Why Giving Pekka Rinne A Huge Contract Was A Bad Idea

The Nashville Predators made waves on Thursday by signing Pekka Rinne to a seven year, $49-million contract. It is the largest contract in the history of one of the NHL’s smallest markets, and it instantly establishes Rinne as the cornerstone for the team for most of the next decade.

Insofar as the deal shows that the Predators’ ownership is serious about building on last season’s successes, this contract does a good job. It shows the team is willing to expend serious money to keep players that it values. Given that the club was facing the possibility of losing a trio of cornerstone players (Rinne, Shea Weber and Ryan Suter) this is a message that will undoubtedly be received appreciatively by the team’s fans.

Unfortunately, the deal also sends two other messages. First, it shows that one of the league’s most financially cautious clubs hasn’t learned the lessons of the last few seasons, where goaltending performance has fluctuated wildly and competent goaltenders can be had for close to nothing on the free agent market. It also shows a fundamental misunderstanding of how goaltending save percentage should be considered. We’ll start with the second point.

Goaltenders make the bulk of their saves in one of two situations: at even-strength, or while their team is on the penalty kill. Even-strength save percentage can vary significantly, but it is shorthanded save percentage that truly jumps around – a goaltender can go from posting an elite save percentage one year to a terrible one the next. This makes sense given that team performance on the penalty kill tends to vary widely depending on both coaching and personnel; the loss of two players can turn a club’s penalty-killing upside down. It also makes sense when we consider the small number of shots involved – a goalie making over 1000 saves at even-strength might make less than 200 on the penalty kill.

Pekka Rinne’s career penalty-killing save percentage numbers are as follows:

  • 2008-09: 163 saves on 188 shots – 0.867 SV%
  • 2009-10: 193 saves on 231 shots – 0.835 SV%
  • 2010-11: 270 saves on 296 shots – 0.912 SV%
  • Career: 637 saves on 726 shots – 0.877 SV%

Last season’s number on the penalty kill is significantly better than in Rinne’s previous two years, and well above the league average.

History shows us that Rinne isn’t the first starting goaltender to post a shorthanded save percentage above 0.900. Four players managed it in 2009-10, three in 2008-09, three in 2007-08 and five in 2006-07. Let’s look at the list, and see how those players did in the preceding and following seasons.

Player 2006-07 2007-08 2008-09 2009-10 2010-11
Brian Elliott 0.812 0.904 0.858
Chris Mason 0.921 0.854 0.885 0.899 0.820
Cristobal Huet 0.907 0.870 0.883 0.886
Dan Ellis 0.918 0.831 0.843 0.856
Henrik Lundqvist 0.878 0.871 0.913 0.883 0.873
Jimmy Howard 0.905 0.875
Martin Brodeur 0.904 0.893 0.848 0.866 0.862
Mathieu Garon 0.859 0.908 0.821 0.891 0.855
Nicklas Backstrom 0.913 0.905 0.918 0.880 0.859
Roberto Luongo 0.906 0.872 0.858 0.860 0.897
Ryan Miller 0.846 0.866 0.879 0.919 0.886
Tim Thomas 0.846 0.871 0.905 0.919 0.889

With two exceptions, every player on that list was one-and-done when it came to posting a 0.900 SV% or better while their team was shorthanded. This holds true not just for lower-tier starters, but also for the high-end guys. Ryan Miller won a Vezina on the back of his penalty-killing save percentage in 2009-10, but he regressed to a number more in line with his career totals in 2010-11 – a regression that was entirely predictable. It was a similar story for some of the best goaltenders in the game – guys like Brodeur, Lundqvist, and Luongo – they all had a single season to shine before crashing back down to more realistic totals.

The two exceptions to this pattern are Tim Thomas and Nicklas Backstrom – Thomas had a two-year run while Backstrom had three years with a save percentage above 0.900 on the penalty-kill. Backstrom’s run left town with Jacques Lemaire while Thomas saw his numbers decline a bit last season.

