Swing-Mirroring: Subconscious Cues Influence Batter Decisions

How do teammates' results indirectly impact hitters?

Have you ever felt a huge burst of energy after seeing your favorite team notch a big hit? Maybe even an urge to dig your cleats out of the back of your closet and don a jersey again? These feelings, unsurprisingly, seem to extend to the players themselves: announcers often speak of hitters “swinging out of their shoes” in order to “join the hit parade” after a teammate’s base-knock. But is the recent performance of a teammate, subtly encouraging more swings, really enough to override other I-should-swing or I-shouldn’t-swing cues?


To Swing or Not to Swing?


To answer this question, I first examined these other cues by asking the following: in general, how do hitters decide whether to swing at a pitch or not? Before the pitch comes in, hitters can rely on their knowledge of the pitcher’s command and offerings to determine if they can reasonably sit on a belt-high fastball. What’s more, hitters can take strategic game information into account—if first base is open and a runner is on second, a hitter might expect to be pitched around, and in most two-strike counts, a hitter might anticipate a waste pitch.

Hitters can also think back to pitches they have already seen in the at-bat. High fastballs, for example, are traditionally followed by low curveballs (or vice versa) because low curveballs and high fastballs perform better in this sequence; low curveballs are said to “set up” and be set up by high fastballs. In the following sequence, Rich Hill starts off Matt Davidson with a curveball in the lower half of the zone for a called strike. Then, he gets Davidson fishing on a high fastball.

After considering game information, pitcher profile, and previous pitches, as the pitcher goes into their windup and throughout a pitch’s early trajectory, the batter can then adjust their initial prediction until they have to decide to swing, ultimately having to do so at least 150 milliseconds before the pitch crosses the plate in order to optimize point of contact. 

But on the first pitch of an at-bat, some pre-pitch information is more limited. For instance, as the new at-bat begins, the pitcher’s sequencing essentially resets—no particular combination of location and pitch type has been “set up” yet. Further, according to Statcast’s “zones,” pitchers threw in the strike zone on approximately 52% of 0-0 counts in 2021. This is the smallest difference from 50% for any count besides 1-1 (49.5% of 1-1 pitches were in the zone). This is important because a 50% chance of seeing a strike would mean a hitter could not rely on count to predict the incoming pitch; it would be a tossup between a strike and a ball. Below is a table of Zone% by count:

Pitcher Zone Percentage by Count

So, it stands to reason that hitters might be more susceptible to the influence of other cues on the first pitch. One such cue that is especially salient in those situations is the outcome of the previous at-bat. In a groundbreaking 1951 study, Solomon Asch defined two types of social influence that can subconsciously sway decision-making (in our case, the decision regarding whether to swing or not). Normative influence pushes people to conform to fit in more with a group, regardless of whether the group is well-informed or not. Informational influence pushes people to conform when they believe the group is more knowledgeable than them. When the previous hitter gets a hit, they exert informational influence on the next hitter; namely, they indicate that it is beneficial to swing. I reasoned that the previous hit with the most influence would be a home run, given its prominence in the mind of a hitter.


Different Hits = Different FPS%


Using 2021 Statcast data, cleaned for inconsistencies, I set out to quantify the effects of this influence. To simplify things, I made sure that the hit did not knock the pitcher out of the game; in other words, the new batter would be facing the same pitcher that just gave up a hit, the pitcher that they had just received “information” about. I also ensured that each hit did not somehow result in an inning-ending out; no one made the third out trying to advance an extra base. This way, the influence would be coming from teammates, and in-group members, in the same half-inning. Below is the mean Swing% on first pitches (FPS%) sorted by the outcome of the previous at-bat:

FPS% by Previous Hitter’s Result

Given that the 2021 data is just a subset of the vast amount of baseball data in existence, I used statistical techniques, factoring in sample size, to determine whether we could generalize these results. All but two of the differences in FPS% across outcomes were statistically significant, or generalizable. The only two pairs of outcomes that did not have a statistically significant difference in FPS% were the post-single and post-double combination and the post-triple and post-double combination. The latter was at least trending towards significance, indicating some potential for a real difference. Either way, every other kind of hit (single, double, and triple) portended a significantly (at the ? = 0.05 level) higher FPS% in the next at-bat than a home run did. Home runs only caused a higher FPS% compared to non-hits.

If informational influence were the only force at play here, home runs would have a larger impact. Instead, these results indicate that, while social influence may still be a factor, there is likely something else going on here. Perhaps hitters following doubles and triples, in addition to being socially influenced, may feel a sense of urgency to put the ball in play in order to either move the runner over or drive him in. 


