Stealing on Catchers (Not Named Molina)

I spent the last week on my newest tool, which analyzes play by play in order to rate catchers on their throwing.

The task is to seperate the catcher’s ability to throw out base stealers from that of the pitchers they are teamed with. My initial table, extracted from the RetroSheet events for 2003-2008, contains IDs for the catcher, pitcher, baserunner, the hand of the pitcher and batter, natural or artificial turf, and the the number of steal opportunities and total of each type of result for each combination of these factors. For each catcher, for each season and base, there is how he did with each pitcher (the observed values). In a WOWY fashion, that is compared to the results for each of those pitchers, over the past six seasons, with every other catcher (the expected values). The sum of all players forms the mean values. Using a variation of the Odds Ratio I’ve called the Inverse James Function, I then calculate what true talent level would give us that observed value, given the expected and the mean.

The mean CS% for the past six seasons is .243. If a player’s observed value is the same as his expected, then his normal value is equal to the league mean. If the observed is higher (or lower) than expected, then the normal will be higher (or lower) than the mean, with the normal value limited to between 0 and 1.

Check out the top 5. What’s on that gene pool? Bengie, who’s always been a little better than average, had a great season despite adverse expectations, and jumped to the head of the list ahead of his brothers. Jason Kendall has been on a roller coaster, being average, average, poor, average, poor and very good the past six years. Despite the fluctations, averaging all that together, he projects as league average for next year. Most of the values are consistent from year to year.

Henry Blanco is the “career” leader of the six year period, with a normal CS% of .445, and a runs allowed above average (RAA) of 10.1 per 1800 base stealing opportunities. He’s followed by Yadier Molina at .442 and Gerald Laird at .406. The worst rates are Gary Bennett .139, Michael Barrett .152 and Mike Piazza .160, with Piazza having the worst R/1800 at -12.1. At age 36, Blanco has shown no signs of slowing down, having a normal CS% over .500 in 3 of the past 4 seasons.

On the other hand, Jason Varitek and Brad Ausmus, two catchers with a past record of defensive accolades, have both shown sharp downward trends. Varitek’s normal CS% has been .419, .255, .211, .160, .139 and .118, dropping from very good to very bad. Ausmus similarly has marks of .308, .222, .180, .169, .103 and .136.

Despite four consecutive years of poor throwing, the rate of steal attempts versus Ausmus has been at or below average, likely based on his past reputation and the ability of his pitchers to hold runners. Ivan Rodriguez, who in the 1990’s routinely had observed CS% of over .500, has the past two seasons only been slightly better than average at .273 and .275. Although the rate attempt rate against Rodriguez has risen to .040 (average=.047), over the last six years he has the lowest rate at .029, followed by Rod Barajas at .036, Joe Maurer .037, Toby Hall .038, Chris Snyder .038 and Ausmus .040. The catchers run against most often have been Mike Piazza .074, Brandon Inge .066, Victor Martinez .060, Paul LoDuca .058 and Michael Barrett .055.