Interestingly, Nashville has seen this happen before – Chris Mason and Dan Ellis both had brief runs as the Predators’ starter, mostly thanks to a strong penalty-killing save percentage. Then they lost the top job when the numbers while shorthanded went the other way. The Predators clearly didn’t pick up on exactly what happened in those two cases.

Given Rinne’s history – two less than stellar years on the penalty-kill, followed by a brilliant performance in 2010-11 – and given what we’ve seen from high-end performers like Lundqvist and Miller, does it make sense to bet on Rinne repeating his performance? Does it make sense to wager on it continuing for the duration of his seven-year contract? No, it doesn’t.

It’s a particularly baffling decision on the Predators’ part given what the goaltending market has looked like the last few years. Pekka Rinne’s current $3.4 million cap hit ranks him 17th among NHL goaltenders – right about average, in other words. Nobody currently makes $7.0 million – we’ve seen high-end guys like Bryzgalov, Luongo and Thomas all sign in the $5.0 – $5.7 million range. For capable but unspectacular starters, the record is even worse; they’ve been going in the $1.0 – $3.0 million range. We’ve even seen some unbelievable bargains, like the Capitals inking Tomas Vokoun to a one-year $1.5 million contract.

This isn’t to say Rinne will implode, or anything like that. In two previous years he’s been an average to slightly above average NHL starter. That’s probably the range he’ll stay in for the duration of his contract.

Unfortunately, that’s simply not good enough for the Predators. They can’t spend to the salary cap maximum, so their dollars are even more dear than they are for most teams, and spending so extravagantly on a good but not elite goaltender may cost them other key pieces.

  • Spydyr

    @ Smokey…I’m with you on needing a defencemen or two. For me it is just not Weber.He goes along the lines of a Hossa or Heatly.Searching for the great while whale.I would rather they traded for some young up and coming defencemen that can grow with the kids.You try for a Pronger or a Weber type player in a few years,when your ready to take a run at the cup.It is too soon for that IMO.

  • Spydyr

    I’d personally take a shot at Weber as UFA next year. And if you get a guy like, I think you would be a contender. Weber is a defensemen that makes a team a contender for a long time.

  • ubermiguel

    Luongo’s cap hit is only $5.3 mill. Rinne’s not worth $1.7 mill more than Loungo to his respective franchise. And I find it odd it’s $7 mill each year, not front loaded like Bryzgalov. Overpay him in 12/13 and save some cash for Suter and Weber.

  • The issue is that 2010-11 is Rinne’s only spectacular season – 2009-10 and 2008-09 were good, but hardly elite-level performances.

    So, does his 2010-11 performance reflect his true level of talent, or was it a result of a great year on the PK? Nashville’s paying Rinne to be an elite goaltender – they’re betting that 2010-11 reflects his true level of talent better than the previous two years. I’m looking at the previous two years, noting a crazy run on the PK, and guessing that the Rinne of the future will look more like the Rinne of 2008-10 rather than the Rinne of 2010-11.

    Yes, PK SV% is a single stat. It’s also a stat that seems to mean something, based on history.

  • justDOit

    Looks like the price for elite goalies is on the rise again, after Philly’s offer for Bryzgotalotov. A big mistake, because with so many goalies coming out of nowhere for a while now, it’s way more important to keep the stud D-men – of which, N’ville has TWO!

    Defensive teams should refuse to pay any goalie more than $3M/yr. Trade em when they get too big for their britches and tell FAs to find some other wall of Weber/Suter (Stevens/Nied) to play behind, if they want more.

    ‘Tenders who excel while playing behind an inferior defense should be the ones who make big $$$! Well, like Khabby, for instance. How many rookies or current AHLers has be played behind already?

    Khabby should be making at least $7M this year, and he’s signed for a reasonable $3.75! Almost half price! Can we afford NOT to trade him now with his value so high? For the benefit of the team’s future?

  • Little Buttcheeks

    I read the first part of the article up until you started getting into the stats. Having not read that part, I agree that my first thought was that it was a mistake. From a publicity standpoint, it would have been hard to walk away from a guy like Rinne, but I would much rather have the horses on D than the the highly paid goaltender because goaltending can be so inconsistent.

