Is Jorge Posada toast as a righty hitter?

In 2011, when Jorge Posada hit right-handed, it was worse than having a pitcher hit. As a righty, Jorge had a BA/OBP/SLG of .092/.169/.108 over 71 plate appearances; pitchers in 2011 had a line of .141/.175/.153.

Posada did not get a hit batting right-handed until June 7, starting the season 0-for-27. Thus, since last May, Yankee beat writers and other local media have been constantly citing Posada’s bad righty splits. The consensus of such observers for a while has been that Posada is done as a right-handed hitter, as was expressed, for example, by Professional Baseball Player Career Adviser Bob Klapisch.

Of course, if you are reading The Hardball Times, you do not need me to prove that Posada’s 2011 righty splits, amounting to about 2.5 weeks of plate appearances, alone are not enough to assert anything with great confidence. Rather, what happened is that Mr. Klapisch’s article annoyed me. It was the straw that broke the camel’s back in terms of incessantly coming across articles about Posada’s $#@%$#@ splits.

So for some winter fun, I used the Retrosheet data to explore the implications of confidently asserting judgments based on 71 plate appearances. The first thing I wanted to do was quantify the degree of confidence one can in general place in 71 trips to the plate. One quick method is to calculate the range of OBP an average 2011 hitter could be expected to have in 71 plate appearances.

In 2011, non-pitchers averaged an OBP of .325. The R formula qbinom(c(.1,.9),71,.325)/71 indicates that with 90 percent confidence, an average hitter would have an OBP in the range of .253-.394. In other words, after 71 plate appearances, an average hitter’s OBP could be anywhere between Miguel Olivo’s and Ryan Braun’s, and we’d really have no way of knowing the difference.

That being said, Posada’s OBP was well below .253, the low end of that range. The binomial distribution probability indicates that the likelihood of an average hitter having an OBP of .169 after 71 plate appearances is approximately .0025.

At first glance, those odds—one out of 400—may make it seem that Posada’s righty skills indeed have gone the way of all flesh. However, these are hardly astronomical odds in a baseball context; thousands of streaks of 71 consecutive plate appearances occur each season, and I would expect such ugly streaks occasionally happen to average-and-above players.

But does my expectation here reflect reality? To study that, I wrote a program that pulls from the Retrosheet data the stats batters compiled in streaks of 71 consecutive plate appearances; for example, a batter’s first “streak” would span from his first to 71st appearances of the season, his second streak would span his second through 72nd, etc. The results will reveal whether average and above-average hitters experience such horrible streaks.

For 2011, excluding pitchers, there were 485 players who had 71 or more plate appearances; these players had a total of 140,813 streaks that consisted of 71 plate appearances. A total of 685 of those streaks resulted in an OBP less than or equal to Posada’s right-handed OBP of .169.

Of course, not all the players in this latter set have at least average OBP skills, so to quickly identify whether players with at least average OBP skills had such bad streaks, I pulled the players with at least 400 plate appearances in both 2010 and 2011. This will not eliminate all players with below-average OBP skills—for example, the results will include players kept in the lineup for real/perceived defensive skills, players on lousy teams, and players with offensive skills other than OBP.

But for this purpose, it is a non-issue if some or even most of the players in the query results are crappy hitters. All we are looking for here is whether a few average-or-above players had such poor streaks. The following table lists the players, their OBPs for the last two seasons, and their lowest OBP in a set of 71 consecutive plate appearances.

Player             2011 - worst OBP, 71 PA  2011 OBP   2010 OBP
Aybar, Erick                .127               .317      .306
Beckham, Gordon             .129               .293      .317
Betancourt, Yuniesk         .141               .271      .288
Dunn, Adam                  .155               .292      .356
Ellis, Mark                 .116               .283      .358
Lind, Adam                  .155               .295      .287
Loney, James                .157               .337      .329
McGehee, Casey              .169               .280      .337
Olivo, Miguel               .155               .252      .315
Raburn, Ryan                .130               .294      .340
Rasmus, Colby               .155               .297      .361
Rios, Alexis                .169               .265      .334
Soriano, Alfonso            .141               .289      .322
Suzuki, Ichiro              .155               .309      .359
Theriot, Ryan               .169               .317      .321
Uggla, Dan                  .127               .311      .369
Wells, Vernon               .155               .248      .331
TOTAL                        na                .293      .332

It is fair to say, I think, that at minimum, Ichiro, Loney, and Uggla can be considered to have at least average OBP skills, despite their horrific 71-PA streaks within 2011. You could make arguments for or against others as well, based on various criteria, but again, I am merely looking for evidence that such streaks happen to a few at-least-average players.

