from Amazon
Thursday, December 08, 2011
Crappy save percentage, just by luck?
This Washington writer asks, and tests:
Vokoun and Michal Neuvirth have an underwhelming .896 save percentage and 3.09 GAA.
Is it possible it’s just two goalies experiencing slumps that just happened to coincide with Vokoun signing with Washington? Not sure, so let’s ask the Monte Carlo Machine.
Specifically, I’m going to simulate 10,000 26-game seasons based on the career save percentages of Vokoun and Neuvirth. For each game they have started in 2011-12, I will use a random number generator to determine what their save percentage was based on their historical performances to date. Then, I will average those games to see if both players were at or below their actual 2011-12 averages. Finally, I will repeat this process 10,000 times and count how many simulations contained averages at or below the actual 2011-12 save percentages, which will give us the probability that the declines of Vokoun and Neuvirth have been due to random fluctuations in performance.
On to the results!
Out of 10,000 simulations, 2,457 contained stretches in which both Vokoun and Neuvirth simultaneously put up average save percentages as low as they’ve posted so far in real life. This means there’s a 24.6 percent chance that Vokoun and Neuvirth currently have the same inherent “save skills” as their post-lockout averages suggest, but have merely gone through two simultaneous slumps.
Now, you didn’t really need a sim to do all that. Vokoun has a .917 career save percentage, and his backup has a career .908. Their weighted average is .914. They’ve faced 815 shots, so, .914 save percentage means 70 goals expected compared to 84 actual goals, or a difference of 14 goals.
One standard deviation is sqrt(.914 * (1-.914) * 815) = 8 goals, or 1.75 standard deviations from the mean. That’s 4% chance of luck. I don’t know where he got 24.6% though. Maybe he used a lower save percentage? 24.6% means only .69 standard deviations from the mean, which would imply a difference of only 5.5 goals, which would imply an expect save percentage of .904 (instead of the .914 I used).
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date