  • Clyde Frog

    I dislike dumping that amount of cash in net, not that I don’t like Rinne…

    I think he is a pretty damn good goalie,just think that the extra 4 mil a season you pay in cap hit could buy back the goal differential and then some a goalie with a .911 versus a goalie with a .929 save % would have.

    I just don’t see a huge statistical difference between elite goalies and good ones. Yes its there, don’t get me wrong. But it just doesn’t warrant a 4-5 million difference in the end impact for the team.

    You can argue mind set or they provide confidence for the team to do things they wouldn’t otherwise in front of them. But again throwing that extra cash at upgrading really decent top 6 free-agent or top 3 defencemen just seems so much more sexy for me.

  • I agree with the idea. I definitely agree signing to that term and that dollar value doesn’t make sense. What I don’t agree with is constantly shipping out players because they don’t meet statistical criteria. At some point you gotta make a run with what you’ve got and I guess they figure that this is the time. I’d be ok with 3 or 4 years 7 million if that’s the case though.

  • Jason Gregor

    Curious why you are basing everything on PK sv%…21.1 of the goals they gave up were on PK last year…

    Year before when you said his SV% was much worse…PK gave up 26% of team goals…

    In 2009 it was 25% of team goals…

    Is using a stat that takes up a quarter of the icetime a fair indicator of how good he is?

    • Jason Gregor

      Agreed. But what I have found out is that the guys who base everything on advanced stats will use any stat to make an argument but neglect others. Advanced stats arguments just make me laugh. Yes they are useful for seeing certain things but I would rather watch the game and see how a guy plays, the kid line puts up points but they do it against soft competition with no dzone starts so its not that good…um ya it is because they put the puck in the net and we win games.

      It was like the RNH PP points argument, well I am happy RNH is getting PP points here because our PP is not the worst in the league anymore and we are winning games.

      But it’s not like advanced stat arguments don’t grind my gears or anything….

      • Ogden Brother Jr. - Team Strudwick for coach

        Oilcan, it is called Confirmation Bias, and it is the predominant issue in any statistical analysis.

        Like any researcher, it is difficult to completely neutral on an issue, so the comment is fair. However, Willis does identify patterns through statistical analysis that are then open to interpretation – correlation vs causation, as an example.

        People can write off stats as much or as little as they like, but there are underlying patterns that are evident in stats. It’s just a question of identifying the correct patterns, which is basically every researcher’s goal in anything that uses statistical analysis (i.e. everything).

        What we have is a lack of information – did Nashvile’s management group use the PK% to arrive at a reasonable contract? If so, Willis’ argument is with merit. If not, then the article is of less value. We can’t answer that question (well, I can’t), so the article is what it is: identification of a pattern with speculation as to its improtance. Nothing more.

  • Hemmertime

    I also think you are putting way too much value in this one statistic. The 7 mill per year hit does set a new high water mark for goal tenders and it is surprising that the Preds and Rinne are the partners that set this standard.Preds have hitched their wagon onto Rinne and either Weber or Suter, or maybe both will be leaving within a year or two. This will fuel speculation that the Oil should take a run at Weber.

  • French Toast Mafia

    Not sure why you would only choose to use penalty killing save percentage to make a point…

    He is an unreal goalie. 7 million a season is a lot of money but if he wins a couple vezinas and is in the top 10 goalies over that period of time then I think the preds will be happy.

    I would take Rinne for 7 years over being stuck with Luongo for the next 10 or Dipietro for 15.

    • misfit

      I think the point was pretty clear that PK save percentage is something that generally swings wildly from year to year for no real reason (aside from luck, I’d say) and banking on it to stay at unusually high levels is a bad idea.

      And if those unsustainable PK numbers are skewing the overall SV% numbers which I would think had some baring on why he was given the contract he received, then it’s definitely something worth looking into.

      Kind of like giving RNH a contract on the basis of him being a 40 goal scorer and maintaining his 22% shooting percentage.

      With all that said, Rinne is a fine goaltender and one of the league’s best at the moment, but I do agree that signing Rinne to that deal was probably a bad move. At best, an unnecessary risk given what’s out there on the goalie market every year.

