2021 MLB Home Run Forecast
Baseball has revolutionized the use of statistics. I’d argue they are years ahead of any of the other major sports in terms of using data to make decisions. One of the major players in the statistical revolution is Statcast. Introduced league wide in 2015, it measures and analyzes numerous player and ball activities on the field. Some buzzwords thrown around nowadays — launch angle, exit velocity, spin rate, sprint speed — are all part of an exuberant plan to gain a better understanding of baseball players. Gone are the days of batting average, home runs, and RBIs; in are the days of Statcast data. What I’m going to focus on in this piece is home runs. The ultimate crowd pleaser, what brings the people to their feet. Can we predict the home runs based on Statcast data? Several oddsmakers have also begun releasing odds on MLB awards and accolades. Maybe this analysis can give us some insight on how accurate their odds are on things like predicted home run leader. Do players with the lowest odds actually have data to back up that prediction? Let’s dig in.
Figure 1 contains the odds for the top 30 players from OddsShark. Obviously, a lot of familiar and expected names are listed. I’d say it’s fairly surprising to see the reigning home run king all the way down at the bottom at +3300, but maybe there are numbers to back that up; 60 games, most home runs in a season for his career, it could be several things. Let’s dig deeper into each of these players using some more advanced analytics. Sure, we’ll look at how many home runs they hit last year, but let’s also investigate other avenues that could accurately characterize a hitter’s power.
Since the implementation of Statcast in 2015, I looked at the correlation between home run rate (HR%) and barrel rate (Barrel%) and hard hit rate (HardHit%). For the purposes of this study I used home runs per at plate appearance. You might think, “well guys who walk will have lower home run rates.” Precisely, if you walk more, you are hitting less home runs. For this analysis it is worse to walk, as you are losing precious plate appearances that could be used for home runs. The reason I looked at Barrel% and HardHit% is because they both have to do with exit velocity. HardHit% is entirely contingent upon a ball being hit at greater than 95 mph while barrel rate is contingent upon the combination of exit velocity and launch angle. Yes, the two variables overlap, but only slightly. Both hold a distinct meaning in this scenario. Figure 2 shows the relationship between these two variables and HR%. The sample includes every season in which a player accumulated at least 150 at bats. It’s messy, but it should give you a good idea on how the variables correlate.
Knowing that there’s a strong correlation between Barrel% and HR% and a moderate correlation between HardHit% and HR%, I dug into these stats for each of the 30 individuals listed in the first table. I also added five other hitters who might find themselves in the race for most HRs. With this knowledge I wanted to look at two main ideas: Barrel%/HardHit%/HR% in 2020 and how each of these rates stand the test of time. In other words, how do each of these metrics trend over the last three years? A progressive trend downwards may be the result of a player aging, a tougher home park, or an injury. A progressive trend upwards may be the result of a player entering his prime, changing his approach/swing, or moving to a more hitter-friendly park. I would feel much more comfortable giving a hefty HR projection to someone who has maintained/improved these numbers than someone who is showing signs of decline.
First, let’s peak into 2020. Figures 3A, 3B, and 3C show the leaderboards of each of the three metrics. I listed the top guys and the bottom guys for each category — the solid dark line indicates a chunk that was removed. You’ll notice there are only 34 hitters as Yordan Alvarez didn’t play in 2020.
Some names you expect and some you don’t. You have to remember that this was in only 60 games as well. Since I’m more confident in Barrel% and HardHit% in determining HR%, I really wanted to see how far off the leaders in these two metrics were from their HR%. The analysis of this concept is shown in Figure 4.
The first column to the right of the name shows where each hitter ranked in terms of HR% in 2020. Jose Abreu had the fifth highest HR%; only 4 other players hit more home runs per plate appearance than him. In the next column, I took the ranking of where each player was for Barrel% and HardHit% and averaged them. It’s not shown here, but Fernando Tatis Jr. had the second highest barrel% and the highest HardHit%; therefore his average rank among the 34 players was 1.5. This calculation was performed for each of the players in the study. The last column denotes the difference between the two preceding columns. The darker the color of the box, the larger the difference. Green means the player underperformed the advanced metrics and red means the opposite. In my opinion those that have larger values — their boxes are shaded darker in either direction — will regress more rapidly. Their true home run potential is tied more to the underlying numbers than it is just to the percentage of home runs they hit in a 60 game season. It’s hard to judge Aaron Judge (funny I know) on his discrepancy because he played only a handful of games. However, his battery mate, Luke Voit, has a full season to tell the story. His advanced metrics and HR% present such a strong discrepancy, I’d expect a strong regression in 2021.
