Short work

We’ve all heard this bit of conventional baseball wisdom, quoted here from an interesting article on “Little things that win big ball games” by college pitching coach Jim Mason:

There is a direct relationship between the quickness with which a pitcher works and his effectiveness…. The pitcher who works slowly will often cripple his defense. Baseball demands a high level of concentration, and this is very difficult to do behind a pitcher who works at a slow pace. The mind tends to wander behind deliberate workers, such as pitchers. Coaches should try to speed up such workers.

Last May, Dan Fox took a look at the adage in an article at Baseball Prospectus. Dan was restricted to using the best guesses of the BP writers as to the fastest- and slowest-working pitchers, though he did check those guesses against the average length of the games started by the pitchers. Dan did not find any significant relationship between the pace of the pitcher and the defense behind him.

This year, however, we have new data at our disposal. MLB Gameday PITCHf/x data includes a time stamp for over 95 percent of pitches that have been thrown in the majors in 2008. This time stamp is down to the second and records when the pitch was in the air. We can use this data to determine how quickly every pitcher in the majors has worked this year.

I already noted some of the fastest and slowest pitchers in a post to THT Live. I don’t think it surprised anyone to see Mark Buehrle at the top of the list or Rafael Betancourt at the bottom. A few Oriole fans remarked to me that Steve Trachsel must be among the slowest workers, but he clocked in at only a couple seconds below average, at 24.1 seconds between pitches.

Here are the data for all 30 teams, including the fastest and slowest worker on each team (minimum 10 innings pitched). The time listed is the average time in seconds between pitches in an at-bat. Pitches that were thrown more than a minute after the previous one were excluded under the assumption that there was a non-pitcher-related game stoppage.

Team            Average Rotation  Pen   Fastest           Time  Slowest             Time
Athletics         20.7    19.9    22.7  Joe Blanton       17.6  Alan Embree         25.3
Cubs              20.7    20.3    21.7  Jon Lieber        17.8  Bob Howry           26.5
Phillies          20.9    20.1    22.5  Clay Condrey      19.1  Tom Gordon          25.4
Nationals         20.9    20.5    21.9  Jason Bergmann    17.7  Jesus Colome        23.8
Padres            21.1    21.0    21.3  Randy Wolf        19.5  Josh Banks          28.1
White Sox         21.1    20.9    21.7  Mark Buehrle      17.2  Bobby Jenks         23.9
Reds              21.2    21.1    21.3  Mike Lincoln      19.6  Edinson Volquez     23.2
Cardinals         21.6    20.8    23.0  Kyle Lohse        19.9  Jason Isringhausen  27.4
Angels            21.6    21.5    21.8  Ervin Santana     20.0  Francisco Rodriguez 24.1
Braves            21.7    21.2    22.3  Chuck James       19.3  Chris Resop         25.7
Marlins           21.8    21.3    22.6  Scott Olsen       19.0  Taylor Tankersley   25.5
Pirates           21.8    21.5    22.5  Zach Duke         20.0  Tyler Yates         26.0
Mariners          21.8    21.6    22.3  R.A. Dickey       17.9  J.J. Putz           27.9
Orioles           22.0    21.6    22.8  Adam Loewen       19.9  Chad Bradford       24.6
Giants            22.2    21.0    24.5  Matt Cain         19.7  Tyler Walker        27.0
Brewers           22.2    21.4    23.9  Ben Sheets        18.4  Guillermo Mota      26.4
Astros            22.2    21.8    23.2  Roy Oswalt        19.3  Jose Valverde       28.0
Mets              22.3    22.0    22.9  John Maine        20.3  Jorge Sosa          25.0
Dodgers           22.4    21.4    24.0  Esteban Loaiza    18.4  Joe Beimel          26.6
Royals            22.5    21.8    24.3  Brian Bannister   20.0  Joel Peralta        29.0
Diamondbacks      22.8    22.2    24.4  Doug Slaten       20.5  Juan Cruz           25.6
Rockies           22.8    21.7    24.8  Franklin Morales  20.7  Kip Wells           26.3
Rangers           22.8    22.1    24.0  Sidney Ponson     19.5  Joaquin Benoit      26.2
Tigers            22.9    22.4    23.9  Justin Verlander  20.4  Denny Bautista      27.6
Twins             22.9    22.1    24.2  Glen Perkins      20.2  Joe Nathan          26.8
Blue Jays         23.1    22.9    23.7  Jesse Carlson     19.5  Jason Frasor        26.6
Rays              23.2    22.4    24.8  Andy Sonnanstine  19.7  Dan Wheeler         27.3
Indians           23.4    22.3    26.1  Aaron Laffey      20.2  Rafael Betancourt   32.0
Red Sox           24.4    23.6    25.9  Justin Masterson  18.9  Jonathan Papelbon   28.4
Yankees           24.6    24.4    24.9  Darrell Rasner    21.3  LaTroy Hawkins      26.7

On the list of slowest pitchers, you’ll notice a lot of relievers; on average, relief pitchers take almost two seconds more between pitches than do starting pitchers. Also, American League East fans, you may be forgiven for wondering why your games drag on.

