Hit ‘em where they ain’t—if you can

Bobby Crosby stepped to the plate knowing what his job was. It was
the eighth inning and the score was tied. Dan Johnson was leading off
second base, having just stroked a leadoff double to right
field. Crosby didn’t need anybody to tell him what to do—his
goal was to move the runner to third base, either by hit or by getting
the ball to the right side of the diamond.

Crosby, a right-handed batter, was hoping to get something on the
outer half of the plate, something he could drive toward the right
side. The first pitch must have looked pretty good, because Crosby hit
it hard to the right side, where it was scooped up by second baseman
Ray Duhram for the routine 4-3, while Johnson easily moved to third
base. Crosby was greeted by slaps on the fanny and high-fives as he
returned to the dugout and the broadcasters (I assume, I was in the
stands and so did not hear them) complimented Crosby on his fine piece
of hitting.

This kind of thing happens all the time, of course, and some players
are better at it than others. These expert bat-handlers are often
batted in the two-hole, since that is where their talent is deemed
most useful. But, how well can batters really control where they hit
the ball? I’ll be honest: I’m skeptical that batters can place the
ball to the degree that we generally believe. I look at it this way:
In 364 trips to the plate last season, on 62 occasions Bobby Crosby
not only failed to hit the ball where he wanted to, he failed to hit
it at all.

Now, of course, I’m being a bit facetious here—in
most cases the batter is just trying to hit the ball hard
somewhere and only occasionally is he trying to hit it to a
particular area. Still, I think the overall point holds: Hitting
the ball at all is damn hard, how can we expect a batter to aim his
hits?

Arguments about the existence of “place hitting” go back more than a
century, and in fact articles debating the issue appeared regularly in
the sporting press in the early part of the 20th century. Some players
were believed to excel at place hitting, also known by the curious
name of “scientific hitting.” Ty Cobb was considered perhaps the best of
the place hitters. Here is the Georgia Peach’s own take on the
subject:

I believe I can truthfully say that I can hit equally well into either
right or left field and generally at will. … I will claim that when
I am going right, I can drive four out of five fast balls within
fifteen feet of where I want them to go.

Holy Shoot! that is some place hitting. Of course, back in the dead-ball era, the power game was non-existent. Punching the ball through or over the infield was really what hitting was all about (mostly).
Let’s hear from perhaps Cobb’s
biggest rival in the game, Tris Speaker. The Grey Eagle did not
profess to be Cobb’s equal in place hitting, but he was convinced that
he had some place hitting ability:

I do not lay as much stress on place hitting as some. I claim no such
magic power as Ty Cobb is supposed to possess. All I know is that I
have made a study of placing hits and can frequently drive a ball
pretty close to where I want it to go.

Many players of the dead-ball era were known for their place hitting
ability, but there were others who were much agnostic about the
subject. Here’s Brooklyn slugger Jake Daubert, who was among the
National League’s better hitters in the second decade of the 20th
century:

Once or twice in a whole season a batter may meet a ball that is just
to his liking and drive it through the particular gap in the infield
that he aims for. That is place hitting, and I won’t say it does not
exist. But it is the rarest thing in the world.

Of course, in today’s game, nobody is claiming to be able to hit a
ball to within 15 feet of some particular spot. Today’s place hitter
has more modest aims: hitting to the right side or, perhaps, aiming
for a hole that opens up during a hit-and-run play. Daubert also
commented on this kind of hitting:

The most any batter can ever expect to do, as a general thing, is to
drive the ball, at will, either to right or left field. That is not
place hitting in any sense of the word, but it is as near as any
batter can come to directing the course of the ball after he has swung
on it with the bat.

Ok, let’s leave the issue of place hitting 100 years ago to the
baseball historians. But, what about today’s game? Are there some
batters who can hit the ball to left or right field at will? We can
try to answer this question by looking at some data. Here’s how.

I tried to come up with some situations where the batter might be
expected to show some place hitting or bat-handling skill. Then, we
could look in the data to see if batters achieve their place hitting
goals in those situations. Maybe the best such situation is the
hit-and-run play, but unfortunately my data sources (mainly Retrosheet
and Pitchf/x data) do not tell me if a given play was a hit-and-run.

I did come up with the following three situations, where we might
check to see to what degree place hitting exists in today’s game:

1. Runner on second base, nobody out.

2. Batter with two strikes.

3. Runner on third base, fewer than two outs.

Let’s take these one at a time.

Hitting to the right side to advance the runner

Ichiro takes aim. Or does he?

As I described in the introduction to this piece, with a runner on second and nobody out, we expect the
batter to try to hit the ball to the right side. Of course, not every
batter will try to do this every time. It will depend on the score,
the inning, the team’s offensive philosophy and all kinds of other
factors. Still, we would, I believe, expect more batters to try to hit the
ball to the right side when there is a runner on second base and
nobody out, than they otherwise would.

This is the only scenario I could come up with where a batter would reasonably be
expected to try to hit a grounder to the right side. There must be
zero outs, because the idea is to get a runner to third base with one
out, where a routine fly ball scores him. Also, there must be nobody on
first base, to take off the double play.

