On Situational Hitting

I live in the New York area, which means I get to watch a lot of Mets games on TV. Their announcing crew is well-publicized and somewhat of an acquired taste- I really enjoy them, but some people find them a little much. Whether you like them or hate them, though, they’re very different than the rotating Cardinals announcing team… most of the time. There’s one instance, though, where the two broadcasting teams are carbon copies of each other. Scene: Matt Carpenter at the plate, runner dancing off of second base. The pitcher reads the signs, wipes his brow, and delivers. Carpenter watches strike three, and Jim Edmonds is dismayed. “You know, you have to put the ball in play here, give the team a chance to make something good happen,” he says (I mean, maybe he said that. I’m paraphrasing, but you get the idea). Scene: Michael Conforto at the plate, runner dancing off of second base. The pitcher reads the signs, wipes his brow, and delivers. Conforto watches ball four and trots down to first base, but Keith Hernandez is dismayed. “That’s an RBI chance right there for Michael, and you’d like to see him swing at a hittable pitch during that at-bat,” he says. The names change, the results change, but the criticism remains the same. You’ve got to put the ball in play! Those runners aren’t going to score themselves.

This concept almost seemed too obvious to me to bother looking into. Still, something about this truism stuck in my brain. It sure seemed to me, a biased and somewhat uninformed observer, that Cards hitters were leaving their bats on their shoulders a little too often, prioritizing walks at the expense of situational hitting when appropriate. Could it be? Was I agreeing with Jim Edmonds and Keith Hernandez about baseball strategy? That scary thought led me to do a little digging.

First of all, the Cardinals aren’t hitting that well this year. That’s certainly no secret- every time Dexter Fowler or Marcell Ozuna comes to the plate, it’s a reminder of what was supposed to be. That’s just plain bad hitting though- it’s not like they are putting up excellent advanced stats and not producing, they’re just not putting up good numbers any way you look at it. By BaseRuns, in fact, the Cardinals are scoring more runs per game than you’d expect from their walk and hit totals alone. This fact doesn’t change how I feel when I’m watching the game, though. Why aren’t these guys swinging? Maybe the BaseRuns outperformance reflects a bit of sequencing luck, and the offense could be outperforming by even more if they’d just choke up and slap one into play.

I’ll attempt to look at this issue in two ways. First, there’s a rich data set available for comparison- every other team in baseball. I took each team’s swing rate with runners in scoring position and compared it to their rate in all other situations. This creates something I’ll call Aggression Factor- how much swing rate increases in situations where conventional wisdom would say you should put the ball in play. I compared this to how much better or worse each team has been at scoring runs than BaseRuns would imply- essentially, a measure of whether teams are better or worse than random chance at converting hits and walks into runs. The resulting distribution looks like just that- random chance.

Swinging more or less with runners on doesn’t really seem to do much for your run production. The Cardinals, for the record, increase their Swing% by about 1.6 with runners in scoring position this year- less than the major league average of 2.9% but comfortably positive. Unsurprisingly, every team in baseball swings more with runners in scoring position. Hitters do love RBI’s.

Maybe swinging more isn’t the right answer. A swing isn’t guaranteed to result in a ball in play, after all. What if we look at how much a team’s contact rate changes when there are men to drive in? I looked at what percentage of swings resulted in a swing and a miss, both with and without RISP. Teams that whiff less often with runners on are more likely changing their style to put the ball in play and make productive outs in these situations.

Well, crap. Looks like there’s nothing there either. To be a bit more formal, Aggression Factor has a .011 r^2 and Contact Factor has a .001 r^2. It would be difficult for these two factors to be LESS correlated to run-scoring prowess.

