Curveballs Are Underutilized Early in the Count

I got the idea for this article thinking about pitching strategy. It makes sense to me that getting to two strikes for a pitcher is an important strategy for good performance. With two strikes, pitchers can get a hitter to swing out of the zone and either make bad contact or miss completely, two of the best possible results for a pitcher. The problem is, how does a pitcher get there without getting knocked around? If a pitcher throws a meatball down the middle in order to get early strikes, good hitters may take advantage and hit the ball hard before the pitcher can get to that good situation. So if a pitcher can throw a strike early, and maximize the chance a hitter chooses not to swing, that seems like the most effective strategy to get to this situation. The research below suggests that if this is the case, throwing a curveball high in the zone early might be a great strategy that almost no one uses.

I initially looked at first pitches going back from the beginning of 2016. I wanted to see which pitches had the highest swing rates on 0-0 counts. I was fairly certain that we would see fastballs with the highest swing rate. To my surprise, changeups have the highest swing rate, despite the lowest zone rate. Curveballs had the lowest swing rate. Below is the breakdown.

Changeup: 34%

Fastball: 29%

Slider: 29%

Curveball: 18%

The changeup swing rate suggests a well-placed changeup on the edges or out of the zone can be a good pitch to throw on the first pitch on occasion. However, with a curveball, you can throw it in the zone and not get a swing a large amount of the time. Given a pitcher’s goal to get to two strikes, the most advantageous count state for him, throwing first-pitch curveballs seems like a smart idea. However, this is not the strategy we generally see from pitchers. Below is percent of pitches thrown on first strikes.

Fastball: 60%

Slider: 14%

Curveball: 9%

Changeups: 7%

These frequencies suggest why changeups are so effective at getting swings out of the zone on 0-0 counts. Pitchers overwhelmingly throw fastballs early in counts, so when the changeup comes, it is very hard to distinguish it from the fastball, which a hitter will expect most of the time.

There are some practical reasons why pitchers throw mostly fastballs on 0-0 counts. First off, they are much easier to command, and as stated earlier, throwing in the zone and getting to two strikes is the main goal for a pitcher early in the count. Offspeed pitches, on the other hand, tend to have much more movement and can be harder to locate. Second, swings and misses aren’t a big deal without two strikes. Fastballs tend to have higher contact rates than offspeed pitches, but contact rates are much more relevant when whiffs lead to strikeouts.

But there are a few reasons why it makes sense for curveballs to be a go-to pitch early in the count. Some pitchers do locate the curveballs very well. Rich Hill is a great example. He famously throws his curveball about 50% of the time, throwing in the zone about 55% of the time the past three seasons. Throwing his curveball so often is probably why hitters swing so little against Hill despite his incredibly high rate of throwing the ball in the zone. Throwing his curveball, especially early in the count, may be a big reason behind Hill’s resurgence.

My next piece of research was looking at pitches high in the zone. I hypothesized that when pitches are located in the part of the zone that moves opposite to the pitch’s movement, hitters would swing less. For example, curveball breaks sharply downward, so a curveball high in the zone will look out of the zone to the hitter, therefore garnering less swings. I think this is logical and probably a well-known concept, but it was something I had never looked into.

I looked at all pitches thrown in the upper third of the zone on non-two-strike counts. Separating out curveballs and non-curveballs, the swing rates were vastly different.

Curveball swing rate: 26%

Non-Curveball swing rate : 65%

The results were overwhelming. There is nearly a 40% difference in swing rate between curveballs and non curveballs high in the zone. Hitters swing a lot high in the zone in general, but with curveballs they barely swing at all.

Very few pitchers utilize high curveballs without two strikes. The ones that do are a mix of bad and good pitchers. Of all pitchers who threw more than 200 curveballs on non-two-strike counts, Carlos Martinez had the highest percentage in the upper third of the zone, 15.3%. Hill is up there as well at 12.7%. But so is Paul Clemens at 14.6%, one of the worst pitchers in baseball. Jake Arrieta was the lowest at 3%, and he’s one of the best.

Early in the count, changeups and fastballs tend to have high swing rates, while curveballs tend to have low ones, especially high in the zone. Pitchers mostly use fastballs early in the count, but sparsely curveballs. While it makes sense to throw curveballs low with two strikes in the count to get swings and misses, this research suggests that a high curveball is an underutilized pitch early in the count.