Here’s how I tried to measure place hitting ability using Retrosheet
play-by-play data. I looked at every plate appearance that occurred with a
runner on second base and first base open where the batter put the
ball in play (excluding home runs and ground-rule doubles). I also
excluded bunts, of which there are many in this situation. I consider
a ball hit to the right side if it was fielded by the first or second
baseman or the right fielder.

The right-side fraction is simply the number of balls hit to the right
side, divided by the number of balls in play. Now, I compare the
right-side fraction for situations with zero outs (when we would
expect batters to be aiming for the right side) with the right-side
fraction when there are one or two outs (when we don’t expect batters
to be trying to hit to right necessarily).

Here are the results, based on data from 2000 to 2007:

+------+-------+-----------+-----------+
| Outs | PA    | RightSide | rightFrac |
+------+-------+-----------+-----------+
|    0 | 20491 |    8411   |    0.41   |
| >= 1 | 80674 |   29186   |    0.36   |
+------+-------+-----------+-----------+

So, here we see some evidence of place hitting ability: Batters in the
sample hit the ball to the right side 36 percent of the time when they
(presumably) were just swinging normally; when they had an incentive
to hit to the right side, they increased that rate to 41 percent.

It would be very interesting to know what fraction of batters really
do aim for the right side, so we could measure their success rate, but
obviously we don’t have that information. Keep in mind, also, that while batters are trying to hit in a particular direction,
pitchers are likely trying to keep them from doing that. So, perhaps even a small increase in balls hit to the right side
represents some kind of true skill.

One might wonder if this place-hitting ability has decreased with
time. After all, a hundred years ago, Ty Cobb was placing his hits
within 15 feet of his intended target (according to Ty). Since nobody today can approach that
kind of ability, maybe we can see a steady decline in place-hitting
acumen over the last 50 years (the period covered by Retrosheet).

The graphic on the right attempts to show the trend. Plotted is the ratio
of the previously described right-side fraction in situations with no outs
to the right-side fraction in situations with one or two outs. The
points represents four-year averages. Hmmm, looks like place-hitting
declined rather sharply during the ’50s and ’60s and has dropped off
more slowly since then.

Foul balls with two strikes

Luke Appling, Hall of Fame shortstop for the Chicago White Sox in the
’30s and ’40s, was famous for his ability to foul off seemingly
innumerable pitches at will. You cannot read anything about the guy
without being regaled with stories of his ability in this
regard. Supposedly, he once fouled off 24 pitches in a single at-bat in an
effort to get a good pitch to hit.

You don’t hear much about Old Aches and Pains anymore, but perhaps he
has an heir in today’s game: a guy named Ichiro who plays up in
Seattle. Dave Cameron of USSMariner.com is one of the keenest observers (and
fans) of Ichiro this side of Hawaii. Dave had this to say about the Mariner
center fielder’s uncanny knack for fouling off pitches:

However, since he occasionally has at-bats where the pitcher isn’t throwing strikes, Ichiro has found a solution to this dilemma – foul pitches out of the strike zone off in an effort to make the pitcher throw him something he can hit. He wants to swing the bat, and is willing to swing at pitches he knows are out of the strike zone in an effort to force the pitcher to throw him one that is.

Wow, that’s some cojones, ain’t it? Ichiro is not only fouling off potential strike-threes, he’s also purposefully swinging at ball four to keep the at-bat alive.

Fouling off pitches intentionally isn’t exactly “hitting them where they ain’t,” but
it does demonstrate superior bat control, so let’s have a look and
see who might have that ability. Actually, John Dewan of Baseball
Info Solutions has already had a look in his excellent “Stat of the
Week” series. Dewan looked at how often batters fouled off pitches
when there were two strikes compared to when there were fewer than two
strikes. He found that batters were 50 percent more likely to
foul off a pitch when he had two strikes on him. Wowie! That’s a big
difference.

Here’s Dewan’s data:

Situation              Pitches    Fouls   Pct.
Fewer than 2 strikes   518,634   82,361   16%
With two strikes       189,088   45,169   24%

Should we interpret this to mean that a large number of
two-strike foul balls are intentional on the part of the batter? Fifty
percent is a lot, it’s a huge effect. If you consider that there must
be a rather wide variation in this ability among players, there must
be a bunch of modern-day Luke Applings out there.

There’s a problem with this interpretation, though—have you spotted
it? The missing piece of information is that batters tend to
swing much more often when they have two strikes than when they have
fewer than two strikes. More swings, more foul balls. The thing you
want to look at to identify bat-handling ability is foul balls per
swing, not foul balls per pitch.

