2012-13 Reasonable Expectations: Miikka Kiprusoff



What should we consider Miikka Kiprusoff to be?

An elite goaltender that seems to be made of Jell-O, keeps his team in games and makes stops at key times?

Or is he a league average goalie who just also happens to be so hyper-athletic people look past his (at times) poor play?

Being the brother of a goaltender (and somewhat of a goaltender myself these days), it would be quite easy for me to be biased in my view of Kiprusoff’s performance – but my own philosophy regarding goaltenders usually boils down to “goaltenders can only lose games, not win them.” That’s extremely reductive, of course, but over the past few years both the performance of Kiprusoff and likewise the performance of other goalies around the league have given me ample reason to stake such a claim.

Perhaps we should rewind a little bit before we go down that path, however.

Kiprusoff had an amazing season last year. He was a top-10 guy in EVSV% (.928), GP (70) and saves (1878) as well as being pretty much the sole reason the Flames were even close to the playoffs in March and April, putting up an insane .938 SV% between the Boston Massacre and the trade deadline.

But – that didn’t continue post-deadline, as Kiprusoff’s play would drop off to the tune of a .919 SV% the last part of the season – with the majority of that poor play coming at even strength. One could blame the high dependence on Kiprusoff and resulting wear and tear on his body for his decline in play, but I’m not convinced – professional athletes are paid to be in shape. No, I think it goes much deeper than that, as most things in this sport do.

Everything Old Is New Again

Some of you might remember this article I penned last year over at Matchsticks and Gasoline. It boiled down to the appearance of these simple numbers:

2005/06 941
2006/07 932
2007/08 919
2008/09 907
2009/10 928
2010/11 916
2011/12 928

More than these ones:

2005/06 941
2006/07 932
2007/08 919
2008/09 907
2009/10 928
2010/11 916
2011/12 928

So, in looking at those, I’ve identified three main questions.

  1. What is the likelihood Kiprusoff performs on the same general level (~.920 SV%, ~.925 EVSV%) this season as he did last season?
  2. What is the likelihood Kiprusoff’s performance drops off below league average (~.915 SV%, ~.920 EVSV%)?
  3. Why will one happen and not the other?

Real-Life Math Sucks

In our business (well, any business), all you really have to help predict the future is the past. Looking at Kiprusoff in the 7 years since the last lockout, we can reduce his play to 4 different and simple performance levels. Twice since 05/06 he’s been what I would consider elite (05/06, 06/07), twice he’s been good (09/10, 11/12), twice he’s been average (07/08, 10/11) and one year he’s been replacement level (08/09).

The issue here is that without the two seasons he’s been elite, you have 2 good seasons and 3 bad ones. Now, it seems mighty convenient for me to just drop those two seasons from the history books, so I won’t – but I will say that 7 years seems like an awfully long time for one to be able to perform at an elite level. The reason for that, of course, is the natural player development curve we see from everyone who has ever put skates on.

Kiprusoff is no spring chicken, and goaltending is a position that is simultaneously undervalued and overvalued – those two things really skew the public perception of what “elite” actually is. But rather than argue semantics – that’s another article entirely – let’s get down to the nitty gritty and see what kind of season should be expected. I would like to draw your attention to these two links.

The first is a list of goaltenders between the ages of 34 and 36, and the second is a list of goaltenders aged 35 and up. Based on what’s highlighted there you’ve probably already figured out what the two links are talking about – .920 seasons put up by goalies aged 34 and above in the history of the NHL. Not very many.

I think what’s more important to us, though, is the prevalence of good seasons after the lockout that goalies have had. 7 out of those 15 or so seasons have happened in the current era, which is definitely a positive trend. But the issue here is that the guys doing this sort of thing aren’t regular goalies. Three of them could conceivably be called at the very least generational goaltenders – Hasek and Roy are two of the best (if not the best) to ever play goal and Belfour was a multiple Vezina and Jennings winner. You’re then left with 22 games of Sean Burke and Mr. Insane himself, Tim Thomas.