There’s obviously more to it than these “advanced analytics.” Both Luke Voit and Aaron Judge had significantly higher HR% at home, where Yankee Stadium is known to have a short porch in right field. However, Giancarlo Stanton had the reverse of these splits, hitting better on the road in his injury plagued campaign. So it’s hard to say exactly how much influence the home field played here, especially when, again, the season was only 60 games.
Before we revisit the HR leader odds for 2021, let’s look at the past three seasons of each of these players. I harped on it several times already, but once again, 1 season — at just 60 games — is not indicative of future performance. People like to think it is, but it isn’t. Since it would take a couple of pages to run through each and every individual in the table, I picked out those that were noteworthy and analyzed them in Figure 5.
Figures 5A and 5B show 16 players — highlighted in green are those with numbers trending up, highlighted in red are those with numbers trending down. Joey Gallo is an exception, but his numbers are so absurd I had to include him. Plate appearances help put some of the metrics into perspective. Take Giancarlo Stanton for example, not a lot of plate appearances in 2018 and 2019, but high Barrel% and HardHit%. Can we trust these numbers in such a small sample size? Maybe, because he’s been a machine for several years prior as well. Some of these guys are not only trending up, but their numbers are also already really high. Miguel Sano absolutely crushes the ball, Joey Gallo was doing just that in 2018 and 2019. Tatis Jr. looks like he’s on the verge of bursting into the top tier. I won’t go through each of the other hitters, but there are a lot of names that should put up some big HR numbers this year. I’d be confident in any of these “green” guys to lead the league in HRs. On the other hand, there are a few spiraling in the wrong direction. J.D. Martinez is aging and I wouldn’t be surprised to see him begin to slow down further in 2021. Nolan Arenado, while carried by some impressive numbers in the mile high city where the thinned air let balls carry like nowhere else, is now situated in a less hitter friendly ballpark. Coupled with his already weak advanced metrics, his HR numbers may decline. As touched upon earlier, Luke Voit, the reigning HR champ, can attribute some of his success last year to luck. With the same Barrel% and HardHit% from 2019, he hit HRs at a clip over twice as high. He’s a powerful guy, but I expect a strong regression from him in 2021. Betts and Torres are a couple others that don’t have the eye popping numbers you’d expect from a HR champ.
So now that we’ve seen all the data, let’s revisit the books back from Figure 1.
The expectation is that Trout has the best chance, then Alonso, then Judge, and so on. My analysis points to a different conclusion. The purpose of this piece isn’t to produce new odds or lock in any one individual to win it, but it does serve as a valuable lesson. Fernando Tatis Jr. is in the middle of the pack, yet he ranked extremely high in Barrel% and HardHit% in 2020, while also showing significant growth over the past two years. His HR total didn’t exactly show that, but with these underlying numbers, it’s hard not to place him near the top. Same goes for Miguel Sano, Jorge Soler, and Giancarlo Stanton. Middle of the pack guys, with top of the pack HR potential. Obviously Stanton would need to stay healthy, but his numbers are off the charts. Similar to Stanton, if Joey Gallo can stay healthy his chances for the 2021 HR title are higher than all of the guys listed above him (even though he is much closer to the top tier of guys than any of the other hitters previously mentioned). Juan Soto has already developed into a superstar; last year we got a glimpse at him unleashing some of his untapped power. It’s hard to look away from his tremendous power potential and at just 22 years old, he can only get better. When looking at some of the guys to really fade — J.D. Martinez, Luke Voit, Nolan Arenado — they seem to be accurately portrayed above. It looks like I’m not the only one not buying into the 2020 hype carryover for Luke Voit, he’s placed comfortably towards the bottom of the list. Finally, while the top guys like Trout, Alonso, and Judge all seem to be the best options based on their past HR performance, their underlying stats suggest they aren’t the best bet to bring home the crown. Especially if you look at the odds that are given, there are several other players that carry a much better chance to win the HR title. We’re less than a month away from baseball. We’ll just have to wait and see how this year pans out.
*All raw data analyzed in this piece was found on: https://baseballsavant.mlb.com/
*Odds were found from: https://www.oddsshark.com/mlb/mlb-home-run-leader-odds