2008 Leaders

Name RAA R1800 nCS% oSB oCS oCS% eSB eCS eCS%
Molina, Bengie 13.0 11.3 0.420 68 31 0.313 106 20 0.160
Kendall, Jason 11.6 9.3 0.419 55 36 0.396 89 26 0.224
Molina, Jose 8.7 12.8 0.408 42 32 0.432 61 22 0.259
Navarro, Dioner 3.4 3.7 0.367 45 23 0.338 53 19 0.263
Molina, Yadier 5.8 5.9 0.364 34 16 0.320 47 12 0.205
Martin, Russell 3.9 3.5 0.325 68 17 0.200 86 15 0.149
Snyder, Chris 4.5 5.2 0.324 49 19 0.279 63 15 0.198
Schneider, Brian 2.7 3.1 0.312 42 16 0.276 54 15 0.217
Suzuki, Kurt 4.4 3.9 0.303 53 18 0.254 82 22 0.214
Mauer, Joe 4.4 4.0 0.301 51 19 0.271 67 18 0.216
Laird, Gerald 0.6 0.8 0.293 52 18 0.257 55 14 0.207
Johjima, Kenji -0.6 -0.8 0.290 51 19 0.271 43 12 0.223
Rodriguez, Ivan 3.2 3.4 0.275 51 22 0.301 64 23 0.262
Torrealba, Yorvit -1.5 -2.8 0.224 45 10 0.182 34 9 0.198
Bako, Paul -6.0 -7.7 0.217 55 20 0.267 40 28 0.409
Pierzynski, A.J. 0.3 0.3 0.212 96 11 0.103 106 14 0.119
Hernandez, R. -4.5 -4.1 0.207 98 18 0.155 75 17 0.181
Doumit, Ryan -4.2 -4.3 0.202 68 15 0.181 61 17 0.221
Ruiz, Carlos -1.6 -2.0 0.202 65 14 0.177 50 14 0.214
Soto, Geovany -1.9 -1.8 0.198 67 17 0.202 63 20 0.240
McCann, Brian -6.1 -5.7 0.189 93 21 0.184 72 21 0.227
Treanor, Matt -0.8 -1.5 0.186 44 11 0.200 31 10 0.241
Mathis, Jeff -6.7 -9.7 0.175 57 16 0.219 44 20 0.306
Napoli, Mike -4.6 -7.5 0.131 52 9 0.148 42 15 0.263
Varitek, Jason -8.2 -8.6 0.118 56 14 0.200 56 33 0.372
Buck, John -8.1 -8.9 0.098 59 9 0.132 51 24 0.322

2003-2008 Leaders

Name RAA R1800 nCS% oSB oCS oCS% eSB eCS eCS%
Blanco, Henry 24.6 10.1 0.4 128 79 0.382 177 47 0.209
Molina, Yadier 38.0 9.7 0.4 115 89 0.436 190 53 0.218
Laird, Gerald 22.9 7.7 0.4 165 93 0.360 212 55 0.206
Molina, Jose 27.7 9.4 0.4 160 104 0.394 205 69 0.252
Martin, Russell 17.6 5.4 0.4 219 75 0.255 256 49 0.160
Schneider, Brian 36.6 6.5 0.4 259 134 0.341 349 96 0.216
Johjima, Kenji 12.6 4.1 0.3 154 72 0.319 174 49 0.218
Rodriguez, Ivan 44.9 7.5 0.3 236 119 0.335 455 140 0.235
Mauer, Joe 17.0 4.5 0.3 150 78 0.342 225 74 0.246
Hall, Toby 24.1 5.4 0.3 241 94 0.281 335 84 0.200
Ross, Dave 9.9 3.5 0.3 140 73 0.343 172 60 0.258