While Kiprusoff has been good, I think it’s an extreme stretch to put him in the same sentence as Hasek, Roy and Belfour (even though that’s what I just did). Burke’s season is obviously an outlier, being that he only played 22 games primarily as a back-up. Thomas’ past four seasons have had associated EVSV% of .927, .947, .913 and .940. Thomas had hip surgery after his .913 season, so it’s logical to conclude that he was suffering at least in part from his hip injury during the season. That’s significantly better than Kiprusoff has been over the past four years, even though Thomas is older. 

So, from this historical perspective, there doesn’t seem to be much that would tell us Kiprusoff will repeat his performance from last year.

Possible Impossibilities

So, now that we know history likely won’t be kind to Kiprusoff, we need to look at what’s changed since last season.

Obviously, the Flames defense as a whole has improved a little – but does that mean that Kiprusoff will do better? If you think about how goals are random events, you’d probably have a hard time buying that the Flames will be able to suppress shots (if such a thing is even measurable and/or significant) to such a degree that Kiprusoff’s numbers dramatically improve.

With the kinds of numbers Kiprusoff had last year, stopping one extra shot per game from being created in front of Kiprusoff would result in about 1 win’s worth of goals – but that’s essentially 4% less shots year-over-year. It’s not unheard of, but over the course of a season that’s probably about 3 extra games worth of shots. It’s a lot in one area and not exactly a lot in another, which is bad news. One win better wouldn’t have changed their draft position.

Of course, the argument could probably be made the other way with everyone on the team being older and Dennis Wideman’s lack of defensive prowess being added to the mix. While I don’t think the dropoff from Dennis Wideman to Scott Hannan is small, I’m unconvinced the gulf is so large the Flames will be able to see demonstrable results on the defensive side of the puck.

So the defense is a little better, but perhaps not in the way we’d like it to be. While most of the game is played at even strength (84% of shots Kiprusoff faced last year came with 11 other guys on the ice), penalty killing and power plays can make a huge difference in a goalie’s game. Whereas EVSV% is basically a goaltender-only stat, PKSV% has a lot to do with the team’s penalty killing strategy. If a team has a tendency play-wise to devolve into a mess of penalty killers, it’s going to cause more goals than a team with a good penalty killing strategy. I don’t think I’m re-inventing the wheel on that one. 

But Calgary was 9th in the league last year in penalty killing, and to my eye there hasn’t been much change in terms of PK personnel over the off-season. I think another 12-7 finish is possible, but once again, the difference between the Flames and the 7th place team (St. Louis) was the equivalent of 3 goals. Once again, a lot in one area for marginal return.

So What?

Kiprusoff is old as far as hockey players go. He’ll be 36 by the time the season eventually starts, and most goalies fall off their respective cliffs well before that age. But Kiprusoff hasn’t really followed the pattern of straight decline we usually expect from goaltenders – or even skaters, for that matter. Maybe that’s due to the fact that most players don’t get second chances to redeem poor play, they’re just out of the league – but maybe there is something different about Kiprusoff. I just can’t put my finger on what, if anything.

In saying that, though, I’d like to propose something: that Kiprusoff has never truly been an elite goaltender – merely a good one with a couple of elite seasons and a bunch of hype due to a number of factors. Cue the comment from Rain Dogs in 3, 2, 1…

For me, it really does come down to expectations. If a player hits his targets, that’s great. If you consider Player X to be good, and he performs “good”, then you don’t care. What you do care about is when a player performs above or below his expectations. Of course, there are different reasons for both. It’s curious to me that expectations haven’t been met by Kiprusoff a couple of times but there hasn’t been any backlash because of it. That tells me one of two things: either people aren’t really evaluating Kiprusoff based on the amount of pucks he’s stopping (which is odd, since that’s his job) or everything else in Calgary has been so bad over the past few years people just haven’t really noticed his play (quite likely). 

So let’s go back to the questions.

