### Thursday, November 08, 2007

## PMR RUN CONVERSIONS, SHORTSTOPS

David Pinto has started posting his Probabilistic Model of Range figures for teams and individuals, starting the individuals off with the shortstops. Here are the conversions (there's an explanation in the relevant section of the sidebar to your right, as well as figures from past seasons):

1. David, along with the individual totals, listed the team totals, which also means that we have the major league-wide totals. This is great; one concern with the first couple of years of these conversions was that I had to guess at what the league average really was in any given season, as David was using multi-year probabilities. This led to weird happenstances where every third baseman was "above average", which makes no sense, so I had to guess at what the average really was. The last two years, David seems to be using one-year probabilities, and in fact major league shortstops were "predicted" to have made 15,904 outs last year when they really made 15,913. This is a negligible difference, and the predicted DER and observed DER are the same for several decimals. (The MLB total also allows me to properly state the number of predicted outs a SS would have in 150 games, which is 490 [actually 491, so sue me for rounding], and what the second column represents.)

2. At some point I need to go back and re-code the various PMR models, but David is using something called the "smoothed visitor model", which is I believe the second version of PMR. He did this last year, as well.

3. As such,

4. In 2005, we saw, in addition to these figures, groundball-only figures for infielders. For the past two seasons, the figures have been for every kind of batted ball. Zone rating and MGL's Ultimate Zone Rating do

Player Runs Runs/490A few points:

Troy Tulowitzki 38.0 33.0

Tony F Pena 23.0 25.1

Rafael Furcal 20.9 23.0

Jason Bartlett 16.9 18.7

John McDonald 12.6 21.0

Jimmy Rollins 12.4 11.8

Jack Wilson 9.7 10.4

Jhonny Peralta 7.3 7.1

Omar Vizquel 4.7 4.6

Orlando Cabrera 3.8 4.1

Julio Lugo 3.7 4.2

Yunel Escobar 2.7 9.9

Adam Everett 1.8 4.1

Alex Gonzalez 0.0 -0.1

J.J. Hardy -0.3 -0.3

Cesar Izturis -0.3 -0.6

Bobby Crosby -0.6 -0.9

Stephen Drew -0.9 -1.1

Mark Loretta -1.0 -2.7

Eric Bruntlett -1.4 -5.0

Ryan Theriot -1.6 -2.5

Marco Scutaro -1.6 -6.4

Royce Clayton -2.1 -5.0

Hanley Ramirez -2.2 -2.4

Yuniesky Betancourt -2.7 -2.8

Khalil Greene -2.7 -2.6

Edgar Renteria -3.1 -4.2

Jeff Keppinger -4.3 -15.4

David Eckstein -6.5 -8.9

Josh Wilson -7.8 -25.3

Miguel Tejada -7.9 -10.3

Juan Uribe -8.6 -8.1

Jose Reyes -9.0 -8.6

Cristian Guzman -10.5 -39.4

Felipe Lopez -14.1 -18.3

Carlos Guillen -14.4 -17.2

Brendan Harris -14.4 -27.9

Michael Young -21.8 -21.1

Derek Jeter -30.6 -32.5

1. David, along with the individual totals, listed the team totals, which also means that we have the major league-wide totals. This is great; one concern with the first couple of years of these conversions was that I had to guess at what the league average really was in any given season, as David was using multi-year probabilities. This led to weird happenstances where every third baseman was "above average", which makes no sense, so I had to guess at what the average really was. The last two years, David seems to be using one-year probabilities, and in fact major league shortstops were "predicted" to have made 15,904 outs last year when they really made 15,913. This is a negligible difference, and the predicted DER and observed DER are the same for several decimals. (The MLB total also allows me to properly state the number of predicted outs a SS would have in 150 games, which is 490 [actually 491, so sue me for rounding], and what the second column represents.)

2. At some point I need to go back and re-code the various PMR models, but David is using something called the "smoothed visitor model", which is I believe the second version of PMR. He did this last year, as well.

3. As such,

**. I believe the smoothed visitor model was introduced in 2005, so it's the same version -- but my "re-center" of the MLB average was an estimate and not an absolute. This caveat will apply for every position.***use caution when comparing these figures to years before 2006*4. In 2005, we saw, in addition to these figures, groundball-only figures for infielders. For the past two seasons, the figures have been for every kind of batted ball. Zone rating and MGL's Ultimate Zone Rating do

*not*account for popups and line drives, so, again, use caution when comparing these figures to runs generated by or estimated from other systems.
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

Advantage #4: Get more with weekly and special alluviation vast bulk go directly into the more innovative Vegas-like picture slots games. [url=http://www.tasty-onlinecasino.co.uk/]online casino[/url] online casinos uk At times the bonus that divine it; Scarface is a definite pull in any casino. http://www.onlinecasinoburger.co.uk/

Post a Comment