from Amazon
Tuesday, September 08, 2009
Baseball Guts, part 2
Three years ago, I took a stab at creating a model of batted ball out conversions, where I guessed this:
Freq___Out Rate___Great Fielder___Bad Fielder___Type of Play
40%___100.0%___100.0%___100.0%___Automatic
10%___97.0%___99.0%___94.0%___Automatic10%___93.0%___98.0%___83.0%___Automatic
5%___80.0%___90.0%___60.0%___Some Effort5%___60.0%___70.0%___40.0%___Some Effort
5%___40.0%___60.0%___20.0%___Alot of Effort
5%___20.0%___40.0%___10.0%___Alot of Effort10%___10.0%___30.0%___5.0%___Highlight Reel
10%___0.0%___0.0%___0.0%___Highlight Reel
100%___70.0%___75.7%___64.7%___ALL
Sky actually recorded 200 balls in play, gave it his own 5-point scale, to come up with:
FREQ OUT_RATE TYPE_OF_PLAY
48% 99% Automatic
18% 90% Some Effort
11% 50% Toss Up
7% 10% A lot of Effort
17% 1% Highlight Reel
100% 70% TOTAL
As you can see, my “baseball guts” model aligns itself fairly well with Sky’s sample of 200 observations. I pretty much presumed they would because I knew that the overall out rate had to come in at 70%. Basically, it would be kind of hard to come up with a model that didn’t match reality (if you had some interest in baseball, anyway).
I love that Sky did the work and reported his observations.
Anyway, he does some cool work, and concludes:
In contrast, the standard deviation of linear weights batting runs in one plate appearance is about 0.43 runs (compared to a SD of 0.19 runs for fielding).
Given that the true spread in talent is double on the hitting side than the fielding side, it is not a surprise that the standard error of observed performance will also be proportional to the talent level.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date