In my column this morning on Baseball Prospectus (subscription required but well worth it) among other things I take another crack at the PITCHf/x Gameday data. In part, inspired by this fascinating article, I created a couple of metrics to quantify plate discipline. They are:
Without further ado here's the list (with the big caveat that only 24% of all pitches in 2007 have been tracked in the system and there is a heavy bias to the AL West because of the parks that the system is installed in) of all players who have 200 or more pitches tracked this season sorted by "Fish".
Name Stand Pitches Swing Fish BadBall Eye
A.J. Pierzynski L 357 0.602 0.469 0.813 0.117
Garret Anderson L 260 0.515 0.467 0.750 0.350
Delmon Young R 200 0.550 0.454 0.630 0.228
Rob Mackowiak L 297 0.525 0.438 0.738 0.240
Hank Blalock L 332 0.521 0.430 0.632 0.183
Erick Aybar L 235 0.532 0.430 0.869 0.217
Carl Crawford L 216 0.519 0.427 0.738 0.171
Jeffrey Francoeur R 337 0.546 0.425 0.775 0.094
Vladimir Guerrero R 403 0.524 0.421 0.771 0.142
Adrian Beltre R 430 0.512 0.403 0.694 0.234
Michael Young R 585 0.523 0.403 0.766 0.284
Eric Chavez L 519 0.503 0.394 0.736 0.225
Kenji Jojima R 270 0.522 0.389 0.878 0.217
Nomar Garciaparra R 436 0.557 0.388 0.800 0.132
Juan Pierre L 431 0.480 0.385 0.916 0.302
Yuniesky BetancourR 345 0.496 0.383 0.806 0.279
Joe Crede R 376 0.503 0.381 0.729 0.238
Jason Phillips R 203 0.493 0.381 0.721 0.303
Vernon Wells R 417 0.501 0.380 0.713 0.216
Jose Lopez R 347 0.478 0.375 0.853 0.288
Ichiro Suzuki L 515 0.435 0.369 0.862 0.326
Khalil Greene R 407 0.514 0.367 0.595 0.247
Orlando Cabrera R 517 0.468 0.362 0.838 0.279
Brian Mccann L 267 0.434 0.358 0.783 0.282
Kevin Kouzmanoff R 363 0.499 0.350 0.724 0.167
Lyle Overbay L 381 0.465 0.349 0.704 0.230
Sammy Sosa R 532 0.492 0.349 0.625 0.205
Royce Clayton R 227 0.515 0.348 0.457 0.194
Adam Lind L 336 0.482 0.348 0.719 0.253
Andruw Jones R 389 0.478 0.344 0.686 0.232
Jose Guillen R 394 0.475 0.344 0.636 0.232
Alexis Rios R 413 0.426 0.343 0.735 0.351
Marcus Giles R 452 0.511 0.341 0.674 0.211
Bobby Crosby R 474 0.451 0.338 0.652 0.327
Casey Kotchman L 463 0.434 0.338 0.814 0.285
Gary Matthews Jr. L 449 0.441 0.334 0.781 0.223
David Dejesus L 209 0.445 0.333 0.816 0.340
Nelson Cruz R 374 0.457 0.330 0.690 0.268
Andre Ethier L 376 0.500 0.329 0.803 0.200
Jose Vidro L 336 0.461 0.329 0.870 0.200
Juan Uribe R 334 0.476 0.328 0.651 0.232
Raul Ibanez L 474 0.456 0.328 0.760 0.229
Gerald Laird R 484 0.459 0.326 0.698 0.290
Jason Kendall R 405 0.432 0.325 0.870 0.373
Jermaine Dye R 452 0.442 0.324 0.753 0.267
Mark Teahen L 213 0.437 0.324 0.636 0.307
Adrian Gonzalez L 590 0.453 0.322 0.767 0.225
Darin Erstad L 372 0.398 0.318 0.797 0.413
Shea Hillenbrand R 330 0.445 0.308 0.821 0.297
Brad Wilkerson L 263 0.414 0.306 0.755 0.303
Paul Konerko R 457 0.420 0.303 0.765 0.290
Richie Sexson R 448 0.446 0.303 0.732 0.211
Mark Ellis R 440 0.409 0.302 0.851 0.380
Mark Teixeira L 501 0.399 0.302 0.657 0.276
Kenny Lofton L 514 0.434 0.300 0.818 0.239
Travis Buck L 377 0.416 0.300 0.662 0.255
Michael Napoli R 346 0.419 0.299 0.683 0.322
Kelly Johnson L 384 0.393 0.292 0.787 0.315
Josh Bard L 283 0.470 0.292 0.878 0.177
Edgar Renteria R 346 0.405 0.290 0.688 0.275
Jeff Kent R 420 0.476 0.285 0.611 0.176