At this point, it’s probably worth noting the text of Rule 8.04:

8.04 When the bases are unoccupied, the pitcher shall deliver the ball to the batter within 12 seconds after he receives the ball. Each time the pitcher delays the game by violating this rule, the umpire shall call “Ball.”

The 12-second timing starts when the pitcher is in possession of the ball and the batter is in the box, alert to the pitcher. The timing stops when the pitcher releases the ball.

The intent of this rule is to avoid unnecessary delays. The umpire shall insist that the catcher return the ball promptly to the pitcher, and that the pitcher take his position on the rubber promptly. Obvious delay by the pitcher should instantly be penalized by the umpire.

Of course, we measured the time between pitches, not the time to pitch after the pitcher gets the ball from the catcher. We also certainly captured some events like the umpire calling time or pickoff attempts. Nonetheless, I think it’s a fair guess that Rafael Betancourt isn’t sitting in his hotel room every night memorizing Rule 8.04.

This information is all well and good and makes interesting fodder for conversation, but does working quickly have any benefit to the pitcher? On a team level, do the teams that pitch quickly, like the A’s and Cubs, have better defenses?

As measured by the percentage of batted balls they turn into outs, some of the quicker-pitching teams are indeed among the best defensively, but if there is an overall relationship between defensive efficiency and quick pitching in this data, it’s a very weak one.

What about for individual pitchers? Let’s take a look at the batting average on balls in play (BABIP) for both starters and relievers. BABIP is basically the opposite of defensive efficiency, and these graphs show the rate for each pitcher relative to his team. However, in addition to base hits, I’ve included “reaching on an error” and “fielder’s choice with no outs” as marks against the defense.

For starting pitchers, there doesn’t appear to be any relationship between defensive support and the time between pitches, and although there might be a correlation for relievers, it’s very weak. However, not all pitchers may be consistent in how quickly they work; maybe they work slower when they get into trouble, or they don’t have a good feel for their stuff on a given day. What if we ignore the identity of the pitcher and just look at BABIP compared to the time elapsed between the previous pitch and the pitch that was put in play? I’ve binned the pitches into five-second groupings:

For the bulk of the pitches thrown between 11 and 50 seconds after the previous pitch, there doesn’t seem to be much of an effect. However, at the extremes, the pitches thrown within 10 seconds after the previous pitch have a notably lower BABIP (.281), and the pitches thrown more than 50 seconds after the previous pitch have a much higher BABIP (.366). This finding is definitely noteworthy, but further investigation is needed to determine how much of the disparity is due to defensive play and how much is due to other situational differences.

The error bars on the graph illustrate the random error due to sample size. There are almost assuredly some additional systematic errors—for example, I did not control for the quality of the hitter or pitcher. I also didn’t look at groundball/flyball splits or control for the percentage of strikes that a pitcher threw. These are possible avenues for further investigation.

Finally, it probably makes sense to look at the data on the level of the at-bat rather than the pitch. If a pitcher works slowly to a batter, or the batter extends the at-bat by taking and fouling off pitches, perhaps the fielders start to sit back a little.

Again, there may be some effect at the extremes: At-bats taking 15 seconds or less have a BABIP of .314, whereas at-bats taking longer than two minutes have a BABIP of .334.

As I mentioned, there are additional ways to sift this data. One of them that I did look at briefly is the presence of baserunners. In this case, BABIP was about 10 points lower with the bases empty (.323) than with runners on (.333). Here’s how that looks for the BABIP vs. time elapsed in the at-bat.

I don’t intend this study to be the final word on whether fielders play better behind pitchers who work more quickly; it appears that such an effect might exist, but if so, it is small. Mainly, I wanted to let the community know of the existence of the pitch time data and to offer (to the best of my knowledge) the first official peek at it. It is a resource that could be used to answer other questions: Do quick workers get more strike calls? Do pitchers take longer between pitches when they are tired? Does a pitcher’s pace indicate anything about which pitch type he is most likely to throw? Etc.

References & Resources
Thanks to Sportvision and MLBAM for providing this data and to Marv White for his response to my questions about it. It’s pretty cool to find something like this that has been lying around the PITCHf/x data set for two months without me even realizing it!