Olivo, Miguel 20.5 4.4 0.3 214 95 0.307 278 87 0.239
Snyder, Chris 15.9 4.8 0.3 181 67 0.270 241 64 0.211
Barajas, Rod 15.8 3.7 0.3 203 90 0.307 297 95 0.243
LaRue, Jason 5.7 1.4 0.3 198 84 0.298 224 71 0.242
Molina, Bengie 23.5 4.3 0.3 328 120 0.268 416 110 0.210
Inge, Brandon 5.1 2.9 0.3 125 56 0.309 95 33 0.257
Miller, Damian 8.4 2.1 0.3 200 76 0.275 237 74 0.238
Navarro, Dioner -1.3 -0.4 0.3 196 69 0.260 183 64 0.261
Matheny, Mike 5.4 1.6 0.3 165 64 0.279 184 64 0.258
Hernandez, R. -0.9 -0.1 0.3 397 123 0.237 375 109 0.225
Paulino, Ronny 3.7 1.5 0.3 165 51 0.236 158 42 0.210
Torrealba, Yorvit 2.4 0.7 0.3 216 71 0.247 207 61 0.228
Posada, Jorge 5.1 0.9 0.3 428 147 0.256 441 141 0.242
Redmond, Mike -1.0 -0.4 0.3 146 45 0.236 126 41 0.243
Moeller, Chad -0.6 -0.2 0.2 174 44 0.202 160 41 0.205
Bako, Paul 0.6 0.2 0.2 166 59 0.262 166 65 0.282
Lo Duca, Paul -8.5 -1.7 0.2 439 136 0.237 359 106 0.228
Treanor, Matt 0.0 0.0 0.2 139 40 0.223 124 34 0.217
Kendall, Jason -6.2 -0.9 0.2 465 133 0.222 447 126 0.220
McCann, Brian -12.5 -3.6 0.2 254 59 0.188 222 71 0.243
Martinez, Victor -13.5 -2.7 0.2 396 120 0.233 298 108 0.266
Lieberthal, Mike -7.0 -1.9 0.2 264 70 0.210 245 75 0.234
Pierzynski, A.J. -2.4 -0.4 0.2 422 96 0.185 446 114 0.204
Ruiz, Carlos 1.3 0.7 0.2 133 33 0.199 120 33 0.217
Napoli, Mike -3.9 -2.1 0.2 139 39 0.219 128 42 0.245
Doumit, Ryan -4.7 -2.7 0.2 121 30 0.199 107 31 0.225
Varitek, Jason -7.3 -1.3 0.2 366 97 0.210 394 128 0.246
Valentin, Javier -4.7 -2.2 0.2 119 36 0.232 112 41 0.267
Johnson, Charles -0.9 -0.5 0.2 112 26 0.188 114 30 0.208
Zaun, Gregg -11.1 -2.8 0.2 323 77 0.193 303 87 0.222
Ausmus, Brad -4.1 -0.8 0.2 310 89 0.223 349 129 0.270
Bard, Josh -10.9 -4.0 0.2 279 47 0.144 229 54 0.190
Lopez, Javy -7.8 -2.6 0.2 203 51 0.201 193 69 0.264
Buck, John -11.0 -2.5 0.2 225 63 0.219 240 103 0.299
Estrada, Johnny -16.1 -4.1 0.2 285 70 0.197 238 93 0.280
Phillips, Jason -12.4 -6.4 0.2 177 32 0.153 140 43 0.236
Piazza, Mike -27.5 -12.1 0.2 272 47 0.147 153 47 0.236
Barrett, Michael -29.2 -6.7 0.2 362 73 0.168 269 101 0.272
Bennett, Gary -10.8 -4.2 0.1 156 28 0.152 160 52 0.247

