Thoughts from Diamond Mind

Welcome to the Diamond Mind weblog. For a while, we've been looking for the best way to publish small pieces of baseball commentary and research, items that may not warrant a full article (such as the ones we've been writing for ESPN.com since 1998) or items that would be outdated by the time our next email newsletter is due to be issued. A weblog seems perfectly suited to this purpose, and we hope you enjoy what you read here.

To provide feedback, email us at blog@diamond-mind.com. We won't be able to reply to those messages individually, but we'll read them all, and we may respond by posting a reply in the weblog itself.


Blog Archives

Weblog Home

This page is powered by Blogger.

 		Wednesday, August 18, 2004

Ball-strike dynamics
 
In June, 2001, we wrote a newsletter essay about a misleading Sports Illustrated article on the subject of batting performance on different ball-strike counts. Someone asked me about that essay recently, and that made me think it might be interesting to revisit this subject with some updated stats.

The main source of confusion in that SI article was the difference between stats when the ball is put in play ON a certain count and stats when the battle between batter and pitcher passes THROUGH a certain count.

To illustrate the difference, let's go over the three things that can happen on the first pitch:

1. The batter concludes the plate appearance by putting the ball in play or getting hit by the pitch.

2. The batter takes ball one. This puts him a little closer to a walk, reduces the ability of the pitcher to go outside the strike zone on subsequent pitches, and puts the batter in a position where he be more selective, looking for a pitch he can drive.

3. The batter falls behind the pitcher. Now the batter is a step closer to striking out, the pitcher has a little more freedom to go outside the strike zone in hopes that the batter will chase a bad pitch, and the hitter has to be ready for anything the pitcher might throw at him.

If you use the "at 0-0" data, you lose most of the effects of 2 and 3 above. The averages are distorted because all of the strikeouts are loaded into the counts with two strikes and all of the walks are loaded into the counts with three balls. It's impossible to walk or strike out on the first pitch, so batting averages are higher and on-base percentages are lower than on other counts.

If, on the other hand, you use the "through 0-0" data, you see the effects of all three of these forces at work. The incremental steps toward walks and strikeouts are reflected in the stats for each count, as is the relative advantage for the batter or pitcher when a ball is put in play.

To illustrate this point, here are the 2003 totals for both leagues with pitcher hitting removed:

                AVG   OBP   SPC     AB     H   2B   3B   HR   RBI    BB  IBB     K

At 0-0         .341  .348  .555  21312  7263 1455  158  932  3790     0    0     0
At 0-1         .321  .330  .487  16002  5139  981   93  496  2328     0    0     0
At 0-2         .167  .179  .243  13198  2209  421   55  156   949     0    0  5770


At 1-0         .342  .344  .574  13604  4659 1001  104  646  2402     0    0     0
At 1-1         .328  .333  .522  15772  5175 1036  106  603  2474     0    0     0
At 1-2         .177  .186  .260  23762  4212  724   82  360  1776     0    0  9755


At 2-0         .343  .343  .622   4882  1673  379   31  307  1000     0    0     0
At 2-1         .332  .334  .551  10118  3361  723   73  448  1652     0    0     0
At 2-2         .195  .200  .308  22761  4434  891  104  489  2084     0    0  8590


At 3-0         .401  .960  .789    279   112   29    2   25   159  4077 1243     0
At 3-1         .347  .680  .604   4403  1526  318   33  250   963  4677   67     0
At 3-2         .230  .470  .380  15263  3506  753   77  462  1983  6923    3  4752


Through 0-0    .268  .337  .430 161297 43249 8707  916 5173 21534 15651 1310 28851
Through 0-1    .239  .283  .372  77513 18524 3622  392 1956  8583  4033   35 19273
Through 0-2    .183  .215  .275  29598  5410 1001  114  500  2452   918    2 12009


Through 1-0    .280  .394  .459  62496 17474 3632  366 2286  9175 11622 1277  9579
Through 1-1    .248  .319  .396  62642 15529 3116  350 1815  7506  6170   64 14356
Through 1-2    .191  .243  .294  42796  8182 1532  172  846  3789  2596    2 16608


Through 2-0    .290  .516  .493  18181  5265 1155   98  785  3076  8580 1247  2497
Through 2-1    .262  .398  .433  30609  8030 1682  183 1061  4193  6873   68  6034
Through 2-2    .207  .307  .333  33128  6850 1404  155  827  3468  4657    3 11827


Through 3-0    .296  .760  .496   3170   937  222   16  127   634  6204 1243   452
Through 3-1    .281  .586  .475   9295  2615  557   59  374  1562  6940   67  1514
Through 3-2    .230  .470  .380  15263  3506  753   77  462  1983  6923    3  4752


For example, the "through 0-0" data shows the results of all plate appearances, since everybody starts with a 0-0 count. At this point in the sequence, the batter is almost twice as likely to strike out as to walk.

If, however, the first pitch is a ball, the batter has a slightly better chance to walk than to strike out. But starting with strike one makes him almost five times more likely to strike out than to walk. In round numbers, that first pitch shifted the K:BB ratio from 2:1 to either 1:1 (ball one) or 5:1 (strike one). That's a big deal.

And batters who reached 1-0 had a batting average 51 points higher than batters who started with strike one, with even larger gains in slugging.

Now, if you'll forgive me for a blatant Diamond Mind Baseball plug ... the ebb and flow of the batter-pitcher confrontation is one of the reasons we rebuilt our game engine seven years ago to simulate every pitch. In DMB games, just as in real life, the advantage swings between batter and pitcher as the count changes.

Another big advantage of pitch mode is the ability to change tactics on every pitch, thus creating an interesting cat-and-mouse game between the two managers. Worried about a possible suicide squeeze? Call for a pitch out, but understand that this will move the count in the batter's favor. Worried about a steal attempt? Make a pickoff throw. Want to steal but are afraid of a pitch out? Give the hitter the take sign and see if you can get ahead in the count.

Last but not least, pitch mode can significantly increase drama and tension. In batter mode, the outcome of a play is revealed immediately. But in pitch mode, the outcome can hang in the balance over an extended sequence of pitches. That suspension of time in a critical situation is one of the great things about baseball,
and we wanted to give you the option to experience the same thing in your DMB games.

Of course, veteran DMB gamers know that this doesn't mean you must play your DMB games in pitch mode. You can toggle between pitch mode and batter mode at any time. In batter mode, the game simulates every pitch, so you don't lose any realism, but it speeds game play by generating the play-by-play commentary only when the ball is put in play or some other notable event takes place.

To get the best of both worlds, many people play the majority of the time in batter mode and switch into pitch mode for big games, big moments in other games, and situations that are rife with tactical options. posted by Tom at 10:07 AM