Wednesday, July 28, 2010

Tomas Fleischmann's Worth

With Tomas Fleischmann just having signed a one year, $2.6 million deal, I thought I'd delve into the statistics to see what the numbers suggest Fleischmann is worth.

Fleischmann had a GVT of 9.9 this past season. GVT is goals-versus-threshold, a rough hockey equivalent of baseball's VORP, measured in goals over the replacement level player (both fewer goals allowed and more goals for). Wins is roughly GVT divided by 6, and the Capitals had 54 wins in 2009-2010, so roughly Fleischmann was responsible for 3.05% of the team's success--1.65 wins (and zero playoff series wins).  GVT, as stated by its creator, Tom Awad, undervalues defensive contributions. By Fleischmann's rate statistics (easily found at Behind the Net), Fleischmann's value is clearly on offense, and hence, other players, like Boyd Gordon and Matt Bradley, are undervalued, so Fleischmann's salary given above is slightly higher than what the ratios say it should be. That being said, the actual amount spent on Fleischmann, $725k, is quite a steal for a 20-goal 50-point wing/center.

The Capitals, I'm guessing, will see regression to the mean in their team shooting percentage, which was abnormally high this season at 11.6%. If they take the same amount of shots but score at a lower rate--say 10.0%--then their goals total on 2693 shots becomes 269 goals for.

Defensively, the Capitals allowed 30.9 shots per game, for 2535 total on the season. If we factor in a small improvement--from fewer penalties and so forth--then, well, we can estimate the shots against per game for next season, arbitrarily, to roughly 30 shots against per game, for 2460 shots against on the season. If we apply an average goaltending save percentage of .910 to that shots against total, we get 221 goals against.

After applying a Pythagorean Winning Percentage Formula,

PTS% = 1.12 * (Goals For)^2 / ( (Goals For)^2 + (Goals Against)^2)

we get pts% = 1.12 x 72361 / ( 72361 + 48841) = 66.9% which equates to 110 points (which is reasonable). With an NHL squad losing, on average, 10 games in overtime or shootout, the Capitals will have gained 100 points by wins, so that means 50 wins (again reasonable). 3.05% of 50 wins = 1.53 wins, just a little less than 2009-2010. 50 wins against $59.4 million in possible salaries leaves about $1.1 million per win, so with 1.53 wins Fleischmann should be valued around $1.7 million.

This analysis, though, was purely based on the assumption that the Capitals will operate very similarly in 2010-2011 as in 2009-2010. This is likely not the case. What may change:
  • Tomas Fleischmann shot 19.0% this season! That's simply not sustainable for a longer range shooter (and even difficult for a guy like Mike Knuble to do). If that goes down, so will Fleischmann's goal and point totals. However, if he shoots more, he could compensate and actually increase his scoring (and his Corsi numbers too). 
  • Tomas Fleischmann, with every passing day, looks like the second line center on opening night. Although he pretty much lacks the strength, defensive acumen, and faceoff ability to effectively play center, his time on ice per game should stay high, he'll get plenty of power play time, and put up points by virtue of playing center (most of the top scorers are centers. I think that's because of their position at least in part, but I also realize that many of the better younger players will play center).
  • Alex Ovechkin on his first extension is getting paid to hit 30 in GVT for around $10 million a season, so 3 GVT per million dollars, or half a win. Nicklas Backstrom is around 3/5 a win per million dollars. Eric Fehr is around 2/3 with his $2.2 million contract (and with more TOI, he should be a huge bargain). Fleischmann at $2.6 million is around 2.5 GVT per million dollars, and hence under half a win per million dollars.
  • I did not factor in games played. Since GVT is an absolute, not rate, statistic, amount matters, and Fleischmann played a lot when he did play, but missed time as well. If he plays more games he should play a bigger part on the team and thus should earn more money.
  • For almost every player signed as a restricted free agent, a contract is done with hope--and compensation--as if the player in question could or will improve a bit the following season or seasons. We should think that Fleischmann will improve a bit more next season, and he should be compensated as such.
  • RFAs on average take around a 40% discount compared to UFAs (numbers courtesy of Gabriel Desjardins at Behind the Net Hockey). If Fleischmann takes a discount on that $1.7 million then he goes to $1.02 million. Definitely too low.
All that being said, $2.6 million seems high. At the RFA discount rate we take it to mean that Fleischmann as a UFA would be valued by the Capitals as roughly $4.3 million player.

I personally have trouble fathoming Fleischmann being worth more than $3 million. But $2.6 million seems okay, especially on a one year contract. More importantly to me, though, it's very trade-able.

(This is really one of my first forays into using statistics. If you have any suggestions on what I could do better, logical errors, and such, then please leave a comment or drop me an email)

    1 comment:

    1. Interesting - interesting to see that a much more statistical analysis than my "seat of the pants" WAG comes up with the same conclusion - at $2.6M the Capitals are likely OVERPAYING "Flash" by between $500K and $1M.

      Also my only other point is re: "Tomas Fleischmann, with every passing day, looks like the second line center on opening night. Although he pretty much lacks the strength, defensive acumen, and faceoff ability to effectively play center,..." No "pretty much" about it - currently based on what he showed last season when playing Center - Flash TOTALLY lacks all those things and is not likely to be effective at the position...