The World Series kicks off Tuesday, a final hurrah before the 2019 baseball season and a rather enjoyable postseason come to a close. The Astros and Yankees, in particular, put on a very compelling series that was ultimately decided by three fly balls. The first was Jose Altuve's Game 6 home run. Obviously. The second was Carlos Correa's walk-off in game two. The third one was not a home run, although it probably should have been.
Trailing 2-0 in the fifth inning of game three, Didi Gregorious launched a fly ball deep to right that looked good off the bat, but died at the wall:
We watched fly balls hit like that leave the yard all season long, but Gregorious fell a few feet short. This was not just one odd non-homer, either. The home run explosion came to a screeching halt this postseason.
In the regular season, there were 9,290 batted balls Baseball Savant qualified as "barrels." These are balls that were hit with a high exit velocity and a quality launch angle. Of those 9,290 barrels, 5,536 turned into home runs, a rate of 59.6%. In the playoffs, just 66 of the 131 barrels have been home runs, a rate of 50.4%. Barrels do not take into consideration the quality of the pitcher on the mound. Yes, the sample size is small, but the stage is large, and a near 10 point drop is telling. By Baseball Prospectus author Rob Arthur's estimate, by Oct. 10, there were 43 home runs hit, when in the regular season, there would have been 67.
And it's not just because of the October weather. Yes, fly balls do not travel as far in colder weather, but not by that much. With every 10 degree increase in Fahrenheit, the ball travels an extra 2.5 feet. The Cardinals' analytics department concluded the ball is traveling 4.5 feet shorter than it did in the regular season. So while the weather is playing a small role, it does not explain the drastic drop-off.
It's indisputable the ball was altered again. Commissioner Rob Manfred insists they are the same balls as the regular season, but they aren't performing the same. The term "juiced baseballs" has been thrown around these last few years, so it does not hurt to explain why they are "juiced." In 2018, the MLB Home Run Committee found the drag of the ball decreased in 2016 and 2017. One of the main reasons why was the laces were thicker, making the ball more spherical and more aerodynamic. There was even less drag on the ball in 2019, and Dr. Meredith Wills deduced it is because the seams are lower and the texture was smoother.
But the ball has had more drag in the playoffs. In fact, the ball has not grabbed this much air at any point since 2016. From Arthur:
Updated drag chart current through yesterday's games. No real reversion. Shaded area is now the 95% confidence interval for each week (derived from bootstrapping). You can see why I said there's a one in a million chance the balls are the same--the intervals are very narrow. pic.twitter.com/Gibc4AUN49
— Rob Arthur (@No_Little_Plans) October 15, 2019
Michael Baumann of The Ringer wrote about how a dejuiced ball would impact hitters across the league. Let's take the same approach from a local angle. What would it mean for the Pirates if the ball was dejuiced? This is going to be a two-part Mound Visit, focusing on just pitchers today. Tune in next time for the hitters.
It is safe to say the juiced ball hurt the Pirates last year. Yes, the other 29 teams had to pitch with the bouncier ball, too, but the Pirates entered 2019 hoping to be carried by their pitching. Instead, it was their detriment. They allowed 241 home runs, a franchise record. Some of that is due to them hanging onto the past mentality of pitching to contact, some of it was due to injuries, some of it was regression, and, of course, some of it was because they trotted out an abysmal middle relief core. But overall, their pitchers allowed a fairly normal amount of fly balls. The difference was those fly balls left the yard at a higher than normal rate.
According to FanGraphs, Pirates pitchers had a home run to fly ball rate (HR/FB%) of 15.7%. That was the fourth highest percentage in the National League, trailing just the Rockies and Reds, who play in bandboxes, and the Phillies.
Now HR/FB rate is not a perfect tool for looking at individual pitchers since it does not take the quality of those fly balls into consideration. A pitcher who gets a lot of cans-of-corn is not "outperforming" his metrics, and a pitcher who allows a lot of hard contact is not necessarily "unlucky" because his HR/FB rate is high. In a lot of ways, it is the pitcher equivalent of BABIP. Still, when you look at the players currently on the Pirates' roster who pitched at least 25 innings in 2018 and 2019, it is clear to see almost everyone's FB/HR% went up:
Keone Kela pretty much held steady from 2018, and Francisco Liriano improved a bit, though that is mostly because he went from being a starter to a reliever. Everyone else did worse.
But what if they used the postseason ball all season? FanGraphs and Baseball Savant qualify fly balls differently, but they both had the amount of fly balls and pop-ups in the regular season result in a 15.3% HR/FB rate. In the postseason, Savant measures the HR/FB rate at 12.9%, 2.4 points lower. That is right in line with the league's 2016-2018 percentages, according to FanGraphs, and Pirates pitchers allowed 1,532 fly balls, per FanGraphs. If their home run rate decreased at the same rate as the rest of the league, their 15.3 HR/FB% would drop to 12.9%.
234 of the 241 home runs the Pirates allowed last year came on what FanGraphs considered a fly ball. With a playoff baseball, that total would have been 198 instead. Assuming those other seven non-fly ball home runs stayed home runs, the Pirates would have finished with 205 allowed on the year.
It gets a little fuzzy to see how it would impact individual pitchers, but it's worth taking a shot. There are five pitchers who I feel had a large enough sample size of fly balls last year to take a stab at seeing how they would have performed with a different ball: Trevor Williams, Chris Archer, Joe Musgrove, Steven Brault and Dario Agrazal. Williams allowed a team high 27 home runs. By reducing his HR/FB rate 2.4 points to 12.1%, he would have allowed 22.5 home runs with the new ball. Archer goes from 25 to 22. Musgrove from 21 to 16.6. Brault from 15 to 12.3, and Agrazal from 15 to 12.6. We won't get into how many total runs would be saved because that can quickly snowball out of control, but roughly 45% of all runs last year came on home runs. So this may only be a handful of home runs each, but it would make a large impact over the course of the season.
Now that goes on the assumption that all fly balls are created equal, which is not the case. Baseball Savant's batted ball data is a little deeper, so let's go the extra step. Last year, 6,378 of the 6,776 home runs hit across the league were on what Baseball Savant called "barrels" or "solid contact." Here is how often those batted balls turned into home runs and how Pirates pitchers compared to the league averages:
The Pirates allowed 323 barrels and 245 solid contact hits in 2019. Let's assume their barrel HR% also drops 9.2 points to 49.9% and their solid HR% goes down 4.1 ticks to 5.7%. If that was the case, they would have allowed an estimated 30 fewer home runs on barrels and 10 fewer on solid contact. That would bring their team total from 241 to 201.
So either way you cut it, the Pirates would have set a new franchise record for most home runs allowed in a season, but it would have been much closer to previous season totals. Some of the blame is due to regression from key players and some is on injury, but now we have a rough estimate on how big a role the ball played.
A dejuiced baseball would not solve all of the Pirates' pitching woes, but it would certainly help. That leaves just two questions: How would it impact the hitters, and would it be good for the team overall?
Data courtesy of FanGraphs and Baseball Savant.
MORE MOUND VISIT
Oct. 15: How Kuhl could break big
Oct. 10: The lesson from Cole/Glasnow
Oct. 8: Ex-Pirates of the postseason
