS .927
3 WARD, CAM .940 17 MILLER, RYAN .926
4 HILLER, JONAS .940 18 FLEURY, MARC-ANDRE .924
5 THOMAS, TIM .939 19 BRODEUR, MARTIN .923
6 ROLOSON, DWAYNE .938 20 HOWARD, JIMMY .923
7 ANDERSON, CRAIG .936 21 LEHTONEN, KARI .923
8 PAVELEC, ONDREJ .936 22 PRICE, CAREY .921
9 HALAK, JAROSLAV .935 23 VOKOUN, TOMAS .921
10 BRYZGALOV, ILYA .935 24 MASON, STEVE .917
11 CRAWFORD, COREY .935 25 NEUVIRTH, MICHAL .911
12 BOBROVSKY, SERGEI .934 26 KIPRUSOFF, MIIKKA .910
13 LUONGO, ROBERTO .933 27 ELLIOTT, BRIAN .906
14 LUNDQVIST, HENRIK .930      

 

Oy. When the game was tied, as his team helped him more, he gave them less. The Flames were a middle of the pack SH% team at 5v5 overall and tied, by the way. Kipper had the 14th best SH% support overall and 13th best when the game was tied, so I have a hard time accepting any sort of argument that his team shot nothing but muffins while he faced unstoppable bullets. 

What should worry people the most is that Miikka Kiprusoff has been underwhelming for several years running by that last metric. I ran a four year composite of SV% for goalies that had played at least 3000 EV tied minutes since 2007/08. There are 19 goalies that met that threshold: 

 

    07/08 to 10/11 5v5 tied SV%  
1 RINNE, PEKKA                              .941  
2 THOMAS, TIM                              .937  
3 BRODEUR, MARTIN                              .935  
4 WARD, CAM                              .934  
5 LUONGO, ROBERTO                              .933  
6 QUICK, JONATHAN                              .931  
7 VOKOUN, TOMAS                              .930  
8 ROLOSON, DWAYNE                              .929  
9 BACKSTROM, NIKLAS                              .929  
10 MASON, CHRIS                              .928  
11 NABOKOV, EVGENI                              .928  
12 BRYZGALOV, ILYA                              .928  
13 MILLER, RYAN                              .926  
14 FLEURY, MARC-ANDRE                              .925  
15 LUNDQVIST, HENRIK                              .925  
16 MASON, STEVE                              .924  
17 KIPRUSOFF, MIIKKA                              .919  
18 PRICE, CAREY                              .919  
19 TURCO, MARTY                              .913  

 

Please note that included in that run was a .935 number in 09/10, which was 6th best amongst his peer group. As well, Kipper had the 5th best shots for/against ratio of those 19 goalies working in his favour, as well as the 5th highest team SH% in front of him during that 4 year run, so again, the team was giving him a reasonable amount of support in terms of shots and shooting percentage. 
 
There’s no nice way to put this, really. Miikka Kiprusoff has been at the heart of the failures of the Flames since 06/07. I’ll be the first to acknowledge that he had a very good 09/10 in the midst of this bad run, but that season is looking more and more as if it were an outlier rather than a bounce back year after being freed of the Iron One.
 
I have no problem in presuming Mr. Azevedo’s reportage from rookie camp is on the level, which suggests that Jay Feaster might not have been completely sincere in his past professions of belief that his number one stopper was of elite status. That’s the correct approach in my view, because no matter what people want to believe or what his cheerleaders might care to suggest, Miikka Kiprusoff hasn’t been within hailing distance of elite during the last several seasons.
 
That leaves the club with a fairly intractable problem for this year, since they weren’t able to move Kipper and they haven’t yet begun the dismantling of the club that next summer might well bring. Absent him pulling out a season from the depths of time, his recent work strongly hints that Miikka Kiprusoff will hinder whatever good the Flames’ skaters are capable of achieving in 2011/12. 

 

 
  • Greg

    Couple of questions:
    1) How much of his stats can be attributed to the horrific slump he hit in January last year? Ie, if you took that bad month or so out when even the fans and media were all over him, do his stats go back to middle of the pack?
    2) are their any advanced stats that would help assess shot quality and could possibly be used to come up with a relative-sv% among goalies?

