Note: This article was originally published at the Statistically Speaking blog at MVN.com on February 24, 2008. Since the MVN.com site is defunct and its articles are no longer available on the web, I am re-publishing the article here.
I’ll warn you from the start that the title is a tad ambitious. I don’t know exactly how Brian Bannister wins in the major leagues with a below-average fastball speed, but I hope to share some of what I have learned on the topic. This article will take the form of a three-part series.
In case you’ve been hiding under the proverbial sabermetric rock the last few weeks–maybe you’re one of those weirdos who believe players are human or you’ve been out of your garage recently to look at the sky–Brian Bannister gave a fascinating three-part interview to Tim Dierkes at MLB Trade Rumors last month.
In Part 3 of the interview, Bannister talked about his opponents’ batting average on balls in play (BABIP).
I think a lot of fans underestimate how much time I spend working with statistics to improve my performance on the field. For those that don’t know, the typical BABIP for starting pitchers in Major League Baseball is around .300 give or take a few points. The common (and valid) argument is that over the course of a pitcher’s career, he can not control his BABIP from year-to-year (because it is random), but over a period of time it will settle into the median range of roughly .300 (the peak of the bell curve). Therefore, pitchers that have a BABIP of under .300 are due to regress in subsequent years and pitchers with a BABIP above .300 should see some improvement (assuming they are a Major League Average pitcher).
Because I don’t have enough of a sample size yet (service time), I don’t claim to be able to beat the .300 average year in and year out at the Major League level. However, I also don’t feel that every pitcher is hopelessly bound to that .300 number for his career if he takes some steps to improve his odds – which is what pitching is all about.
In the interview, Bannister postulated a reason for his success on BABIP.
So, to finally answer the question about BABIP, if we look at the numbers above, how can a Major League pitcher try and beat the .300 BABIP average? By pitching in 0-2, 1-2, & 2-2 counts more often than the historical averages of pitchers in the Major Leagues. Until a pitcher reaches two strikes, he has no historical statistical advantage over the hitter. In fact, my batting averages against in 0-1, 1-0, & 1-1 counts are .297/.295/.311 respectively, very close to the roughly .300 average.
My explanation for why I have beat the average so far is that in my career I have been able to get a Major League hitter to put the ball in play in a 1-2 or 0-2 count 155 times, and in a 2-0 or 2-1 count 78 times. That’s twice as often in my favor, & I’ll take those odds.
This interview has gotten a lot of buzz in sabermetric cyberspace. Several people have taken a look at BABIP at different ball-strike counts, including my colleague at StatSpeak, Pizza Cutter. There seems to be some ability for the pitcher to control the count on which hitters put balls into play, but it looks like a fairly small effect on average. (Pizza, correct me if I’m summarizing your conclusions incorrectly.)
Bannister also mentioned to Dierkes that getting two strikes on the hitter gives him the strategic advantage in terms of pitch selection.
It is obvious that hitters, even at the Major League level, do not perform as well when the count is in the pitcher’s favor, and vice-versa. This is because with two strikes, a hitter HAS to swing at a pitch in the strike zone or he is out, and he must also make a split-second decision on whether a borderline pitch is a strike or not, reducing his ability to put a good swing on the ball. What this does is take away a hitter’s choice. If I throw a curveball with two strikes, the hitter has to swing if the pitch is in the strike zone, whether he is good at hitting a curveball or not. He also does not have a choice on location. We are all familiar with Ted Williams‘ famous strike zone averages at the Baseball Hall of Fame. It is well-known that a pitch knee-high on the outside corner will not have the same batting average or OBP/SLG/OPS as one waist-high right down the middle. Here is a comparison of the batting averages and slugging percentage on my fastball vs. my curveball:
We do know from John Walsh’s work something about batting average and slugging percentage against the typical major-league fastball (.330/.521) and curveball (.310/.471). If Bannister is correct in his numbers, he’s doing quite a bit better than the league with both the fastball and curveball. But is Bannister correct in the numbers he quotes and assertions he makes?
