Monday, November 27, 2006
PMR, RIGHTFIELDERS
Pinto.
The conversions are as follows. David has presented an alternative method of dealing with flyballs, using distance instead of velocity, so I will also present that. Players are sorted by the original method.
This is also not merely an artifact of the disagreement on Brian Giles. The distance-based PMR method is closer to ZR on nearly every player on the list. Take Giles out of the equation, and the original PMR correlates at .454, and the improved at an impressive .704.
I know that there is something unique and odd about Petco's rightfield, but I don't know if PMR is correct to downgrade Giles so far vis-a-vis what zone rating says.
Pinto.
The conversions are as follows. David has presented an alternative method of dealing with flyballs, using distance instead of velocity, so I will also present that. Players are sorted by the original method.
Original Method Alternative MethodThere are some ... stark differences for some of these players, though the correlation between the two methods is .75. This leads me to believe that velocity and distance are fairly strongly correlated, but mistaking one for the other can lead to some big mistakes. Let's see how both methods compare to zone rating:
Player Runs Above Average Runs Above Average
Total Per 285 Opp Total Per 285 Opp
Reggie Sanders 16.3911 30.9922 4.8739 8.4560
Casey Blake 15.6340 23.2528 1.4971 2.0489
J.D. Drew 13.9413 14.8473 6.6347 6.8461
Alex Rios 10.8281 15.0340 6.0052 8.1136
J. Encarnacion 8.6676 11.8302 10.2412 14.1030
Carlos Quentin 8.6506 28.7245 6.8218 22.0984
Mark DeRosa 8.5060 21.0801 3.5045 8.2625
Jose Guillen 8.0722 14.8895 5.2057 9.3971
Vlad Guerrero 7.9616 9.3132 2.9005 3.3121
Damon Hollins 7.9616 18.2049 7.4768 17.0185
Kevin Mench 7.7149 21.3617 -6.2434 -14.9101
Jacque Jones 7.1876 7.6851 6.8984 7.3665
Austin Kearns 6.8984 5.8186 2.7815 2.3129
Ryan Freel 5.1546 15.4737 6.8984 21.1652
Jay Gibbons 4.5422 14.1232 -3.3939 -9.5778
Emil Brown 2.9686 7.9434 1.3525 3.5555
Moises Alou 2.6879 5.0786 5.2482 10.1180
Russell Branyan 0.7911 2.6194 -1.3950 -4.4852
Bobby Abreu 0.3402 0.3314 0.2127 0.2070
Trot Nixon 0.0425 0.0572 -1.9309 -2.5682
Joe Borchard -0.0510 -0.1730 1.7692 6.1552
Jay Payton -0.3573 -1.1386 5.4864 18.9414
Jeff Francoeur -0.7911 -0.7091 -1.3525 -1.2099
Brad Hawpe -0.9016 -0.9143 -1.1653 -1.1804
Ichiro Suzuki -1.0292 -1.1677 7.4938 8.8550
Randy Winn -1.4630 -2.2451 1.8203 2.8526
Jason Lane -1.4800 -2.6912 -2.2796 -4.1203
Shawn Green -1.9479 -2.4974 -5.8862 -7.3927
Milton Bradley -2.9005 -4.2521 0.7315 1.0965
Nick Markakis -3.6831 -4.2962 -0.5104 -0.6045
Jermaine Dye -4.7719 -4.3784 -0.2977 -0.2779
Bernie Williams -5.1121 -14.0078 -5.8181 -15.8161
Geoff Jenkins -5.9882 -6.7180 1.5141 1.7597
Jeromy Burnitz -7.3492 -16.2820 -14.7409 -30.5917
Jeremy Hermida -8.0297 -13.7494 -2.5943 -4.6197
Mike Cuddyer -12.0615 -13.2631 -5.8521 -6.6216
Xavier Nady -13.0057 -18.3233 -4.2190 -6.2638
M. Ordonez -19.7850 -20.0480 -9.1354 -9.6882
Brian Giles -29.3287 -25.1404 -17.4798 -15.6388
Player ZR-R PMR-1 Diff PMR-2 DiffThe original PMR correlates very poorly to ZR, only .165. The new, distance-based method correlates at .381. This does not surprise me, as distance, not velocity, is a parameter in the calculating of zone rating.