    • SmellOfVictory

      Re: shot quality: The general rule of thumb is that shot quality does not differ significantly year-to-year or team-to-team in terms of what proportion of shots taken or given up are ‘good’ shots. If a team is bad, they tend to give up proportionately more shots overall rather than simply a greater proportion of high quality shots.

      • Sworkhard

        If it’s true that a bad team will give up proportionally more bad shots, why do we care about scoring chances. We’ve seen it a lot that one team would outshoot the other significantly but the outshooting team would have fewer or similar scoring chances. To me sv% as a raw number doesn’t mean much, as there is a correlation between sv% and shots against, but isn’t really one for shots against and goals against. I’m sure if someone did the work, they would find that there is a lot better correlation between scoring chances and GA than SA and GA.

        • SmellOfVictory

          My understanding is that fenwick/corsi are strongly correlated with scoring chances, and since both are just indicators of shots taken, we can logically conclude that sv% (based on shots) is nearly as accurate as a true scoring chance/sv% measure.

          Scoring chances are certainly better, but not by a wide enough margin to say “Kiprusoff was a good goaltender last season and that fact was just hidden due to the ineptitude of the team in front of him.”

          That’s my understanding, anyway. Scoring chances may differ heavily from the shot count on a game-to-game basis, or even for streaks, but once you have a large enough sample size, like most things, they will even out. 82 games, or 2000+ shots per side, should be a large enough sample size for that to occur. (I’m admittedly half-assing my background info here, so I hope if I’m going the wrong direction with this, someone more dedicated/knowledgeable will correct my post)

          • BobB

            Even if this is true, there has to be a distinction amongst teams, and I’ve yet to see it enter that equation. That Puck Prospecus article I linked ^^^^ is quite elegant. It makes sense why Roberto Luongo could go from a awful FLA team to an increasingly excellent VAN team and see his sv% on average drop. Or how Minnesota would always be a counter attack team and say “we make our shots count” (as if they are trying harder?!?!)

            Puck pursuit on the back-check (something CGY is horrendous at) Young… Fast… Tenacious.

            Hypothetically. If one team plays five forwards and every other SA is a breakaway yet they get 40 SF a night, your goalie is going to have about a .788sv% and your team will have a impressive FOR/AGAINST %. (an extreme example that doesn’t exist)

            Higher shot ratio the better at any cost, and if the goalie lags behind, he’s the weak link.

            And defense does…? back-check pressure is important…?

            Again, (sorry for repitition) if your primary comparison is vs all CGY tenders you start getting a truer idea of the “level” CGY goalies are facing. Do they ALL suck?

            Kipper didn’t have a great year but he doesn’t play in a vacuum.

            If we start seeing Kipper play less than we expect (like a Huet), vs his annual 10-15 games more than we initially plan (because we can’t lose) then the numbers will be backed up. Until then, I remain skeptical his decline is SOLELY him and that his personal decline is as vast as numbers may suggest.

    • ChinookArchYYC

      This is pretty much what I was thinking. I’d really like to see his performance stats form January to the end of the year. I didn’t see much of a differnce in his play, but the rest of the team sure raised their level of compete and confidence. I can only count 3 players that played consistently throughout 2010/11 on the Flames: Tanguay, Jackman and Kiprusoff.

  • Greg

    Trade deadline people is better than Christmas. Tis the season when GM’s go mad & overpay. It’s when buddy Jay needs to blow the SOB up if we are bottom 3rd in the league. Iggy, Kipper, Sarich, Langkow, Joker, Bourque. No one is untouchable. You may want to grab your private parts but embrace the fact that Calgary has nothing to attract prime free agents, Edmonton is starting to have more attraction than Calgary because they now have pieces that can go the way of Chicago or Pitt. The quicker we can clear out the geriatric ward the quicker fans will become excited about the future of this hockey club. With the exception of Stajan, the day of salary dumps should be over for this franchise after 2011-12 season.

