When are batters most aggresive?

Over the last few weeks I’ve preoccupied myself with the changing nature of the batter-pitcher match-up during the last three decades or so. In that time, I’ve noticed that batters have become considerably more patient at the plate. Perhaps this is a consequence of an increased emphasis on on-base percentage, or a growing wait-your-pitch approach after home run rates sky-rocketed, or possibly even the slow decriminalization of the batter’s strike out.

Whatever the reason, batters nowadays are letting more called strikes pass by than ever.

For some fans, this is a bit of a buzzkill. In a recent Cubs-Giants game at Wrigley field, I witnessed Anthony Rizzo take strike three looking to end a rather frustrating inning for the struggling home team. From behind me, I overheard some of the more articulate denizens of the cheap seats decrying the effects of this phenomenon:

Get the bat off your shoulder and swing, Rizzo! You [disappointing] [bum]!

But with two outs and no one on in the fourth inning of a one-run game, I wondered if it made sense to bother swinging at a borderline pitcher’s pitch, rather than wait on the possibility of a mistake deeper in the count.

This prompted me to ask the question, when are batters most likely to swing? And when are they more likely to let a pitch go by? Using PITCHf/x data from 2010-2012, I examined the change in swing rates by base state, out state and inning hoping to understand how each of these factors might influence hitter aggressiveness. Some of the results seemed very intuitive, but a few things caught me by surprise.

I first looked at swings-per-pitch in each of the eight base states:

Swing rate by base state

Base state Swing% Swing+miss% Called strike% Ball% Foul% In play%
___ 44.4% 8.7% 19.2% 36.1% 17.2% 18.8%
1__ 45.8% 8.8% 16.8% 36.5% 18.1% 19.6%
_2_ 45.3% 9.4% 14.6% 39.5% 17.6% 18.9%
12_ 47.0% 10.0% 15.8% 36.4% 18.4% 19.2%
__3 46.0% 10.0% 13.4% 40.4% 17.6% 18.6%
1_3 48.1% 10.1% 14.1% 37.4% 18.6% 19.7%
_23 44.9% 10.1% 11.8% 43.2% 17.3% 17.6%
123 49.9% 10.7% 15.0% 35.0% 19.4% 19.9%

Batters are clearly at their most patient with the bases empty. In those situations, hitters swing at just 44 percent of the pitches they see, and appear to be quite comfortable allowing almost 20 percent of all pitches sail right pass them as called strikes.

Naturally, batters are similarly patient in situations where the pitcher is more likely to issue a walk. We see that swing rates are considerably lower in critical situations where first base is still open. With runners on second and third, for instance, batters offer at just 45 percent of all pitches, while almost 43 percent of pitches are called balls. This much we’d expect.

Other than the first-and-third situations, however, it seems that batters generally opt to attack pitches more often when there are more runners on base and even more so with runners in scoring position. Is the thrill of the RBI causing batters to make poor decisions at the plate?

In fact, batters are most aggressive when all the glory is on the line.

When the bases are loaded, batters attack just about half of everything that they see. They swing and miss most often, they foul off pitches most often, and they put balls in play at a higher rate with the bags packed than in any other base state.

Perhaps batters are simply swinging more because there are more pitches in the strike zone with the bases loaded? We might be tempted to presume that pitchers are more likely to stay inside the zone in bases loaded situations to avoid walking in the run, but I have a hunch that this isn’t the reason swing rates increase in these scenarios.

Looking at PITCHf/x data from that same three year sample, it appears that pitchers do not necessarily attack the strike zone any more than usual with the bases loaded:

Pitches in the strike zone by base state

Base state Zone% vs. RHB Zone% vs. LHB
___ 53.5% 51.9%
1__ 53.0% 50.5%
_2_ 49.0% 46.9%
12_ 51.4% 49.9%
__3 46.5% 44.7%
1_3 50.0% 47.4%
_23 44.3% 41.6%
123 51.6% 50.2%

It would seem that two powerful forces are wreaking a war of attrition upon one another, while the end result is that not much changes. The tug-of-war between a pitcher’s efforts to avoid contact while simultaneously fearing an RBI walk forces the pitcher to remain inside the zone at normal rates (more or less). And yet, batters still swing substantially more often.

Does this make sense?

Last September, I looked at how batted ball rates may change across the eight base states for Beyond the Box Score. In the comments of that article, Russel Carleton noted that fly ball rates also experience a curious surge in bases loaded situations, suggesting that perhaps the allure of the vaunted grand slam is too irresistible.

(Of course, fly ball rates in bases loaded situations tend to drop a bit with two outs, suggesting that some of that increase in fly ball rate is owed to sac flys.)

But the madness doesn’t necessarily stop there. If we look at the rates at which batters swing at pitches outside of the strike zone, we find that they are most reckless when the prize is the greatest:

Swing rates at pitches outside the zone by base state

Base state Swing% outside zone RHB Swing% outside zone LHB
___ 12.3% 12.3%
1__ 12.8% 12.7%
_2_ 13.6% 13.3%
12_ 13.7% 13.6%
__3 14.9% 14.9%
1_3 14.5% 14.6%
_23 14.8% 14.4%
123 14.7% 15.0%

Even as little leaguers we are taught to widen our strike zone when we find ourselves at the plate in RBI situations. We are instructed to put bat on the ball in some manner to drive ’em in. If it doesn’t land for a hit, at least we can return to the dugout knowing we helped the team’s cause. But what is the cost of swinging at an excess amount of ‘bad’ pitches outside the zone with bases loaded?