  1. What is the likelihood Kiprusoff performs on the same general level (~.920 SV%, ~.925 EVSV%) this season as he did last season? Probably pretty good. He keeps his body in shape and situationally not much has changed. But – goaltending is extremely variable year over year, and he’s had issues in the crease in past seasons.
  2. What is the likelihood Kiprusoff’s performance drops off below league average (~.915 SV%, ~.920 EVSV%)? Probably pretty unlikely, but he has had poor seasons before and is at an age where there isn’t much in terms of historical precedent leading to quality seasons.
  3. Why will one happen and not the other? One is likely in in a historical context, and the other is likely in a personal context. More often than not in this sport, those two tend to find a middle ground.

As a side note, every time I do one of these things that involves me looking at the 08-09 season, I get a little mad. That’s probably the 2nd best Flames team ever constructed and they lost out in the playoffs due to some untimely injuries as well as the worst season their goaltender has had to date. One always tends to wonder a bit.

I think Kiprusoff finishes with 60 GP (pro-rated) and a .918 SV%, marginally worse than what he put up last year. I doubt that’s enough to get the team to 40 wins (pro-rated) and I wouldn’t be surprised if they finished bottom-10 in the league. There will be more factors then average goaltending contributing to it, sure, but average goaltending is something the Flames continue to get and something they seem to be okay with. I’d be more of a proponent of trading Kiprusoff if I didn’t feel the time had passed, but there you have it.

What do you think?

  • 2nd best Flames team ever constructed

    I’ll quibble here and say a number of the late-80’s, early ’90’s teams were better. Probably the best collection of skaters post-2004 though.

    I think Kiprusoff finishes with 60 GP (pro-rated) and a .918 EVSV% for the Flames though.

    Did you mean .918 overall SV%? Because .928 to .918 ES SV% is a big drop (although I consider that wholly possible as well).

    As for 60-games played…You have ot be assuming Leland Irving takes a big step forward to spell off Kipper that much.

    • Derzie

      whoops, yeah. I’ll fix that.

      I don’t necessarily think that irving will take a huge jump forward, I just think that the flames might bow to pressure and start kiprusoff less.

      @bol/mike: basically what sov says. I’m convinced most starters see enough pucks over the course of a season that shot quality is mostly irrelevant.

      • SmellOfVictory

        Irving may not take a big jump forward ,but management has got to give him a chance to prove himself once and for all.That will mean 20 games.Not sure if that,s bowing to pressure or simply finding out where your prospect,s are in regard to readiness.As far as Kipper,s success i would have to agree that his stat,s are,nt spectacular,but he does come up big when he needs to.That,s what makes him a fabulous teamate

  • I find the idea of EVSV% to be somewhat inadequate as a stat. You have argued that it is all a goaltender impacted statistic. Until shot quality can be factored in I find it unreliable at best to solely judge a netminder on. All defenses and systems are not created equal, there is no current way to “balance out the playing field” between goalies. This is one of the few occasions the ” eye test” may be the best way to analyze a goaltender. To say a goalie cannot win a game for a team did not watch enough jonathon quick last year. I’m not really upset with your argument or results, I just hate seeing EVSV% used as an end all stat.

  • I reject the notion that goalies don’t win games, only lose them. While I do agree that a goalie doesn’t win you games, he’s very much responsible for PRESERVING games.

    Kipper has preserved his fair share of games over the years, but it’s hard to believe he’ll play at the pace he did last season ever again.

    • SmellOfVictory

      That’s what he’s saying, though. A goaltender is essentially incapable of scoring goals (I don’t believe a goal has ever been scored on one tender by another – only empty netters), and thus can only delay loss (what you term “preserving” a game).

      @Mike Arcuri: The problem with the eye test is that it’s incredibly fallible. Way, way more fallible than statistical analysis, and EVSV% is the best stat available.

  • delaying loss still implies losing. I think while Azevedo is right that a goalie can lose a game, he can prevent you from losing it too.

    I think we all might be saying different versions of the same thing, but I felt it needed clarifying.