  • French Toast Mafia

    Not sure why you would only choose to use penalty killing save percentage to make a point…

    He is an unreal goalie. 7 million a season is a lot of money but if he wins a couple vezinas and is in the top 10 goalies over that period of time then I think the preds will be happy.

    I would take Rinne for 7 years over being stuck with Luongo for the next 10 or Dipietro for 15.

  • BobB

    Two seperate issues here.

    1. One ignores how good or bad Rinne is (he’s very good btw). I don’t think any team can/would offer him $7/year. Nashville overpaid… and did so early. Doesn’t make sense to me.

    2. How good is he? I understand Willis and his effort to show that Rinne’s pksv% last year was an outlier, which pumps up his base sv%, but the effect is very slight.

    Rinne’s evsv% in ’10-’11 was .932 (very, very elite).

    His pksv% was high (an outlier) at .912.

    His final sv% was incredible at .930.

    Even if he allowed three more goals on the pk… massively dropping his pksv% to a high, but more “normal” .892 (a big difference in pksv%, but a slight move in overall sv% — .928

    four more pk goals GA = .888 pk sv% (15th of regular starters) and overall sv% = still .928

    The idea that Rinne’s pksv% was very high and an outlier is true, however, the idea that if it was more “normal” that he would somehow have given an average line of stats, is false.

    He still would be second in overall sv% at .928 (tied with Luongo but with more shots against) even if his pksv% was .888 vs being .912 in ’10-’11.

    This further demonstrates why we should be looking at evsv% for our basic comparisons.

  • Romulus' Apotheosis

    Why did Willis only use PK stats? Because it best backs up his opinion. Just like most of his articles.

    My opinion is that Rinne is not nearly as good without Weber and Suter playing almost half of the game in front of him. I don’t need stats to make that statement. The funny part is that signing him to this contract almost guarentees that they will not be signing both Suter and Weber. This contract is destined to fail.

  • French Toast Mafia

    As Scotty Bowman said, “Show me a good defensive system and I’ll show you a good goalie”.

    So either Webber and Suter see this as a commitment from management to win or see it as the nail in the coffin because they lost some ability to get high end scoring help and have depth. In today’s market that is a lot of money for a goalie and not necessary.

  • Ogden Brother Jr. - Team Strudwick for coach

    “Why did Willis only use PK stats? Because it best backs up his opinion. Just like most of his articles.”

    Guys, guys, guys, guys.
    Lets quit sitting on our shoulders.

    Rinne’s Overall save % last 3 seasons:
    .917, .911, .930

    He is now getting paid for the .930.

    Rinne’s SH Save% last 3 years:

    The point being made is the .912 from last season drove up his overall number. This makes Rinne look elite rather than good, and that the .912 is NOT repeatable.

    • Did the SH% drive up the overall numbers? Or did him being a good goalie drive up all the numbers? Or did having a better defence drive all his numbers up or down each year?


      You say the .912 isn’t repeatable, but Willis’ chart clearly shows that it can (and has) been done twice by very good goalies on very good defensive teams.

  • misfit

    If you’re going to use stats to make a point or to research whether or not a contract is valued correctly, you use ALL available stats. Otherwise you ruin the credibility of your research. Willis has continually used what stats he seems fit to make his points. Which is why he holds no credibility with me. That’s just my opinion.

  • You’re right,
    I shouldn’t have typed “NOT”, “probably not” would have been better.

    But, last season,
    Rinne’s ES SP went up 1%
    His PP SP went up 11%
    over the prior two seasons.

    Maybe Rinne is a magical SH goaltender,
    7 x 7 is a large bet on scanty evidence.

  • Kudos to the blog Jonathan. Couldn’t agree more with everything you said about advanced stats, the goalie market, etc. Keep them coming! Giving players long term contracts decreases motivation over the short term. Once things start going badly, the pressure and expectations can ruin a player. We all know how finicky goalies can be and the fact that Nashville cant score wont help Pekka’s cause over the long term.