If batters maintained their discipline with the bases loaded, would that not lead to better contact, more hits, more walks, more extra base hits, and ultimately more runs?

Of course, we need to remember that bases loaded situations happen most often with two outs and batters also clearly feel more pressure to make something happen when time is running out. With two outs in any base state, we see a remarkable jump in swings, swing and misses, and swings outside the zone:

Swing rates by out

Outs Swing% Swing+miss% Swing% outside zone RHB Swing% outside zone LHB
0 44.8% 8.3% 12.5% 12.3%
1 44.8% 9.3% 12.9% 12.9%
2 49.4% 12.2% 15.2% 15.3%

During the course of a game, however, batters do not tend to show this sort of panic that their run-scoring opportunities are becoming increasingly limited. When we break down swing rates by inning, we find that there is no egregious tendency for batters to begin hacking away at balls outside the zone as the game nears its final moments:

Swing rates by inning

Inning Swing% Swing+miss% Swing% outside zone RHB Swing% outside zone LHB
1 43.2% 7.9% 12.1% 11.9%
2 44.6% 8.8% 12.5% 12.3%
3 44.4% 8.3% 12.5% 12.5%
4 46.1% 8.8% 13.3% 13.1%
5 45.6% 8.6% 12.9% 13.1%
6 46.0% 8.9% 13.1% 13.0%
7 45.7% 9.5% 13.1% 13.0%
8 45.9% 10.1% 13.3% 13.3%
9 45.5% 10.3% 13.1% 12.7%

Swing-and-miss rates do surge significantly as games mature, but this is likely the effect of opposing managers sending better relievers to the mound later in the game. The eighth and ninth innings are typically reserved for those relievers with the best ‘stuff’ in the pen, and the table seems to reflect that assumption.

It is true that swings outside the strike zone do not occur nearly as often in the earliest innings of a game, but that lack of discipline seems to plateau near the fourth inning or so. Swing rates outside the zone do not see the same increase that swing-and-miss rates see in the eighth and ninth innings, implying that perhaps batters are just getting beat inside the zone more often, by harder-throwing uglier-looking closers with definitive velocity advantage.

I’m interested to hear your thoughts on this. Are batters costing their teams more run-scoring opportunities by swinging at bad pitches in certain situations?

References & Resources
Thanks to Baseball Heat Maps for the PITCHf/x data.


8 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Jim
10 years ago

I see you have both decimals and percentages.  How come they don’t add up?

James Gentile
10 years ago

Swing% should equal swing-and-miss + fouls + in play, so adding ball% and called-strike% to swing% should get you 100.

jesse fruchter
10 years ago

How strongly are base- and out-states correlated? It seems like bases empty situations are going to be tied to zero-out situations.

To me, the key issue here is that walks are inherently more valuable with fewer outs and with fewer runners on. The “a walk is as good as a single” mantra works fine provided that no one is on base.

It works less well when there are runners on, so it makes sense that batters would be less patient with runners on. And once you have two outs, you’re much more likely to swing because—even without runners on—the run expectation of an extra-base hit should greatly exceed the run expectation of a walk.

hoch
10 years ago

Some good stuff.  I actually just posted about something similar on Apr 18 (see link below).  My analysis focuses on only 3 of the 24 base/out states (leadoff approach vs. the two RBI approaches).  We’re all in love with large data sets, but I would caution about lumping situations together.  A hitter’s approach with 2 outs and a runner on 3rd is much different than with 1 out.

http://diamondchartsllc.com/sabr/?p=63

I’ve got a follow-up analysis that I’ll release sometime next week once I’ve got a chance to write it up.

Overall, good work James.

Greg H
10 years ago

Good work.  However, one piece of data missing that I’d love to see is the swing rate on the first pitch of the at bat. Do hitters swing at the first pitch more often with the bases loaded?  With runners in scoring position? How about with a runner or runners in scoring position but with first base open? I would argue that swinging at the first pitch is the most aggressive thing a hitter can do.

The numbers back up the popular notion that pitchers are far more likely to pitch around hitters with first base open, trying to get batters to chase a pitch out of the zone, and ultimately settling for a walk if the hitter doesn’t bite. If hitters in turn are more likely to swing at the first pitch in these scenarios, then the pitchers are poised to get the upper hand in the head game of baseball. In the most critical situations, pitchers may be getting ahead in the count early due to hitter impatience.

James Gentile
10 years ago

Thanks, Greg. It would be no problem to run that data. I’ll either post the results here or in a follow-up with other reader suggestions if I decide to go that route.

David
10 years ago

I’d also wonder whether score and batting order position would have an effect on this.

No ma'am we're musicians
10 years ago

‘MLB Now’ today had a chart up showing that taking pitches means nothing with respect to winning games.  The teams in 2012 with the highest pitch count per batter split equally into top and bottom teams.

the Head Game is not about waiting, but keeping the other guy off-balance.  Knowing that batters are taking means a free shot at a strike, bringing up strikeout rates.  Watch for a switch into swinging early by the end of the season.