So far, most people are accepting what Bannister said at face value. Let’s take a closer look and see if we should believe his numbers and conclusions. We’ll draw on two data sets from the 2007 season. One is the standard pitch-by-pitch result data for all of Bannister’s 2603 pitches in 2007. With this data set we can examine results on balls in play and how Bannister performed in various ball-strike counts. The second data set is the detailed PITCHf/x trajectory data recorded for 1304 of Bannister’s pitches, or about half of his starts. With this data set we can identify pitch types and reliable strike zone location information in order to gain a greater understanding of Bannister’s pitching strategies.
First, what does it mean when we say that Bannister had abnormal success with BABIP in 2007? The work of Voros McCracken and others over the last decade has shown that pitchers in general have limited control over the results of balls in play, such that on a single-season level, chance plays a bigger role in batted ball outcomes than does the pitcher’s own skill. To oversimplify the argument, a pitcher who performed better than the league average for BABIP is considered to have gotten lucky and should be expected, on average, to have a BABIP next year that is much closer to the league average BABIP than to his own performance in BABIP from the preceding season.
The American League BABIP in 2007 was .304. (I’m including sacrifice flies as outs in my BABIP calculations.) Brian Bannister had the lowest BABIP among starting pitchers in the American League at .262. If Bannister had performed at the league average for BABIP, he would have allowed 164 hits on balls in play instead of the 141 he actually allowed. Allowing 23 extra hits on balls in play would translate into roughly 15 extra runs, or 0.80 runs per 9 innings. In other words, if you attribute the superior BABIP performance entirely to luck, when predicting how Bannister will do in the future, you’d be better off acting as if he had a 2007 ERA more like 4.60 rather than the 3.87 mark he actually posted.
Researchers familiar with this topic will know that we can’t entirely attribute a pitcher’s BABIP performance to luck. Any given major league pitcher has some control over BABIP. However, since that control is small, we regress BABIP performance heavily to the mean when using BABIP data for making predictions of future performance. We might also want to take into account the quality of the fielders that played behind the pitcher or the park in which they played. I’m not going to go into detail on either of those two fronts since those topics have been covered thoroughly by others more able than me, but suffice it to say that neither effect comes close to explaining Bannister’s BABIP performance in 2007. From Bannister’s 2007 sample, regressed to the mean, we might estimate his BABIP skill at a performance level .006 better than league average. The Royals fielders (1379/4493) in 2007 were just slightly below average at converting balls in play into outs, with a team BABIP of .307. So we’re still left wondering how Bannister managed a BABIP of .262.
Yes, that’s the stat for me!
It stands alone as the stuff of luck
What of Bannister’s proposition that it’s all in the ball-strike counts in which he pitches and induces balls in play? Does he actually pitch in favorable counts and get balls in play in those counts more often than other pitchers? If he does, does that explain his BABIP performance, or are we still left attributing most of it to chance?
Before we can answer those questions, we need to establish a method for measuring a pitcher’s (or the whole league’s) performance at a given count. The most accurate and comprehensive way is to determine the run values of various events using linear weights. For sake of brevity, I won’t go into the details of determining the run values for different counts in this article. You can find a lot of good material on the subject at Tango’s Book blog, this thread on Joe Sheehan’s recent article being one example.
Here are the 2007 American League average run values for a pitch thrown at each count. The values for a ball, a strike, or a ball in play at a given count are shown relative to the overall run value for that count. (Strikes here include foul balls at a two-strike count.)
You can see that the counts overall most favorable to pitchers are 0-2 and 1-2, followed by 0-1, 2-2, and 1-1. When a ball is put in play, every count favors the hitter on a net basis except 0-2. When adjusted for expectations at that count, the hitter also comes out behind at 2-0, 3-0, and 3-1. Why does putting the ball in play at these counts favor the pitcher? Because the hitter is fairly likely to get a walk or bat in another hitter’s count if he takes the 2-0 pitch. The hitter does better than the average for all counts when putting a ball in play at 2-0, but not better than the average of the other results he gets after the count reaches 2-0. So putting the ball in play at 2-0 is a net negative at that point, on average.