2009 Projections

Name Age RAA R1800 pAtt% pCS% Size pSB pCS pPK
Molina, Yadier 27 10.2 10.4 0.042 0.413 4695 43 30 6
Laird, Gerald 29 5.0 6.4 0.046 0.375 3848 41 24 1
Hundley, Nick 25 4.4 8.8 0.037 0.374 1297 21 12 2
Molina, Jose 34 5.4 7.9 0.046 0.367 3313 36 21 3
Martin, Russell 26 7.3 6.5 0.046 0.358 4755 59 33 2
Chavez, Raul 36 2.3 7.8 0.048 0.352 1327 16 9 1
Schneider, Brian 32 4.9 5.6 0.043 0.338 5431 45 23 1
Olivo, Miguel 31 2.9 6.3 0.043 0.326 4438 24 12 2
Johjima, Kenji 33 3.3 4.0 0.049 0.324 4316 50 24 1
Mauer, Joe 26 5.6 5.0 0.039 0.323 4689 53 25 1
Snyder, Chris 28 4.4 5.0 0.042 0.315 4055 45 21 1
Cash, Kevin 31 1.4 3.8 0.049 0.313 1491 23 10 0
Ross, Dave 32 1.4 3.4 0.042 0.313 3145 22 10 1
Barajas, Rod 33 3.0 4.4 0.038 0.312 3933 32 15 1
Molina, Bengie 35 4.9 4.2 0.040 0.310 5592 57 25 2
Rodriguez, Ivan 37 5.1 5.4 0.033 0.306 5588 38 17 3
LaRue, Jason 35 0.5 1.4 0.044 0.294 3343 20 8 0
Quintero, Humberto 29 2.0 4.9 0.047 0.290 1538 24 10 1
Navarro, Dioner 25 2.2 2.4 0.052 0.279 4164 62 24 2
Suzuki, Kurt 25 3.6 3.2 0.037 0.277 3124 55 21 0
Inge, Brandon 32 1.1 2.3 0.054 0.267 1729 36 13 1
Paulino, Ronny 28 0.3 1.2 0.053 0.256 3242 21 7 0
Hernandez, R. 33 -0.1 -0.1 0.053 0.251 5662 78 26 1
Kendall, Jason 35 -0.5 -0.4 0.051 0.247 6670 86 28 0
Torrealba, Yorvit 31 -0.2 -0.4 0.052 0.237 3547 39 12 0
Flores, Jesus 24 0.0 0.0 0.047 0.234 2086 44 14 0
Martinez, Victor 30 -0.5 -1.2 0.061 0.232 4560 34 10 0
Coste, Chris 36 0.2 0.3 0.049 0.232 2126 40 12 1
Shoppach, Kelly 29 0.9 1.1 0.037 0.231 2688 43 13 1
Nieves, Wil 31 -0.7 -1.6 0.060 0.229 1414 36 11 1
Bako, Paul 37 0.6 0.8 0.047 0.228 3044 51 15 2
Pierzynski, A.J. 32 -0.1 -0.1 0.043 0.227 5849 63 19 0
Towles, J.R. 25 0.5 1.2 0.043 0.223 1195 23 7 2
Quiroz, Guillermo 27 -0.1 -0.4 0.050 0.223 1249 28 8 0
Ruiz, Carlos 30 0.4 0.5 0.052 0.219 3109 57 16 3
McCann, Brian 25 -2.0 -1.8 0.051 0.218 4756 78 22 2
Treanor, Matt 33 -0.5 -1.0 0.054 0.217 2582 39 11 2
Soto, Geovany 26 -1.2 -1.2 0.049 0.214 2455 73 20 0
Iannetta, Chris 26 1.5 1.8 0.037 0.214 2684 44 12 1
Doumit, Ryan 28 -2.5 -2.5 0.050 0.211 2837 70 19 1
Zaun, Gregg 38 -1.5 -2.7 0.049 0.206 3916 39 10 1
Napoli, Mike 27 -1.4 -2.3 0.051 0.202 2787 45 11 0
Mathis, Jeff 26 -3.1 -4.5 0.054 0.176 2370 55 12 5
Varitek, Jason 37 -2.8 -2.9 0.044 0.176 5425 63 13 1
Bard, Josh 31 -1.7 -4.2 0.056 0.174 3045 35 7 1
Ausmus, Brad 40 -1.0 -1.9 0.042 0.168 4614 33 7 1
Saltalamacchia, Jarrod 24 -2.6 -5.6 0.052 0.164 1728 36 7 1
Buck, John 29 -2.3 -2.5 0.041 0.163 4966 57 11 1
Montero, Miguel 26 -1.4 -3.8 0.051 0.146 1762 29 5 1
Baker, John 28 -2.2 -4.6 0.051 0.138 1262 38 6 0
Riggans, Shawn 29 -1.9 -6.2 0.048 0.128 1059 24 3 0



Print This Post



Brian got his start in amateur baseball, as the statistician for his local college summer league in Johnstown, Pa, which also hosts the annual All-American Amateur Baseball Association. A longtime APBA and Strat-o-Matic player, he still tends to look at everything as a simulation. He has also written for StatSpeak and SeamHeads You can contact him at brian.cartwright2@verizon.net


18 Responses to “Stealing on Catchers (Not Named Molina)”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. azruavatar says:

    Brian —

    Am I to read R1800 as the rough equivalent of a 162 game season or do you already prorate a catchers’ season since they rarely catch 162 games?

    Thanks.

    Vote -1 Vote +1

  2. David Pinto says:

    Brian,

    Good stuff, but you need to learn HTML tables. The data is unreadable the way you’ve presented it.

    Vote -1 Vote +1

  3. Peter Jensen says:

    I am skeptical of the values for eCS%. There is almost a 4 to 1 range which seems way to large for a group of pitchers. I warned Tango that using WOWY without doing several iterations might result in some erroneous conclusions. I have a feeling that the large range in eCS% may be an example of this. Also, could you please identify how you have changed the Odds Ratio method to produce “the Inverse James function”?

    Vote -1 Vote +1

  4. Peter Jensen says:

    Also – Could you tell us how you defined a base stealing opportunity?

    Vote -1 Vote +1

  5. Brian Cartwright says:

    Base stealing opportunity is runner on 1st and no runner on 2nd, or runner on 2nd and no runner on 3rd. Fpr catchers, attempts (SB or CS) do not include when pickoff flag is true – those are counted seperately.