I don’t have a perfect reason for why this is true. I keep going back, though, to an interview I heard about pitch calling last year. A catcher (Tyler Flowers, I think, but as I’m not 100% I’m just going to call him a catcher here) was asked whether he would adjust his pitcher’s gameplan to go after some weaknesses evident in an opposing batter’s swing. It seemed like a pretty straightforward question- if a guy is weak to high fastballs, why not throw him a bunch of high fastballs? The catcher immediately responded- just not the way I expected. He always called to the pitcher’s strengths rather than the batter’s weaknesses. If a batter were to be particularly good against the pitcher’s best pitch, he might prioritize the second-best pitch instead, but that was about as far away from sticking with the routine as he was willing to go. Why, he was asked? Well, it’s really hard to throw a strike in a major league baseball game. Pitchers spend their whole lives repeating motions for a few pitches. That’s how they become major league pitchers. Asking them to try to do something they’re less comfortable with at game speed throws their bodies out of sync, and often results in a worse outcome than just sticking with what works. It seems only reasonable to me that hitting is the same way. Major league hitters have spent tens of thousands of hours perfecting their batting eye and swing. Their mechanics are tuned just so, and they rely on tiny visual clues from a white ball sixty feet away to decide whether to swing. Maybe, just maybe, adding a voice to their head saying ‘Hey, it’s first and third with no one out, you should be a little bit more likely to swing at a borderline pitch here, definitely don’t miss it though, maybe choke up and try to flare it’ is more trouble than it’s worth.

Even if you think that hitters can teach themselves situational awareness, though, it would be useful to know which situations are best for contact hitting. Some of them are easy. A man on third and one out is a great time for a productive out. Others are trickier. First and second, no one out. What should you do? Second and third with one out sounds pretty good. A walk to load the bases, however, sounds pretty great as well. First and second with one out, in the event of a strikeout, isn’t even the end of the world. To come up with a consistent way to do this, I considered two fictional hitters. Let’s call one of them Catt Marpenter and one Suchiro Izuki. Catt Marpenter is blessed with the batting eye of a Greek God, but he’s never put a ball in play in his life. Half of his plate appearances end in walks and half in strikeouts. Suchiro Izuki, on the other hand, is a master of contact. He puts the ball in play every at-bat, and it either is an out that advances all the runners or a single. If there are two outs, runners can get two bases on his singles; otherwise they get one. These are totally fictional characters- any resemblance to real life is unintentional.

Okay, so we’ve got our two protagonists. Catt Marpenter is definitely the guy you want with no one on base, no one’s disputing that. The question is how often Izuki’s balls in play need to fall for a hit for him to equal Marpenter given various base-out states. I’ll call this ‘Breakeven Hit Rate’ because there aren’t enough bad names for stats in this article already. In any case, the situation is incredibly reductive, but it’s a good way to see where contact is most useful. I took run expectancy values from Baseball Prospectus to figure out what happens in the various outcomes. Here are the results:

I wouldn’t pay too much attention to the numbers specifically, since the situations they’re describing are pretty much nonsense. What I would pay attention to is the ranking of the situations. There are four situations where Izuki beats Marpenter even if he is making a productive out 100% of the time. Those are definitely situations where you want a contact hitter at the plate. They all involve a man on third and less than two out, so that old piece of received wisdom seems pretty safe. The situation where you most want walks-and-K’s (among situations with runners in scoring position) is first and second with one out. That also checks out- a walk is as good as a single there.

Okay, so we have two somewhat conflicting ideas here. On the team level, trying to swing more or hit for more contact doesn’t seem to do much. I didn’t control for the types of pitches that each team saw, but the relationships were so poor that there seemed to be no point. Simply put, trying to make more contact as a team doesn’t seem to work. On the other hand, there are very clearly situations where you want someone who specializes in making contact at the plate. Where does that leave us? Well, I’m not sure. Off the top of my head, it probably means you want to have a guy who makes a lot of contact available off the bench, because situations come up all the time where that skill set is invaluable. It also means I’m going to be a lot less annoyed if Tommy Pham strikes out with a runner on second, or if Marcell Ozuna is hacking away with a guy on first. Letting a player be himself is probably the best way to set him up to succeed.

Now, I’ve always heard that the best articles have a conclusion that the rest of the story is written around, making the whole narrative coherent and holding readers’ interest all the way through. That sounds like it would be an awesome article! A.E. Schafer writes three times a week- I’d suggest you check those out, they’re great. As for this article, I don’t really have a conclusion for you. Situational hitting exists! Teams who do things that I consider to be situational hitting don’t seem to do any better! Cats and dogs living together! I’m as confused as you are. Maybe the Cardinals will just hit a home run next time and the announcers won’t have anything to say.