Tango on Baseball Archives

© Tangotiger

Archive List

Forecast 2003 - Interim Results (July 31, 2003)

A huge thanks to Primate Sylvain was kind enough to compile the current stats for the players from the Forecast study. That's the study where we pit the baseline estimates (based on prior 3 seasons unweighted stats, plus an age adjustment) for 32 hard-to-forecast players against the professional forecasters' proprietary engines, and the Primer Readers back of the envelope calculations.

The interim results are:
- For hitters, the baseline estimate has an average error of .074 OPS, while the professional forecasters have an average error of .084, and the Primer readers check in at .096.
- For pitchers, we get the opposite results: Primer readers are off by .75 ERA, professional forecasters are off by .84, and the baseline is off by 1.05 runs.

Results follow:

						 Delta 	 	 Delta 	 	 Delta 	 	 Delta 
Player Baselin Profes Readers Current Baseline Forecasters Primer Reader Average
Barry Bonds 1.231 1.248 1.268 1.240 0.01 0.02 0.03 0.02
Jim Thome 0.979 1.030 1.019 0.922 0.05 0.12 0.10 0.09
Gary Sheffie 0.949 0.952 0.939 1.013 0.07 0.05 0.07 0.06
Troy Glaus 0.949 0.883 0.887 0.807 0.14 0.08 0.09 0.10
Luis Gonzale 0.934 0.931 0.871 0.947 0.02 0.01 0.07 0.03
J,D, Drew 0.895 0.865 0.873 0.888 0.00 0.02 0.01 0.01
Pat Burrell 0.895 0.892 0.956 0.699 0.19 0.20 0.26 0.22
Moises Alou 0.869 0.816 0.816 0.858 0.01 0.04 0.04 0.03
Richard Hidalgo 0.858 0.821 0.798 0.959 0.10 0.13 0.16 0.13
Jeremy Giambi 0.840 0.872 0.898 0.702 0.14 0.18 0.20 0.17
Sean Casey 0.808 0.786 0.766 0.783 0.02 0.01 0.01 0.01
Roberto Alomar 0.797 0.776 0.793 0.698 0.10 0.09 0.10 0.09
Jacque Jones 0.795 0.822 0.814 0.811 0.02 0.02 0.01 0.02
Torii Hunter 0.790 0.837 0.843 0.765 0.02 0.08 0.08 0.06
Rich Aurilia 0.773 0.790 0.798 0.718 0.05 0.08 0.09 0.07
Rondell White 0.770 0.726 0.776 0.791 0.02 0.06 0.01 0.03
Adam Kennedy 0.766 0.740 0.750 0.696 0.07 0.05 0.06 0.06
Jeromy Burnitz 0.741 0.737 0.724 0.885 0.15 0.14 0.16 0.15
Jeff Cirillo 0.740 0.689 0.723 0.566 0.17 0.13 0.16 0.15
Jose Hernand 0.717 0.831 0.836 0.658 0.06 0.18 0.18 0.14
Marquis Grissom 0.673 0.717 0.692 0.822 0.15 0.10 0.12 0.12
League OPS 0.760 0.750 0.751 0.757
ERROR 0.074 0.084 0.096 0.085
......
......
Freddy Garcia 3.41 3.81 3.86 5.17 1.82 1.24 1.19 1.42
Javier Vazquez 3.41 3.67 3.61 3.77 0.42 0.02 0.04 0.16
Ryan Dempst 4.19 4.62 4.56 6.54 2.41 1.80 1.86 2.02
Kip Wells 4.32 4.23 4.10 3.62 0.64 0.73 0.60 0.66
Chan Ho Park 4.59 4.70 4.64 7.58 3.05 2.76 2.82 2.88
Matt Clement 4.60 3.84 3.95 4.08 0.46 0.12 0.01 0.20
Jamey Wright 4.76 4.89 5.10 5.17 0.47 0.16 0.05 0.23
Aaron Sele 4.77 4.65 4.83 5.17 0.46 0.40 0.22 0.36
Shawn Estes 4.91 4.34 4.79 5.89 1.04 1.43 0.98 1.15
Kenny Rogers 5.35 4.41 4.52 4.94 0.35 0.41 0.30 0.35
Todd Ritchie 5.62 4.77 5.16 5.08 0.48 0.19 0.20 0.29
League ERA 4.49 4.31 4.31 4.43
ERROR 1.05 0.84 0.75 0.88



--posted by TangoTiger at 10:07 AM EDT
Posted 10:10 a.m., July 31, 2003 (#1) - tangotiger
  The deltas were calculated as follows, and I'll take Bonds as an example:

Baseline: Bonds = 1.231, lg=0.760, comp=+0.471
Current: Bonds = 1.240, lg=0.757, comp=+0.483

delta = Absolute value of (+.471 minus +.483) = .01

We can do it as a percentage as well, but I think this will do for now.

Posted 11:37 a.m., July 31, 2003 (#2) - adam
  could you post some standard deviations as well?

Posted 11:54 a.m., July 31, 2003 (#3) - tangotiger
  standard deviation, in the above order:

hitters: .094, .099, .115
pitchers: 1.00, .83, .83