    1800 attempts was a number I picked that represented a typical full season, but short of 162 games. A handful of catchers got up to 2200 in a season. It gives a method of scaling the performance as a rate, and showing what how much the best and worst performances vary over a full seaosn in terms of runs (and thus wins). Switching from Yadier to Piazza would cost a team 2 – 2.5 wins in opponent’s base running, something to be considered when valuing defense.

    Peter, what you say about the eCS% makes sense, and I did not think about iteration, something to be considered. I haven’t done a “normal” run for pitchers, but looking at their observed values it looks like there is a much greater variance in CS% among pitchers than catchers, who appear to be between .1 and .5. The eCS% for 2003-2008 ranges from Russell Martin at .160 to John Buck at .299

    I think there could be distortions when a good and bad catcher are paired on the same team, such as Nick Hundley and Josh Bard. I think Hundley may not be as good his one partial year. Another year and new pitchers on the staff should help.

    Dave Pinto, I have plenty of experience writing scripts that generate html tables from csv files, but Mr Appelman says to (at the moment) use blocks of text. I previewed it in IE before I posted.

    Vote -1 Vote +1

  6. Peter Jensen says:

    Brian – I am not surprised that there is a larger variation in CS% for pitchers than catchers, but I am surprised that the variation for an entire pitching staff would be as large as you report. The left handers and right handers and fastballers and off speed pitchers should cancel each other somewhat out, I would think. The way you define Base Stealing opportunities seems fine, but they are not all created equally, of course. One would hope that each catcher would have about the same proportion of opportunities by base out state, lineup position, etc., but it might be worth checking to see if you need to figure the steal % for some of the factors separately. How about the formula for your “inverse James function”?

    Vote -1 Vote +1

  7. Brian Cartwright says:

    Inverse James Function

    Sorry, it should be log5 instead of odds ratio, but they are very similar
    http://www.diamond-mind.com/articles/playoff2002.htm

    After Bill James published it originally in the Abstract, which tells you what pct a given combination should produce. I solved for one of the other variables in the equation.

    N = (O*M*(1-E))/(O*M+(E*(1-O-M))) (all being binomials)

    Vote -1 Vote +1

  8. Xeifrank says:

    I wonder how these (and others?) results would line up vs taking a stop watch and determining on average how long it took for a the ball to go from the catchers glove to the fielders glove and 2nd/3rd base. Controlling for pitch outs and perhaps really bad pitches and the like. You could even throw in a measurement for accuracy of throw. There seems like too many factors that are out of the catchers control and that are not consistent that could make a pbp study difficult to do.
    vr, Xeifrank

    Vote -1 Vote +1

    • Brian Cartwright says:

      I believe this is something scouts do, as they are looking at amateur players, and want an objective measure of ability.

      But, with the pros, what we have is pbp, and so I’m working on tools to derive the most I can from that pbp.

      Vote -1 Vote +1

  9. Peter Jensen says:

    Brian – Log 5 works less well the farther percentages get from .5. Thats why it is OK to use for team winning % where the mean is .5 and the range is relatively small, but really shouldn’t be used for other analysis.

    Vote -1 Vote +1

  10. Xeifrank says:

    Looks like you have some fairly decent agreement with Tango’s Fan Scouting reports for 2008. The fans like Y.Molina the best, followed by J.Mauer, J.Molina and I.Rodriguez. B.Molina probably drops down to the bottom half of the top 10. The fans don’t think much of R.Martin’s accuracy, and wouldn’t make the fans top 10 at all. Nice work!
    vr, Xeifrank

    Vote -1 Vote +1

    • Brian Cartwright says:

      Thanks. My goal in writing these pbp tools is to get them to agree well with what others have done for major leaguers, so that I can then apply them to minor leaguers with a known degree of confidence.

      Vote -1 Vote +1

  11. Sam says:

    Add to win values?

    Vote -1 Vote +1

  12. Nathan says:

    Why would the type of turf matter? The baserunner would be running on dirt and the catcher would be throwing from dirt no matter the type of grass.

    Vote -1 Vote +1

  13. MGL says:

    Brian, great work! Have you done the same thing for pitchers?

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>