In terms of overall offense, the following table shows some putrid 71-PA streaks in 2011 for players with arguably average-or-above skills.

Player                  BA    OBP    SLG   wOBA
Uggla, Dan            .090   .127   .104   .111
Raburn, Ryan          .104   .130   .119   .114
Aybar, Erick          .101   .127   .130   .119
Loney, James          .109   .157   .125   .124
Suzuki, Ichiro        .118   .155   .162   .126
Smoak, Justin         .079   .155   .111   .134
McGehee, Casey        .123   .169   .123   .142

As expected, average-or-above players do experience 71-PA streaks during which they hit worse than pitchers. But for further validation, I wanted to see whether such hideous streaks happen to proven good players. So I ran the program for all players who have accumulated 6,000 at-bats and have played at some point since Posada’s career began. The following table lists selected horrific streaks of name-brand players.

Player               Year     BA    OBP    SLG   wOBA
Konerko, Paul        2003   .079   .169   .079   .124
Suzuki, Ichiro       2011   .119   .157   .164   .126
Beltran, Carlos      2009   .116   .141   .159   .126
Rollins, Jimmy       2005   .104   .155   .104   .129
Helton, Todd         2005   .085   .229   .102   .139
Sosa, Sammy          1991   .116   .129   .203   .141
Bonds, Barry         1988   .136   .183   .167   .144
Suzuki, Ichiro       2003   .118   .155   .162   .146
Jeter, Derek         2004   .123   .174   .185   .151
Gonzalez, Luis       1992   .092   .155   .169   .152
Delgado, Carlos      2008   .098   .211   .115   .152
Ibanez, Raul         2002   .134   .169   .194   .154
Piazza, Mike         2001   .132   .169   .191   .154
Ortiz, David         2009   .118   .155   .221   .166
Guerrero, Vladimir   2006   .155   .155   .268   .178
Guerrero, Vladimir   2005   .134   .169   .284   .184
Rodriguez, Alex      1999   .119   .169   .299   .201

The data here speaks for themselves, but I do have a couple of comments.

First, to put the 1-in-400 odds that led to this exercise in context, let’s look at Alex Rodriguez’s 1999 streak. A-Rod’s on-base percentage in 1999 was .357; the previous year’s was .360 and the following year’s .420. So it is reasonable to think that his actual on-base skills were something in the range of .357-.420. Using the lowest end of the range, .357, as the probability, the binomial distribution indicates the odds of A-Rod performing as poorly as he actually did were 1 in 2,500.

Second, as a Yankee fan, I wish ESPN writer Bill Simmons had been the Red Sox GM during Big Papi’s 2009 slump, which included the above 71 plate appearances.

Circling back to Posada, the 1-in-400 odds clearly do not mean his 71 plate appearances are predictive; his hideous 71 righty plate appearances certainly could be due to a collapse in skills but also to random fluctuation or bad luck.

Let’s look at luck. Posada’s BABIP batting right-handed was .146, suggesting bad luck may have been a factor. For fun, let’s play the role of the Baseball Gods and give Posada six more hits to get him to a typical BABIP of around .300; with that, his OBP would be .254.

Coincidence or not, .254 happens to be the bottom of the expected range that pbinom() calculates as the expected OBP of an average player. Does this prove that luck was behind Jorge’s abominable righty hitting? Of course not; the .146 BABIP could reflect a washed-up hitter capable of only weak contact; it merely suggests the possibility of horrendous luck.

Since Posada’s righty 2011 stats alone are not useful to gauge his skills, what I next checked were splits from the last few years. Was Posada showing signs of a righty-hitting collapse before 2011? The following table lists Posada’s righty splits over the last few years, along with his career righty splits.

Year  Hand    PAs     BA  BABIP    OBP    SLG    wOBA
2009   R      139   .290   .397   .360   .476    .366
2010   R      153   .257   .300   .340   .493    .360
2011   R       71   .092   .146   .169   .108    .132
Career R     2052   .288   .341   .371   .482    .375

In 2009-2010, Posada’s right-handed hitting, as measured by wOBA, was slightly below his career average but still very good. However, the bulk of his career was in a higher-scoring era than 2010. In terms of park factor, the new Yankee Stadium has been more of a hitter’s park than the old stadium—however, Yankee Stadiums old and new favor lefty hitters, so it also would be useful to look at adjusted stats with and without park factors.

The following table shows both actual and adjusted wOBA for Posada’s right-handed hitting, adjusted against the AL 2011 environment of 4.46 runs-per-game.