  • SmellOfVictory

    I agree the eye test is fallible. Too many homers and biases, but to say shot quality is negated by the total number of shots is false. A goalie playing in Boston’s system will have greater numbers than one playing in columbus’s. If a goalie is facing 2 on 1s and shots from the slot because jack Johnson is defending infront of them rather than Zdeno chara their save % will be lower. An objective scout I would trust much more than a number with a limited scope.

  • JayP

    The problem is that for goalies (and skaters too) the “eye test” is vastly biased towards exciting and athletic players. There’s a reason guys like Kiprusoff and M-A Fleury are still considered by some to be top end goalies when their numbers would indicate otherwise.

    Athletic goalies are able to make the amazing save a positional goalie simply can’t, but lets in average goals at a higher frequency. Which play do you think the eye test is going to remember?

    They also make routine saves look great due to a lack of fundamentals and positioning (Derek Jeter syndrome).

    • JayP

      We know Derek Jeter is a poor defensive shortstop because of advanced baseball statistics. Zones are divided up on a field and shortstops range is determined by how he gets to a ball others(Jeter) just cant get too. It actually measures the difficulty of a play unlike EVSV%. Without measuring the degree of difficulty, EVSV% is essentially fielding percentage. It really doesn’t tell the entire story (and once again Jeter wins a gold glove)

  • BobB

    “Cue the comment from Rain Dogs in 3, 2, 1…”

    Does it make this otherwise fairly strong article better to feature me in it in this way? Calling me out?

    I mean … I’m flattered that you’ve thought of me all summer. But wouldn’t it be stronger to not involve me personally and instead stick to refuting points/stats?

    I try to add value to the conversation. Add critical thinking. I’m always surprised by the negative reaction that gets.

    Try imagining we’re having beers, debating and having a good time.

    Unfortunate. No? Oh well. Maybe there’ll be no hockey this year and I won’t have to update the evsv% spreadsheet.

    • Derzie

      joke (noun): something said or done to provoke laughter or cause amusement, as a witticism, a short and amusing anecdote, or a prankish act: he tells very funny jokes.

      that comment is exactly the reason why I put it in there. chill, dude.

      • BobB

        Right. Mine was a joke too! (Worse than you making the comment is not standing up to it.)

        It’s easy to say, “it was a joke”. Also, easy to say “yeah, I knew you would respond that way and that’s why I made the joke.” But why does the article need it? Some people may say that writing something specifically at a commenter to create a response from said commenter is “trolling”. Right?

        “A troll is someone who posts extraneous messages with the primary intent of provoking readers into an emotional response.”

        My point is, as your writing evolves, have confidence in your analysis and your writing. You put it in there because you think that “joke” makes your article better. You wouldn’t put it in because it makes it worse. Your article would be better if you said “here’s why:” and not…. “Cue the response from….”

        I don’t read professional writers who make “jokes” about intelligent contribution to the community in other serious venues. You may respond to me: “This isn’t a serious, nor professional venue” …..but…

        I question whether I’m the one who needs to chill when I haven’t commented on here in months…. maybe 5 months, and it’s still en vogue to spar with me/troll?

        and I should chill? Write. and Analyse. You wanted comments and feedback from me. You got it. Worth it? I doubt it.


  • RKD

    Right now Kipper is still better than any other alternative. Rammo is still in the KHL for another season and Irving is still progressing.

    A couple of elite seasons is one way to look at Kipper. He’s the only goalie since Mike Vernon to lead the Flames to a Stanley Cup final. He has 7 consecutive seasons of 35+ wins in a season. He ranks 23rd all time for goalie wins at 311. In fact, he reached 311 wins in 7 seasons, when it took Hasek 16 seasons to reach 389.

    Brodeur and Roy never really had spectacularly high save percentages. Brodeur’s highest was .927 and Roy’s highest .925.

    Had Kipper been a starter earlier on, he would be well past 400 wins. He might have even had more Vezinas.