So, let’s add the missing data to Dewan’s numbers. The following
table shows how often the batter swung with two strikes and with fewer than
two strikes, as well as the rates of fouls/pitch and fouls/swing:

+------------+--------+--------------+-------------+-------------+
| NumStrikes | np     | swings/pitch | fouls/pitch | fouls/swing |
+------------+--------+--------------+-------------+-------------+
| <2         | 216432 |         0.40 |        0.15 |        0.38 | 
| 2          |  79343 |         0.61 |        0.24 |        0.39 | 
+------------+--------+--------------+-------------+-------------+

You can see that batters swing about 50 percent more often when they have two
strikes on them. That's natural, of course: They can't be very
selective when a stroll back to the pine is only one called strike away.
When measured by fouls/swing, batters hit virtually the same number of
foul balls with two strikes than with fewer than two strikes (39 percent
versus 38 percent). Essentially all of Dewan's 50 percent was due to more
swings, not to more fouls per swing. (Note that my numbers for fouls/pitch
match up well with what Dewan found, despite that fact that I
used a different data source.)

Frankly, I'm not very surprised. Look, it's hard to hit a baseball at
all; the idea that the average batter can hit a ball in such a way to
send it foul, without missing it, well, it doesn't seem very plausible
does it? If a batter had that kind of control, why wouldn't he hit it
fair somewhere?

This is not to say that nobody excels at
fouling off two-strike pitches. Let's have a look at Ichiro,
bat-handler extraordinaire:

Ichiro Suzuki
+------------+------+--------------+-------------+-------------+
| NumStrikes | np   | swings/pitch | fouls/pitch | fouls/swing |
+------------+------+--------------+-------------+-------------+
| <2         | 1412 |         0.37 |        0.15 |        0.39 | 
| 2          |  463 |         0.71 |        0.31 |        0.43 | 
+------------+------+--------------+-------------+-------------+

Wow, Ichiro really swings a lot more with two strikes, doesn't he? He
almost doubles his swing rate, compared to the average increase of
50 percent. In terms of fouls/swing, Ichiro does appear to hit more fouls
with two strikes, but the increase is a modest 10 percent. That's probably a
real effect, but it hardly seems like Ichiro is fouling off bunches of
pitches at will.

Going for the sacrifice fly

When a batter comes to the plate with a runner on third base and
fewer than two outs, he's looking to hit a fly ball to the
outfield. At least, they always say so in the post-game interview: "I was trying
to drive the ball out of the infield." (When I hear this, I always ask
myself: Aren't, like, 95 percent of batters always trying to drive the
ball out of the infield? Is anybody (except on bunts) trying to hit
the ball to the infield?)

An error of just an inch or two in bat position during the swing will turn a long fly ball into a two-hopper
to short, so if a batter can hit a fly ball at will,
then he is demonstrating some serious bat control. Can a batter
really hit a fly ball when he wants?

Ah, this question holds some nostalgic value for me, as it was the subject of my very first
Hardball Times

article, more than two years ago. I looked at batted ball types,
as well as strikeout and walks, for hitters in sac fly situations and
in all situations. I found that the flyball rate in sac fly
situations was about 5 percent higher. Sounds like batters are able to lift
that ball to the outfield when they need to, right? Well, they also
increased their rate of ground balls by about 6 percent. And pop flies
increased slightly (1 percent), though line drives dropped by 5 percent.

How do we make sense of this? What's happening is that batters strike
out less often (-12 percent) and walk less often (-6 percent) in sac fly
situations. They are indeed putting more balls in play, because they want to
score the run, but they are not hitting more fly balls per ball in
play.

Here are the batted-ball rates per ball in play for the two
situations:

Batted Ball Rates per Ball in Play (percent)
Situation      F     L     G     P
---------    ----  ----  ----  ----
All          28.4  19.0  44.4   8.1
Sac Fly      29.0  17.4  45.5   7.9
(F = fly ball, L = line drive, G = ground ball, P = popup)

Note that the flyball rate is virtually identical for sac fly situations and for all situations (29 percent compared to 28.4 percent).
Batters just don't seem to be able to hit more fly balls even when they
have a strong incentive to do so.

The last word

Look, I'm not saying that batters have zero control over where they hit the ball. I'm convinced that some players are very good at hitting the ball to one side of the infield or the other. An interesting study would investigate which hitters demonstrate a repeatable ability to direct their hits. I found that the average batter does not seem to show this ability to any great degree, but as the saying goes, the average human being has one breast and one testicle. (How valuable such a skill is in the grand scheme of things is a whole different question, one that I will not attempt to answer here.)

On the other hand, I remain skeptical that batters can foul off pitches at will or hit fly balls when they really want to. A lot of people believe that batters do possess these abilities, but the data do not support these beliefs. To sum up my viewpoint, let me plagiarize Speaker: "I do not lay as much stress on place hitting as some." Well said, Spoke.

References & Resources

• Peter Morris' excellent A Game of Inches provides a fascinating account of place hitting in the early days of the game.

• The quotes from Cobb, Speaker and Daubert all come from Baseball Magazine, many issues of which are available on-line. This is a great resource for amateur (and professional) historians of the game.

• For those of you interested in foul balls, the sabermetrician known as Pizza Cutter has everything you wanted to know about them in a three-part series that starts here.

• Thanks to Dave Cameron for his input on Ichiro.