Home Runs Were Down in April, but by How Much?

Seemingly in the blink of an eye, a month of baseball is behind us. With nearly 30,000 plate appearances taken and more than 18,000 batted balls put into play, a month of data is plenty to begin examining league-wide trends and to make some predictions for the rest of 2021.

One big question going into this season — and a topic already examined here by both Ben Clemens and Justin Choi — was what the impact of the new baseball would be on the overall offensive environment. As both Ben and Justin found and detailed, the new baseball is bouncier, yielding higher exit velocities than in years past, and also possesses more drag, as it is not traveling as far. I want to focus on that second point. If the ball isn’t traveling as far, we should be seeing fewer home runs hit in 2021 — and we are. But can we pinpoint just how many home runs will be hit this season? That takes a bit more guesswork, but before getting into that, let’s first see how April 2021 stacks up to prior seasons, as well as identify where exactly we lost those home runs.

Home runs in April 2021 were down compared to 2019, at the difference of 0.45 percentage points. Raw home run totals are at their lowest levels since April 2017, though some of this change can be attributed less to the baseball and more to how the game is played. With fewer batted balls, there will be fewer home runs hit, regardless of the composition of the baseball. We can see this using one simple metric: the percentage of plate appearances ending in a batted ball.

So to consider the impact of the baseball on home run totals, we need to look at home runs on a per-batted ball basis. But to consider just how many home runs will be hit this season, we also need to evaluate the rather significant decrease in the number of balls that are being put into play overall. Brendan Gawlowski touched on this phenomenon in discussing the scarily low league-wide batting average, but it’s important to reiterate: the fewer balls put into play, the less offense we should experience overall. (Walks do counteract this a little bit, but strikeouts have increased at a faster rate than walks.)

Before diving deep into the numbers and making predictions for the 2021 home run total, let’s take a look at where we shaved off the 0.45 percentage points of home run-per-batted ball rate. To do this, I scraped every fly ball hit in April 2021 and April 2019 and, similar to what Ben did in his study, binned them by exit velocity and launch angle. Then I calculated the probability of hitting a home run based on each exit velocity-launch angle combination and put them into the following two heatmaps. (Note that there are some small samples in some of these bins, but the general trend is what’s important.)

First, April 2021:

Now, April 2019:

There’s a lot going on here. To distill it down, it’s better to evaluate the changes by looking at a comparison plot. This third heatmap shows the difference in home run probability in April 2021 compared to April 2019:

As you can see, we’ve lost a lot of home runs in the 100–104 mph, 20–24 launch angle degree group. In April 2019, these types of events went for home runs 58.7% of the time. In April 2021, that’s down to 24.8% — a 33.9-point drop. And it’s not as if these batted balls are rare: In 2019, 121 fly balls fit these characteristics, but in April 2021, 101 did. That’s roughly the difference of 30 home runs right there.

As you’ll notice, though, the majority of the change comes at lower launch angles, which makes sense, as these would have less carry over the wall. A baseball with more drag would, in theory, affect these types of events the most. In 2019, April fly balls that were hit below 30 degrees went for homers 45.4% of the time; in 2021, it’s just 35.6%. A change of nearly 10 points is significant considering that over 1,000 fly balls were hit in that range in each year; that represents a difference of more than 100 home runs in 2021 compared to 2019.

Now that we know where we’ve lost the homers, let’s consider the second part of my initial inquiry: How many home runs should we expect to be hit this season? Looking at data from 2015 to ’19, the April home run-per-batted ball rate was, on average, 0.3 percentage points short of the full-season figure. (It seems as if there has been less of a difference in recent years, making me wonder if weather plays a somewhat smaller role in suppressing home runs with a more juiced ball, but I don’t think we have enough information to conclude that.)

It’s a simple method, but after applying that average to the 2021 returns, we can predict that the home run-per-batted ball rate would be roughly 5%. This would fall short of the crazy 2019 season, when it was 5.4%, but it would still be noticeably above all other seasons from ’15 to ’18. (The 2020 season will forever remain an outlier, but it’s worth noting that the rate last year was 5.24%.) Even if you apply the largest difference between the April and full-season home run rates — a 0.39 percentage point jump — 2021 would still fall short of both ’19 and ’20 but above ’15 through ’18.

But effectively evaluating just how many home runs we will see this season is not just about the home run-per-batted ball rate. We also need to consider what percentage of plate appearances end in batted balls, another stat that has remained relatively stable from April versus each full season.

Here, too, the difference is pretty small, with an average bump of roughly 0.56 percentage points. Unlike the home run data, where there are increases from April to the full-season rate in the entire dataset due to weather, this difference looks more random. For our purposes, let’s just apply the average jump to 2021 in BBE/PA%.

Putting it all together, at a 66.91% BBE/PA rate and a 5.0% HR rate, we would expect roughly 6,240 home runs hit in 2021 if hitters received the same number of plate appearances as they did in 2019. (This assumption may not hold true.) This would represent a roughly 8% decrease in raw home run output, but it would also rank second in the homer-happiest seasons in baseball history.

So while home runs were down compared to our most recent season’s worth of April data, we’re still seeing balls leave the yard at a brisk pace. For just the third time in baseball history, we could see another season of 6,000-plus home runs, but as we’d expect with a ball with more drag, we’re losing some home runs in spots with lower launch angles and exit velocities.