Strikeouts are the most efficient way to pitch

Royals broadcaster Rex Hudler often complains that pitchers should pitch to contact in order to get deeper into games. He’s wrong, and here are the numbers to prove it.

Royals broadcaster Rex Hudler has been consistent about two things all year. First, he thinks hitters need to shorten (or, after speaking with team Hall of Famer John Mayberry, “cut back on”) their swings. The other is that while strikeouts are sexy, they’re an inefficient way to get outs and lead to pitchers being unable to pitch deep into games. However, that begs the question: If strikeouts are so inefficient, then why are teams (including the Royals) chasing them?

Counting pitches

First, I decided to dig into the numbers of how many pitches it takes to get a strikeout. I discovered after messing around with Baseball Savant’s search functionality for a while, that there is a pretty consistent number of pitches associated with batted ball outs vs strikeouts. After all, Rex contends specifically that it takes more pitches to strike someone out than to achieve a batted ball out. So I checked how many pitches it took for Royals pitchers to strike a batter out or get a batted ball out in 2008 - the first year Statcast data was available - 2023, and 2024 prior to Tuesday’s game. I also broke out just 2024 Cole Ragans, frequently the pitcher on the mound when Rex talks about the strikeouts, to see if he varied significantly from the other data sets.

I discovered that the numbers are actually remarkably consistent in all four data sets, even once I break them out just for Ragans. It takes roughly 4.9 pitches to get a strikeout and 3.4 pitches to get a batted ball out. That’s a difference of 1.5 pitches per at-bar or, to put it another way, it takes 50% more pitches to strike someone out than it does to get a batted ball out. That makes sense because you have to throw at least three pitches to get a strikeout (barring the recent inclusion of pitch timer violations by a batter) but a batted ball out can happen on only one or two pitches. So, case closed! Rex is right, right?

Naturally, if it were that simple, this article wouldn’t exist.

Counting the rest of the pitches

There are more outcomes to a plate appearance than those, and those also feature remarkably consistent numbers. It takes roughly 5.4 pitches to offer a free pass to a batter (I combined walks and hit batters for this number) and 3.4 pitches to allow a hit. But, here’s the thing, the pitches that lead to a walk or a batted ball hit don’t get the pitcher any closer to the end of the game.

Baseball games, as we all know, don’t end after a set amount of time as in many other sports. They end when one team completes 27 (or more) outs with the lead. In football or basketball, there might be some value in making the other team take longer while playing offense; in baseball, every pitch that doesn’t lead to an out is wasted, in a sense.

Let me put this another way, I counted up Cole Ragans pitches from his last start against the Guardians. And, yes, it took him 32 pitches to get his six strikeouts (5.33 pitches per K,) and only 19 pitches to get seven batted ball outs (one of which got him a double play, so 2.71 pitches per batted ball that resulted in at least one out but only 2.38 pitches per out recorded via batted ball.) But the real problem for his pitch count was in the 42 pitches he spent allowing runners on base (six hits, one walk, one hit batter.) That’s almost half of his pitches! Ragans was ultimately pulled before finishing the fifth inning due to ineffectiveness and a pitch count approaching 100. But his pitch count problems on Saturday weren’t because he was striking batters out, they were because he was allowing too many runners to reach.

Even averaging 5 pitches per strikeout, if a pitcher doesn’t allow any base runners, they can get 20 outs - almost seven innings of work - and only throw 100 pitches. 27 outs would require 135 pitches, which is probably more than you want a pitcher to throw, but if he was literally striking everyone out, you might be OK with it. I’d hate to be the manager who had to pull a pitcher who had struck out every batter he’d faced for seven or more innings.

Not all outcomes are equal

Here’s the other thing about that pitch data that I find particularly interesting. It takes almost the same number of pitches to allow a hit as it does to induce a batted ball out. That means a batter will see roughly the same number of pitches before putting the ball in play regardless of whether that results in a hit or an out. This makes a lot of sense when we consider the entire reason for a pitching stat such as Fielding Independent Pitching (FIP.)

FIP exists to try to calculate a pitcher’s effectiveness based on the things he can control. It strives to only credit the pitcher with the outs he achieves via strikeout, the runners he allows via walk, or the home runs he gives up. Any ball hit into play, the creators of FIP reasoned, could have any kind of result. I think anyone familiar with baseball recognizes that as true. We’re all familiar with 118 MPH screamers hit right at a fielder for an out or 20 MPH dribblers that turn into hits because they just happened to be hit to the right place against the right defenders in the right alignment. And that’s to say nothing of the different outcomes depending on the skill of the fielders playing behind the pitcher.

So, a pitcher who doesn’t get many strikeouts is going to have a worse FIP because the randomness of batted balls is presumed to be outside of his control and will likely go against him sooner or later. Alternately, a pitcher who has been giving up a lot of hits but not walking many and striking out a fair few will have a better FIP because of the same reasoning about batted balls.

The logic behind FIP and the implications of the pitch count data agree: there is no real way to tell whether a batted ball will turn into an out or a hit.

Bringing things together

And that takes us back to Rex’s argument that strikeouts are pitch-inefficient. If you only compare them to the batted ball outs, that may seem true. But that’s not a good comparison. Because there is little if anything to differentiate batted ball outs from batted ball hits,* they should not be separated. At first, you might think, “So what, they’re the same number of pitches, so it’s the same thing!” Except it isn’t, because, as we established before, allowing a hit doesn’t just cost you pitches, it doesn’t get you any closer to the end of the game.

This means we should actually be comparing the average number of pitches to get a strikeout - which always results in an out - to the number of pitches it takes to give up a hit AND to get batted ball outs, but only divided by the number of batted ball outs.

*And that’s to say nothing of walks, which are the most inefficient outcome of a plate appearance for a pitcher even before you account for the fact that they don’t get you any closer to the end of the game.

So, to go back to the Ragans example above. Yes, it took him 5.33 pitches per strikeout. But it took him 50 pitches to get 8 outs (7 plays, including one double play) via batted ball. That’s an average of 6.25 pitches per out by batted ball. If you do the same math over the entirety of the Royals 2024 pitching results - keeping in mind that they’ve allowed only slightly more than the average number of hits - it has taken them 4.87 pitches per at-bat that ends in a batted ball to turn them into an out.

By that metric, strikeouts (which, as noted earlier, take approximately 4.9 pitches to achieve) and batted ball outs require roughly the same number of pitches. But that isn’t the whole story, either. Strikeouts never* turn into baserunners or runs, while batted balls often do. For context, Royals pitchers have allowed a combined BABIP of .298 this season, which means three out of those ten batted balls turn into baserunners, not including errors. None of their strikeouts have resulted in a baserunner.

*Yes, I know strikeouts can result in a runner reaching base, but that happens so infrequently as to be statistically irrelevant.

I am aware that Rex preaches not just pitching to contact, but seeking to induce weak contact. That, however, makes no sense. As we established earlier, even a weakly hit ball can turn into a hit and a pitcher has very little control over that. Beyond that, any tactic a pitcher might use to fool a hitter into being less likely to square up a ball is the exact same tactic they’d use to attempt to fool a hitter into missing completely, just performed less well.

The worst outcome for a pitcher is allowing a baserunner. Doing so not only gets them no closer to their goal of ending the game but also puts them in greater danger of allowing runs. Therefore, seeking strikeouts is the most efficient way to pitch because it takes roughly the same number of pitches to get an out no matter how you do it, but a strikeout never allows a baserunner.

Sorry, Rex, that myth is...