Ian Kinsler R 591 0.423 0.280 0.784 0.268
Terrmel Sledge L 285 0.446 0.272 0.609 0.217
Rafael Furcal L 371 0.391 0.270 0.857 0.353
Shannon Stewart R 437 0.410 0.269 0.810 0.327
B.J. Upton R 225 0.431 0.269 0.528 0.227
Frank Thomas R 478 0.377 0.267 0.838 0.335
Troy Glaus R 306 0.376 0.264 0.706 0.330
Jose Cruz Jr. L 285 0.428 0.262 0.591 0.216
Aaron Hill R 417 0.415 0.253 0.729 0.302
Willie Harris L 205 0.400 0.248 0.765 0.239
Russell Martin R 506 0.401 0.241 0.757 0.290
Luis Gonzalez L 428 0.395 0.236 0.889 0.256
Nick Swisher L 384 0.365 0.232 0.559 0.307
Tadahito Iguchi R 462 0.422 0.230 0.678 0.243
Frank Catalanotto L 309 0.372 0.228 0.821 0.319
Mike Cameron R 539 0.399 0.224 0.569 0.237
Reggie Willits L 274 0.328 0.219 0.818 0.455
Brian Giles L 381 0.396 0.218 0.804 0.269
Jim Thome L 330 0.358 0.203 0.545 0.239
Bobby Abreu L 203 0.374 0.185 0.545 0.274
Jack Cust L 294 0.310 0.175 0.563 0.342
Wilson Betemit L 208 0.327 0.155 0.700 0.282
Dan Johnson L 387 0.282 0.148 0.737 0.346
Of course the interesting thing is that you can then plot the Fish and Eye metrics on a graph and then split the graph into four quadrants. Each quadrant creates a little profile that can be used to characterize a hitter's plate discipline. Note that the bottom left is the sweet spot.
5 comments:
Great stuff, Dan. Could you post the averages for the 4 metrics so far?
I wonder if you might not be better served in the long-run by inverting either Fish or Eye, so that a high score is either good or bad on both. It's a little confusing to have them reversed.
Will the combined metric be called "fisheye"?
When we have more data, you might consider breaking this down by home and away. A big part of home-field advantage is often a K and BB differential. It would be interesting to see if pitchers actually throw more strikes at home, or if visiting hitters have more trouble reading the strike zone, or some combination.
Dan, thanks so much for the hat tip. I sent you an e-mail with some comments. Good work!
Guy, the averages (for all players with 100 or more pitches seen) are
Swing: 44.8%
Fish: 32.1%
BadB: 73.2%
Eye: 26.7%
I also created another metric I didn't publish called Square which is the percentage of pitches in the zone that were swung at and made contact with. The average is 87.1% and there are several players at 100% including Chone Figgins, Placido Polanco, Derek Jeter, and Esteban German. On the flip side Russell Branyan is 61.2%, Rob Bowen 65.4% and Jack Cust 69.7%.
I totally agree with you on inverting one of them. I think Fish is the one to invert.
Great idea on breaking it down by home and road. Thanks as usual.
Pizza Cutter, you're welcome.
I'll be publishing the "Square" metric shortly which you referenced in your email as "Holes". Great idea.
Dan: another part of the home/away story might be umpires giving more calls to the home team. You may have enough data already to look at that at an aggregate level. According to your earlier article, umpires call a strike on 87.4% of actual strikes, and call a ball on 90.7% of actual balls (allowing for the 1-inch error margin). If there's a home-team bias, we'd see a lower called-strike rate and/or a higher called-ball rate when home team is at bat.
Post a Comment