How the Red Sox Are Limiting Home Runs

The Red Sox caught us all by surprise by jumping out to an incredible start en route to an early-season lead atop the AL East. As Tony Wolfe wrote, the strong performance was largely thanks to consistent run-scoring and a dominant bullpen. Fast-forward to mid-May, and Boston is still there, first in the division and continuing to climb up various weekly Power Rankings (including our own). As impressive as the Red Sox have been at the plate, though, the rotation seems to have been overshadowed. I get it; there’s not a lot of name brand recognition. Their two best starters from 2019, Chris Sale and Eduardo Rodriguez, pitched a grand total of 0 innings in 2020. But after taking three of four from the Angels over the weekend, the Red Sox lead the American League in FIP (3.29) among starting pitchers.

Leading the charge is Nathan Eovaldi. His most recent start extended his streak without giving up a homer to 50 innings; he is the only qualified starting pitcher who has yet to do so. That helps make up for a modest 4.50 ERA and strikeout-per-nine rate of 8.46; his FIP is 2.15. Since he’s been in Boston, Eovaldi has struggled with home runs, allowing an average of 1.86 per nine over the past two seasons. That makes sense, as he’s always allowed a lot of balls to be put in play with a penchant for giving up the occasional dinger. But so far this season, the expected value in terms of xwOBA on those balls suggests that Eovaldi is eliciting softer contact, which is supported by his peripherals.

The barrel rate against Eovaldi is just 4.1%, well below the league average of 8.2%. He’s been much more effective with his cutter this season and also mixed in more of his curveball, which proved effective in 2020 and has remained a strong pitch. On top of that, his cutter has increased in velocity by about 1.5 mph and has less drop (about seven inches in comparison to the ‘20 version), making it more difficult for hitters. It’s become an effective weapon to elicit swings and misses; he’s used the pitch to strike out 10 batters already this season, second to only his four-seam fastball.

Still, no home runs against Eovaldi? Even Jacob deGrom, Gerrit Cole, and Corbin Burnes have all allowed at least one homer. But according to Statcast, it’s not much of a fluke: His expected number of home runs based on batted ball data is 2.1, with a resulting xHR/9 of 0.38.

Eovaldi is not alone in overachieving expectations in the home runs allowed department. Red Sox starters are tops in all of baseball with a 0.65 HR/9. Two of them (Eovaldi and Nick Pivetta) are on pace for a career-low HR/9 rate, and Martín Pérez is posting a rate well below his career average. This may warrant a deep dive into each pitcher’s peripherals, but let’s pump the brakes on that for the moment. There may be a simpler answer: Fenway.

The park the Red Sox call home is generally considered a hitters’ park, but it grades lower than average for home runs. This is especially true in recent years and has been the case in 2021 again: Thus far this season, according to Statcast’s new Park Factor tool, Fenway Park has a home run factor of 72 for right-handed batters and 80 for lefties. That’s really low.

That’s not unprecedented at Fenway, though. In 2014, Fenway had a home run park factor of 73, third lowest in the American League; in that season, Red Sox starters allowed 1.01 HR/9.

To understand better the impact a park has on home runs, let’s revisit Statcast’s xHR metric. Based on batted ball data (namely launch angle, exit velocity, and direction), it projects whether or not a batted ball would be a home run in each of 30 ballparks throughout the major leagues and averages the results to calculate the expected number of home runs allowed. Eovaldi’s 2.1 xHR means that batted ball events against his pitching would result in an average of 2.1 home runs. He’s benefitted from some luck, but it’s also thanks to his home park.

The new baseball has also had a significant impact on home run rates. The HR/FB rate has dropped from 14.8% in ‘20 to 13.3% in ‘21, and at Fenway Park, the HR/FB rate has dropped from 16.6% in ‘20 to 12.3% in ‘21. League-wide rates have certainly gone down, but on the surface, fly balls hit at Fenway seem to be more impacted than elsewhere in the league. And while history suggests Fenway supports a lower-than-average rate for left-handed hitters, the impact on right-handed hitters has been profound, with a huge drop from an average park factor of 98 over the previous five seasons.

To understand the impact of the new ball on fly balls hit in Fenway, let’s refer to Devan Fink’s recent analysis of the drop in homers for certain launch angle and exit velocity combinations — specifically, the bin defined in Fink’s analysis as batted balls with an exit velocity between 100–104 mph and a launch angle between 20–24 degrees, which has suffered the greatest decrease in potential to become a home run. For that category, the home run rate has actually gone up. But it turns out that balls with an EV between 90–99 mph and a launch angle of 30–39 degrees have seen a sharp decrease, with the probability of a homer plummeting from 22.1% in 2019 to just 6.3% in 2021.

Intuitively, when considering the Green Monster, balls will need to be hit higher in the air to get over its 37-foot height. Home runs hit over the Monster have a finer range of launch angles: Where a line drive in most parks would get the job done, in Fenway, a line drive to left could bounce off the Monster and result in a mere single. Earlier this year, Justin Choi weighed in on how the new ball was impacting batted ball results and noted that it had a bigger impact on fly balls compared to home runs:

“The difference between a home run and a warning track flyout is sometimes as small as a few feet. On the other hand, there’s probably not much difference between two line drives set apart by the same distance.”

This finding has some pretty big implications at Fenway, where hitters have to “launch” the ball higher to get it over the wall in left. Case in point: watch the following two clips.

These were nearly identical swings: Christian Vázquez’s double left his bat with a 93.4 mph exit velocity and with a 36-degree launch angle; Xander Bogaerts‘ home run had a 93.3 mph EV and a 35-degree launch angle. Yet one stayed in the park, and the other left it.

Left field at Fenway creates opportunities for a lot of fringy home runs on balls that would be warning track fly outs in many parks. You can see that in Vázquez’s double, as the increased drag of the new ball slowed it down and kept it from going over the fence. And it wasn’t the only one.

The increased drag of the new ball is having a greater effect on the home-run rate in Fenway because these home runs typically have a higher launch angle and spend more time in the air. Does this provide an added home field advantage for the Red Sox? So far, it has played in their favor: At home, they have hit 27 home runs and allowed only 11. Boston will have a hard time continuing to hit two and a half times the amount of home runs as opponents in Fenway, but so far the new ball has played to the team’s strengths.