    • SmellOfVictory

      Salary dumps will be entirely unnecessary for next offseason since most of the Flames’ players are UFA. Any trade made at the deadline should be for assets, and only assets.

  • otto

    Shots on goal don’t tell the whole story.In a game against Van last year(Jan 5)we lost 3-1 but out shot them 44-21 and Kent had the scoring chances at 13-10.Luongo padded his stats with 31 easy saves while 1/2 the shots Kipper faced were quality chances.

    • BobB

      +1….The problem with looking at sv% and comparing evenly across teams is assuming that all teams are equal environments – an unsafe assumption.

      Boston. They allowed 32.7SA/game last year 29th in the league! Are they the 29st best defensive team? They’re worse then the Islanders? No way. I’d assert that not all teams are equal, but how much difference exists: I don’t know!

      Comparing sv% across teams should be secondary information to comparing within the team. In fact, against career should preface against other teams.

      BOS evsv% – Thomas .947, Rask .925,
      VAN evsv% – Luongo .934, Schneider .933

      Why are they so high? Cause they’re good teams! Team sv% vs goalie have any value?

      Logically, this makes sense. Kipper always plays 10-15 more games than is the goal at the beginning of the season. If he was the weakest link…. then are the coaches insane?

      This isn’t Turco getting usurped by Crawford,
      This isn’t Huet getting usurped by Niemi, This isn’t Goalie B out playing Kipper. Instead of Kipper preventing us from being in the playoffs, what if Kipper is preventing us from being worse?

      Comparing Miikka to any goalie on Boston (or other quality teams) doesn’t really tell us anything except Boston is waaaay better than Calgary even without Thomas in net.

      Ok, I already know that.

  • SmellOfVictory

    Good information and all, however does it have indication of the quality of the shots/chances involved?

    Who’s not to say the Flames are good at blocking/limiting shots, but prone to second chance goals that should have been cleared away, or defensive breakdowns that lead to a 1v1, or odd man rush?

    Also… the flames by the eye seem, to play like a team on the perimeter as opposed to by the hash marks, or in the circles. A lot of the games last year, the flames did not ‘feel’ dangerous on the attack, even when getting control and shots on net.

    Even though I know the information provided is accurate, I don’t think it paints the whole story. I do think Kipper has declined, but I definitely don’t think he is the reason for the flames downfall.

  • Matty Franchise Jr

    Is there such a thing as Scoring Chance Save %, as opposed to the Shot on Goal Save % that we’re all used to?

    I think that might clear things up a little, if we knew what % of scoring chances were being stopped.

    • BobB

      That’s an interesting idea but I would be wary of results. Even if they were my results! I started doing something like this… but…

      I personally think there is enough scorer bias subjectivity in just shots to question things… scoring chances would be a nightmare, (like goal-line debates). Also sample size info? You’d have to start looking at building a database of years to compare to… and for all goalies.

      I quickly gave up thinking maybe someone had a better way, because If I saw .800% of legit scoring chances at even, tied, in 71 games and .805 in 17 … and that’s all I had?

      If someone has it, or better ideas, it’d be interesting to look at, but …

  • HockeyFan

    I do not quite agree with this article as it is only based on numbers. In practice, it is quite a different story.
    I agree the Flames out shoot their opponents most of the times, but the truth is, the shots are mostly coming from the periphery. They are not quality shots. The goal tenders against the Flames have a clear view of the shots. No Flames in the slots to deflect shots or block the goaltender’s view.
    On the other side of the rink, in the Flames D zone, there is also another story that the numbers in this article does not mention. It is the Flames breakdown. The attention of details, like covering the point as an example. How many times did I see an opponent d-men pinch in without being covered and score a clinching goals. You see the Flames wingers just stand there and just look stunned. They have not look over their shoulder to see that the d man was coming.
    All this to say that Kiprusoff may have played poorly at times, but overall, he is a very consistent goalie, which is a pretty precious thing to have in this league.

  • HockeyFan

    The team has played like hot garbage in front of him for the past 3-4 years. Suffice it to say, if he were playing behind the Red Wings, Flyers or Sharks he would have been perennially championship bound.