                                                    Adjusted wOBA
 Year    R/G   Park Factor    PAs    wOBA     (Park Factor) (No Park Factor)
 1996   5.39      1.01          5    .180        .165           .165
 1997   4.93      1.00         43    .350        .337           .337
 1998   5.01      0.97        122    .421        .408           .403
 1999   5.18      0.98        106    .367        .349           .346
 2000   5.30      0.99        186    .404        .381           .380
 2001   4.86      1.02        175    .375        .362           .365
 2002   4.81      1.00        157    .419        .408           .408
 2003   4.86      0.98        144    .436        .428           .425
 2004   5.01      0.98        169    .385        .371           .368
 2005   4.76      0.98        195    .372        .367           .364
 2006   4.97      1.01        158    .356        .343           .344
 2007   4.90      1.01        168    .396        .382           .383
 2008   4.78      1.03         61    .308        .297           .300
 2009   4.82      1.05        139    .366        .350           .356
 2010   4.45      1.05        153    .360        .354           .361
 2011   4.46      1.08         71    .132        .127           .132
Total    NA        NA        2052    .375        .372           .372

This perspective suggests Posada’s 2011 righty performance was not part of a trend but rather either an outlier or a falling off of the cliff.

In terms of looking for decline patterns, it’s worth also checking his lefty-hitting over the years.

                                                  Adjusted wOBA 
 Year    R/G   Park Factor    PAs    wOBA      (Park Factor)  (No Park Factor)
 1996   5.39      1.01         10    .072           .072           .072
 1997   4.93      1.00        181    .342           .330           .330
 1998   5.01      0.97        287    .318           .308           .304
 1999   5.18      0.98        331    .317           .302           .300
 2000   5.30      0.99        438    .404           .381           .380
 2001   4.86      1.02        382    .355           .341           .344
 2002   4.81      1.00        441    .343           .334           .334
 2003   4.86      0.98        444    .394           .385           .382
 2004   5.01      0.98        378    .395           .382           .380
 2005   4.76      0.98        351    .341           .335           .332
 2006   4.97      1.01        386    .398           .381           .382
 2007   4.90      1.01        421    .427           .412           .413
 2008   4.78      1.03        134    .347           .335           .338
 2009   4.82      1.05        299    .387           .369           .376
 2010   4.45      1.05        298    .353           .347           .353
 2011   4.46      1.08        316    .362           .352           .362
Total    NA        NA        5097    .369           .363           .364

Posada’s .352 lefty wOBA in 2011 would have been a positive at any position, the point being that there is nothing to suggest a general physical breakdown in 2011. Of course, it is possible that some physical issue impacted some part of Posada’s body that only affects his righty hitting.

Finally, apart from statistics, one might argue that the Yankees only giving Posada 71 plate appearances as a righty, while giving him a representative number of lefty plate appearances, suggests manager Joe Girardi considered him washed up as a righty. But as someone who pays too much attention to the Yankees, I suggest that this playing pattern more likely reflects an accident of roster construction and injuries.

What I mean is that Posada lost his job as righty DH in mid-May after 30 righty plate appearances with zero hits and six walks. But he was slumping from both sides of the plate at the time, and I suspect that if Eric Chavez was not injured at the time, Posada would have been benched from both sides of the plate. After all, that is what Girardi did in August; he benched Posada for awhile when he again fell into a slump and used Chavez as the lefty DH (or sometimes Chavez at third base and A-Rod at DH).

Posada later played himself back into the lineup as the lefty DH for the stretch run and playoffs, showing that Girardi was inclined to periodically change lineups in order to play the hot, and sit the cool, hand. Andruw Jones, and later Jesus Montero, hit so well as righty DH options that Girardi never had a reason to give Posada another shot in that role.

Given Posada’s rumored retirement, we may never get a chance to see whether his 2011 righty splits were predictive or mere noise. However, a careful reading of the details suggests the rumors are conditional—according to the New York Times, Posada “told teammates that he plans to retire rather than attempt to play for another team at 40.” The trading of Montero may open a door for Posada, because the trade opens up the DH slot.

It seems likely the Yankees would prefer someone willing to be a part-time DH so they can give a fair amount of DH time to A-Rod and other starters for the purposes of occasional rest. Will some of the rumored candidates such as Carlos Pena be willing to play such a limited role? It’s conceivable that Posada may emerge as the best available candidate for that type of role.

In weighing the Posada option, I would expect Brian Cashman to evaluate factors such as Posada’s age and the degree to which Posada’s ability to play a decent first base and be an emergency catcher provides Girardi with roster flexibility when compared with other options. If this is more than mere idle speculation on my part, I would also expect Cashman and Girardi to view Posada’s switch-hitting as a potential benefit, not to assume that 71 horrible plate appearances mean he is done as a righty.