Those are the numbers for the league as a whole. What about Brian Bannister in particular? Here are his linear weight run values by count for 2007. The values for a ball, a strike, or a ball in play at a given count are shown relative to the league average overall run value for that count (shown in the previous table).
Since Bannister performed much better than average in run prevention, it’s no surprise that he did better than average in 8 of 12 counts. The counts where he did the best compared to expectations are 0-2 and 2-0. The counts where he did the best overall are 0-2, 1-2, and 0-1. Compared to the average pitcher, he performs much worse when the count is 3-1.
Now that we’ve established a baseline, we can check whether Bannister throws his pitches in favorable counts more often than other pitchers.
Bannister throws a first-pitch strike 4% more often than average. He also pitched more often than average in a 1-1 count and less often than average at 0-2 and 1-0.
We are now in position to evaluate one of his claimed reasons for success: “[H]ow can a Major League pitcher try and beat the .300 BABIP average? By pitching in 0-2, 1-2, & 2-2 counts more often than the historical averages of pitchers in the Major Leagues.” In fact, in 2007, Bannister pitched less often in those counts than the average pitcher, entering the 0-2, 1-2, or 2-2 count with 63% of batters he faced, compared to the league average of 66%.
So maybe he wasn’t accurate about which counts give him an advantage, but perhaps he pitches in advantageous counts overall anyway? As my son would say, “Let’s see!”
It turns out the answer is yes, at least a little bit. Depending on whether you use Bannister’s own performance in each count or the league average performance in each count, you end up with an advantage to Bannister between 1.3 and 2.6 runs based on pitching in favorable counts. Let’s settle on a figure half way in between since I don’t currently have any better idea of how to regress these numbers to the mean. He gives up an extra 2.2 runs by pitching less often than average at an 0-2 count, but more than gains that back by pitching less often at 3-0 (-1.5 runs) and more often at 0-1 (-1.1 runs). He also pitches less often at 3-1 (-0.7 runs) and 2-0 (-0.6 runs).
A two-run difference hardly begins to explain the 15 runs by which Bannister outperformed league average BABIP in 2007. In fact, it may have nothing to do with BABIP at all. We could have guessed that Bannister didn’t pitch in 3-0 or 3-1 counts very often since he allowed less walks than average. In the interview Bannister was mainly talking about balls in play, not pitches rung up for balls or strikes, so let’s get to the heart of the matter and focus on balls in play.
Bannister claimed he’d been able to get batters to put the ball in play 155 times with a 0-2 or 1-2 count versus 78 times at 2-0 or 2-1. In 2007, batters put the ball in play against Bannister 78 times at 0-2 and 1-2 versus 61 times at 2-0 or 2-1. The numbers he quoted were for his career, so to check them, we need to add in his 2006 numbers, 19 and 17 respectively, to get a total of 97 career balls in play at 0-2 or 1-2 and 78 career balls in play at 2-0 or 2-1. The second number matches what Bannister said, but the first number isn’t even close. Did Bannister leave out a count? I can’t find any other combination of counts that adds up to the numbers Bannister said. You can look for yourself.
Inconsistencies in Bannister’s arithmetic aside, the main thing to note is that he did not induce balls in play in advantageous counts more often than average. He comes out at slightly less than a one-run disadvantage overall, including, notably, a 1.6-run disadvantage at 0-2. When looking at his run values for balls in play, we need to remember that small sample size affects the reliability of these numbers. You might be able to chalk most of this effect up to chance. In any case, Bannister is certainly not gaining any advantage here.
Where does that leave us? Are we left knowing nothing more than when we started other than that the reasons Bannister offered for his BABIP performance don’t hold water? In Part 2, we’ll move on to the PITCHf/x data set and see if learning more about his arsenal can help us answer the BABIP question.