Alex Rios 11 11 0 6 5
Austin Kearns 5 7 2 3 2
Bernie Williams -7 -5 2 -6 1
Bobby Abreu -4 0 4 0 4
Brad Hawpe -9 -1 8 -1 8
Brian Giles 12 -29 41 -17 29
Casey Blake 4 16 12 1 3
Damon Hollins 3 8 5 7 4
Geoff Jenkins -2 -6 4 2 4
Ichiro Suzuki 7 -1 8 7 0
J.D. Drew 5 14 9 7 2
Jacque Jones 4 7 3 7 3
Jason Lane -8 -1 7 -2 6
Jeff Francoeur -1 -1 0 -1 0
Jeremy Hermida 4 -8 12 -3 7
Jermaine Dye -2 -5 3 0 2
Jeromy Burnitz -12 -7 5 -15 3
Juan Encarnacion 9 9 0 10 1
Kevin Mench -6 8 14 -6 0
Magglio Ordonez -2 -20 18 -9 7
Mark DeRosa 2 9 7 4 2
Michael Cuddyer -7 -12 5 -6 1
Milton Bradley 7 -3 10 1 6
Moises Alou -6 3 9 5 11
Nick Markakis -2 -4 2 -1 1
Randy Winn 2 -1 3 2 0
Reggie Sanders 6 16 10 5 1
Russell Branyan -5 1 6 -1 4
Trot Nixon 0 0 0 -2 2
Vladimir Guerrero -5 8 13 3 8
Xavier Nady 0 -13 13 -4 4
This is also not merely an artifact of the disagreement on Brian Giles. The distance-based PMR method is closer to ZR on nearly every player on the list. Take Giles out of the equation, and the original PMR correlates at .454, and the improved at an impressive .704.
I know that there is something unique and odd about Petco's rightfield, but I don't know if PMR is correct to downgrade Giles so far vis-a-vis what zone rating says.
Labels: David Pinto, fielding, probabilistic model of range
Comments:
視訊做愛視訊美女無碼A片情色影劇aa免費看貓咪論壇彩虹性愛巴士金瓶梅影片交流yam視訊交友xxx383美女寫真kyo成人動漫tt1069同志交友網ut同志交友網微風成人論壇6k聊天室日本 avdvd 介紹免費觀賞UT視訊美女交友自拍密錄館sex888情人輔助品哈啦聊天室豆豆出租名模情人視訊視訊交友網視訊交友90739影片 圖片av168成人日本A片免費下載 金瓶梅影片交流免費A片下載85cc免費影城85cc日本a片情色a片無碼女優 免費色情電影同志聊天室38ga成人無碼a片小魔女免費影片玩美女人影音秀台灣18成人網18禁成人網聊天室ut歐美嘟嘟情人色網影片18禁地少女遊戲a383禁地論壇成人影城18禁av影片無碼線上LIVE免費成人影片sex女優松島楓免費影片咆哮小老鼠論壇色咪咪情色網 視訊熱舞秀ut台中聊天室貓貓論壇豆豆情色風暴視訊xxx383美女寫真? 線上漫畫免費線上a片無碼dvdxvediox日本美女寫真集免費成人電影小魔女自拍天堂av1688影音娛樂網0204movie免費影片咆哮小老鼠論壇85cc免費影城85ccfoxy免費音樂下載免費視訊免費影片成人影城免費a網 免費視訊辣妹彩虹頻道免費短片av1688天使娛樂網辣妹妹影音視訊聊天室視訊網愛聊天室後宮電影電影院蜜雪兒免費小說洪爺情色論壇sexy girl video movie視訊交友90739無碼dvd維納斯成人用品辣妹貼圖a片天堂月光論壇sexy girls get fucked中國性愛城sex520-卡通影片383movie成人影城ut正妹 聊天室倉井空免費a影片伊莉論壇tw 18 net18禁成人網免費性愛影片影音視訊聊天室av168成人視訊交友視訊美女視訊交友
# posted by 
: 12:32 AM
: 12:32 AM
This is also not merely an artifact of the disagreement on Brian Giles. The distance-based PMR method is closer to ZR on nearly every player on the list. Take Giles out of the equation, and the original PMR correlates at .454, and the improved at an impressive .704.
cheap wholesale t shirts free shipping
buy bulk shirts for printing
Post a Comment
cheap wholesale t shirts free shipping
buy bulk shirts for printing