More Urgency = More FP Swings


To determine if urgency alone portends significant changes, I looked at FPS% with men on base but not after a hit. These results are listed in the table below:

FPS% by Men on Base

All of the differences in FPS% between the conditions were significant, and FPS% also followed the same pattern as above (increasing as the baserunner advanced), indicating that urgency likely has an impact. However, compared to their counterparts in the other table above (i.e., runner on first vs. post-single), all conditions had a significantly lower FPS%. The major difference between the two tables is that in the first, the previous batter had gotten a hit. This essentially controls for urgency across the two tables and points to informational influence as the driving factor behind at least the inter-table differences.

One more factor left to control for is the number of outs; the previous batter in the non-post-hit conditions was significantly more likely to have made an out than in the post-hit conditions. This makes sense intuitively, but the real question is, what kind of effect does the number of outs have on FPS%? The table below provides some answers:

FPS% by Number of Outs

There was no significant difference between the one- and two-out FPS%, but both were significantly larger than the no-out FPS%. Similar to the influence of runners on base, more outs may portend a greater sense of urgency. Or, fewer worries that swinging may lead to a double play. So, if we had to generalize, more outs would typically mean a higher FPS%. Yet, even though there were more outs following a non-hit on average, batters following a non-hit posted a lower FPS% on average than those following a hit. This indicates that the power of informational influence is able to overcome the opposing subconscious force wrought by the number of outs.




Practically, what does this all mean? These results seem most useful to pitchers; perhaps they should feel less pressure to throw a first-pitch strike after allowing a non-home-run hit. The results for first-pitch O-Swing% after hits and non-hits and grouped by the number of outs show similar patterns to general Swing%:

First-Pitch O-Swing% by Previous Hitter’s Result
First-Pitch O-Swing% by Men on Base
First-Pitch O-Swing% by Number of Outs

However, there is no real difference here between post-hits and post-non-hits. Additionally, home runs don’t portend a higher O-Swing% on the first pitch of the next at-bat, even compared to non-hits and no one on base. This doesn’t knock the power of informational influence, but it indicates that the influence may have more weight when it comes to first pitches that are in the strike zone. Sure enough, that seems to be the case, as the trends in the following tables more closely mirror the original ones:

First-Pitch Z-Swing% by Previous Hitter’s Outcome
First-Pitch Z-Swing% by Men on Base
First-Pitch Z-Swing% by Number of Outs

This makes sense given that, since hits typically come against pitches in the zone, the social influence should be at its peak for first-pitch strikes. So, if pitchers can’t rely on a higher likelihood of hitters chasing, how can they use the effects of informational influence to their advantage? Their best option may just be to avoid pouring one in on the first pitch after a hit; aim for the corners and don’t assume the first pitch is a freebie, even for typically passive hitters.


Case Studies


Let’s take a quick look at the starting pitchers with the best first-pitch approaches according to my advice. In 2021, the qualified starter with the lowest number of hits allowed, Max Scherzer, gave up 119, and the qualified reliever with the highest number of hits allowed, Nabil Crismatt, gave up 87. So, to capture mostly starters, I used a cutoff of 100 hits allowed. There were 112 such pitchers. Here are the 10 with the lowest rates of middle-middle first-pitch strikes after hits:

2021 First-Pitch Post-Hit Middle-Middle% Laggards

And the 10 with the highest rates of corner strikes on the first pitch after a hit:

2021 First-Pitch Post-Hit Corner% Leaders

The only two pitchers who appear in both tables are JT Brubaker and Zac Gallen. Brubaker limped to a 5.36 ERA and 5.16 FIP last year, while Gallen put up a 4.30 and a 4.25 in those categories, respectively. Those were hardly stellar results, but that doesn’t necessarily mean their post-hit first-pitch approaches were to blame. In fact, Brubaker netted the 10th-highest swinging-strike rate on these pitches. Here’s a perfectly-placed first-pitch slider from Brubaker that prompted a Marcell Ozuna whiff:

Gallen came in at merely average, ranking 43rd in swinging-strike rate, but he also generated the 22nd-most foul strikes after allowing a hit (Brubaker garnered the 34th-most). With runners on the corners, a well-located first-pitch Gallen changeup yielded this weak (66.5 MPH exit velocity) foul fly from Austin Riley:

To truly determine the effectiveness of this first-pitch post-hit approach, more case studies are needed. But Brubaker and Gallen present preliminary evidence in favor of it. If it became widely adopted, hitters could counter this move by forcing themselves to be more passive against tougher strikes to hit on the first pitch, even in the face of strong subconscious influence. As with many other aspects of the strategic battle between batter and pitcher, if pitchers choose to heed my advice, the result could turn into another war of adjustments.


Feature image by Michael Packard (@CollectingPack on Twitter) / Photography by Frank Jansky / Icon Sportswire

Alex Eisert

Alex graduated from Vassar College, where he served as the paper's sports editor, in 2022 with a psychology and cognitive science double major and economics and philosophy minors. He is especially interested in how and why people make decisions, something he clearly struggled with when determining his course of study in college.

Leave a Reply

Your email address will not be published. Required fields are marked *

Account / Login