    Regarless of the context, I blame Phaneuf.

  • HockeyFan

    I see where your coming from, but it seems too much like your desperately looking for someone to blame for the Flames woes. So you turn on one of the two players who give the Flames a chance night in and night out? I disagree whole heartedly. This isnt baseball where stats (ala Moneyball) can win you championships in my mind. The game is more about flow and as much as I love stats, they only get you so far. Remember a couple years ago when the Flames were looked at to win the cup because on paper they looked great. Well, we all know what happened there. The issues with the team are far greater than Miikka having a couple off nights.

    However, I do respect the time and research you took to try and prove your point. I just dont agree.

  • BobB

    @ Many

    I think we can challenge the variables which contribute to a stat, and I think we can challenge how much “noise” is in stats, how much team effect…

    but if we talk about whether Kipper displays emotion, plays with heart, or is a good guy, is that based on anything, or going to bring us anywhere?

    If we’re going to use stats to praise Miikka, we should be prepared for others using them to take him down. Confirmation bias on both sides…. it’s a b**ch.

    I personally believe that advanced stats much more clearly illustrate the offensive game and thus tend to get people focusing on that, as a result, giving offense artificially increased importance. When we want to focus on defense we go straight to the goalie, because their numbers are so clear and measureable and somewhere in their is missing an entire analysis of the defensive game (because it’s just THAT complex). I think far to often, we use advanced stats to discover who are the “good guys” and who are the “bad guys” instead of focusing on how to use those guys to achieve a greater understanding of the game.

    … BUT, talking about Kipper’s “try” sure as heck isn’t going to help us get to understanding anything, let alone more about the defensive game in hockey.

    Thanks Robert…. even if I disagree.

  • HockeyFan

    Team play has way more impact then you can statistically quantify. It is why sub average tenders can put up good numbers on great teams.

    How many rebounds are being cleared. Can they minimize net presence. How long does the other team get in the ozone.

    Kipper didn’t have a great season by any metric. But to call him the weakest link is a big stretch.

    • BobB

      …and back pressure on the back-check. Which, I think this is such a key.

      It makes:
      -possible for defenders to play more aggressively,,
      -clearing of rebounds quicker,,
      -forcing of N-S hockey, vs East-West opening of ice,,
      -set-up more difficult from reduced time,,
      -collapsing and protecting easier with numbers,,
      -turnovers with man advantages on counter,,
      -keeping shots to the perimeter.

      It comes from:
      -fitness and speed,,
      youth (who normally possess more endurance),,
      -coaching.

      Nearly none of those things sound like Calgary in the defensive zone and all of those things in the offensive zone.

      Maybe Sutter tries, but he can only play with what he is dealt.

      I can tell you Calgary was 6th worst in give-aways (805) and 27th in the league in Take-aways VS Give-aways @ -227, with other such successful teams as Toronto (28th) and Edmonton (29th)…. although LA was 30th???

  • RKD

    2010-11 Calgary NHL 71 37 24 6 4156 182 1935 2.63 .906 6 2 – 2 – –
    2009-10 Calgary NHL 73 35 28 10 4235 163 2035 2.31 .920 4 – – 1 – –
    – – –
    2008-09 Calgary NHL 76 45 24 5 4418 209 2155 2.84 .903 4 2 – – – –
    2007-08 Calgary NHL 76 39 26 10 4398 197 2096 2.69 .906 2 – – – – –
    2006-07 Calgary NHL 74 40 24 9 4419 181 2190 2.46 .917 7 – – – – –
    2005-06 Calgary NHL 74 42 20 11 4380 151 1951 2.07 .923 10 – – – – –
    – – –
    2003-04 Calgary NHL 38 24 10 – 2301 65 966 1.69 .933 4

    He’s been facing 2000 shots a season, his save % has been in the low .900s. The only difference I’ve really noticed is that he’s been beat by goals in the past that would not have beaten him before. In other words, he’s giving up more weak goals than usual. He’s still a strong goalie in my eyes. On one side you can see a guy like Brodeur declining, but we just saw a 37 year old Thomas hoist the Cup.