-
MGL I don't know John, but I think there is little doubt that pitchers are NOT trying to throw a pitch right down the middle on the average or in almost all cases (however you want to couch it), even with a runner on first. For one thing, there are times with a runner on first, that the pitcher does not mind the walk and will almost definitely not throw one "down the middle." For example, with runners on 1 and 3 or even 1 only with 2 outs and someone Pujols at the plate. Do you seriously thinj that a pitcher does not mind grooving one in those situations with those kinds of batters?
And you think that because your average location is the middle of the plate, that means that pitcher is trying to throw down the middle? Perhaps he is trying to throw down the middle 60% of the time, and 20% around the middle symmetrically? Or any combination of percentages you want. As long as the locations are symmetrical, the average location is going to be zero, right? That does not tell you anything about there the pitcher is trying to throw (other than his locations are symmetrical around the middle).
And game theory is an important consideration here. It is not that pitchers should and do try and throw 3-0 pitches down the middle because batters rarely swing at them. Batters rarely swing at them because pitchers don't try and throw 3-0 pitches down the middle (among other reasons, of course). If pitchers did indeed try and throw most 3-0 pitches down the middle, then batters would swing more.
I don't know how much data there is, but you want to at least only look at 3-0 counts to pitchers, and even then in perhaps situations where the pitcher definitely does not want to walk the opposing pitcher (for example, with 2 outs and a runner on third, I am not sure that the pitcher is going to throw a 3-0 pitch down the middle).
Bottom line here is that while pitchers probably can't throw strikes more than 75 or 80% of the time even when they are trying to throw down the middle, you are not going to get at that number by looking at 3-0 counts to all batters, even with a runner on first, since many of your assumptions are clearly incorrect, the main one (that is incorrect) being that almost all of the time a pitcher is trying to throw a pitch right down the middle, with a 3-count, runner on first, and when he throws a fastball. In fact, I would venture a guess that less than 75%, maybe close to 50%, of the time a pitcher is trying to throw that 3-0 fastball right down the middle of the plate, even with first base occupied.
As I said above, you should at least start with 3-0 fastballs to pitchers (as batters). Even though you will not have a whole lot of data, that is going to be a lot closer to a pitchrer's "limit of accuracy" than 3-0 counts against all batters.
-
walshj58 Hi Mitchel,
You make some good points. I think the fact that the ball fraction is the same for cases when 1B is open and when it is occupied supports the notion that pitchers are throwing down the middle (in both cases, actually).
Looking at at 3-0 counts with pitchers at the plate is a good idea. I did something easier, but similar: the ball fraction when the ninth batter is at the plate. I have some 1500 of those since 2000: the ball fraction drops just a little to 33%.
A more comprehensive study of the ball fraction (or pfx standard deviation) as a function of the quality of batter would probably help as well. I didn't do it for this article for lack of time.
-
-
conorsg Good points by MGL. One other thing I would like to add is that watching where catchers set up is not a good indicator of accuracy. Some batters are known to be peekers and oftentimes catchers will set up in a location to throw off the batter if they suspect they are peeking to pick up location. Even wild pitches and passed balls are not a good indicator since a good portion of the time they are the result of "cross-ups".-
walshj58 Good point about catchers setting up against peekers. Still, I have watched many catchers receiving pitches and you can generally tell when the catcher moves his glove towards where he expects the pitch, a very smooth and subtle positioning of the glove, and when he moves to catch a pitch that was off target. That's a much less smooth motion.
Anyway, I believe in the majority of the cases the catcher's glove is in the position where he wants the pitch at the time the pitcher releases the pitch. You can see this pretty easily on TV.
-
-
zukatesta John, thank you very much for the article. I read it all the way through and found it very interesting, but I have to agree with MGL that the premise--pitchers intend to throw the ball down the middle on 3-0--is flawed.
If I could, I would suggest another study: take 3-2 fastballs, where the pitcher's intent will often be to locate a quality strike. Then look at fastballs thrown at roughly the same velocity. I wonder if a study of location of a few hundred or so 90 MPH fastballs on a 3-2 count would indicate anything? You might find tight clusters on the corners, or you might find the same standard deviation of 8 inches bracketing each corner.
I also imagine that pitchers that through 96 might target the upper tiers of the strike zone. This might be interesting, to see how well 96 MPH heaters on 3-2 are clustered around the letters.
-
walshj58 Thanks for reading it!
Your idea is a very interesting one and I will have a look. Actually, I already had a very quick look at the horizontal location of 3-2 pitches -- I do see a sort of double-peak structure as you suggest, but it is superimposed on a much larger and broader distribution. Definitely needs more study, but's it's an interesting approach.
Studying 0-2 pitches might be interesting as well. You might assume (MGL would probably object for game-theory reasons!) that a pitcher definitely does _not_ want to throw the pitch over the middle of the plate on 0-2. Of course, lots of 0-2 pitches go down the middle: what does that say about control and can we quantify it?
Thanks for the interesting comments.
-
The 3-0 count
THE BOOK--Playing The Percentages In Baseball —
... I love all articles on the 3-0 count because the number of choices to the pitcher and batter are very limited, while the payoffs are more apparent. Here’s John Walsh: ...
Monday Mendozas
ESPN Feed: neyer rob —
... ) •How much control do pitchers really have? As John Walsh writes, maybe not as much as you think . Speaking of which, I once read a quote from a major league pitcher -- I wish I could remember which one -- in which he addressed this very point. People think major leaguers can throw the ball wherever they want every time, but they really can't. The difference between a good pitcher and a lousy pitcher is that the good pitcher can throw the ball within six inches of where he wants, and the lousy pitcher might often miss by a foot. •Finally, ...
Link Dump: 12/1
South Side Sox —
... John Walsh looks at the extent to which pitchers actually have pinpoint control. It's a pretty interesting idea and application of pitch f/x stuff. I would also bet that the metric he uses there would line up pretty well with my balls called/strikes called command metric. ...
Around The Web: Kurt Streeter Is Overpaid
Fire Ned Colletti Now —
... care about the economic state of the country as long as Dodger fans keep pouring in to their stadium. Baseball America: Two links that show all the recent minor league transactions that the Dodgers have been involved in. Fan Graphs: Peter Bendix examines Matt Kemp's high BABIP and wonders whether or not it's legit. I agree with his conclusion. Driveline Mechanics: This has to be the weirdest pitching mechanics i've ever seen. The Hardball Times: John Walsh takes a look at control and what it really is. Sports ...
![Derek Jeter: Sportsman of the Year [Rumor Monger]](http://images.dailyradar.com/media/uploads/ballhype/story_preview/2009/11/27/derek_jeter_sportsman_of_the_year_rumor_monger.jpg){kind=small}
