An Early Look at the New Baseball
I guess it wouldn’t be baseball in these trying times without a debate about the state of the ball. This year’s rendition started in February when Eno Sarris and Ken Rosenthal of The Athletic reported that they obtained an internal memo Major League Baseball had sent outlining changes to the baseball that would “reduce offense slightly in the 2021 season.” Specifically, Rawlings loosened the tension of the ball’s first wool winding, reducing the weight and bounciness of the ball as measured by COR, or the coefficient of restitution.
How would the new ball affect the league’s offensive environment? At that point, we could only speculate. Included in Sarris and Rosenthal’s article is a cautionary tale from the Korean Baseball Organization, which experienced a crash in league-wide offense after minor reductions to the ball’s COR. But the article also cited Dr. Meredith Willis, who believed that because MLB intended to reduce the ball’s COR along with its weight – the KBO actually increased the weight of its ball by one gram – the effects would be less severe. As for MLB, its memo included an independent lab that found minor decreases in fly ball distance with the new ball.
Then came spring training, and along with it the first uses of the new baseball. As March closed out, Rob Arthur and Ben Lindbergh published an article for The Ringer entitled “The New Baseball Still Seems Juiced.” Using data from spring training games, they made two key observations: (1) home run per contact rate had increased, not decreased, from last spring, and (2) the new ball seemed to have a high drag coefficient. “Higher drag should translate to less carry and fewer home runs,” Arthur and Lindbergh wrote. “Yet the higher-drag balls also have a higher home run rate on contact, because they have a substantially higher exit speed.” If the ball’s COR was really reduced, they added, the opposite phenomenon should occur.
And now, here we are, approximately one week into the 2021 season. It’s far too early to draw conclusions about individual players or even teams, but taking a glance at league-wide offensive changes won’t hurt. As of writing, major league hitters have put 4,727 balls into play. By dividing the number of home runs hit so far by the number of balls in play, then comparing it to that of other years, we can see if home run rates are down or up. This is identical to Arthur and Lindbergh’s HR per contact metric; the wording is just different. Here’s how league-wide home run rates have fluctuated throughout the Statcast era:
The results are a bit surprising. Having read The Ringer article, you might have expected to see the home run rate increase or remain stagnant. Yet, as far as home runs are concerned, league-wide offense has declined. If this rate remains constant throughout the remainder of the season, it would represent a return to around 2016 or ’17, seasons with environments slightly in favor of hitters but manageable for pitchers nonetheless. It also aligns with MLB’s goal of limiting offense some but not so much that the league is turned upside down.
So are Arthur and Lindbergh wrong? Not necessarily. It’s undeniable that MLB introduced a higher-drag ball during spring training, but several factors outside of their control may have contributed to the discrepancy. For one, the level of competition in spring games is different than that of actual major league games. Five teams reportedly added humidors to their stadiums this offseason, which could explain why the home run rate has been suppressed beyond what one might expect. Also, batted ball distance is influenced by the lower temperatures of April, as Arthur explained. When our sample is relatively small, even minor factors can have an outsized influence.
But there’s a twist. The rate of home runs is down, yes, but the higher exit speeds indicative of higher-drag balls are still there. They’re most notable in line drives, which I’ve defined as balls hit at launch angles between 10 and 25 degrees. Going back to 2018, here’s how they’ve fared in a few key metrics:
| Year | wOBA | EV (mph) | Dist (ft) |
|---|---|---|---|
| 2018 | .719 | 92.7 | 270 |
| 2019 | .731 | 92.9 | 272 |
| 2020 | .729 | 92.9 | 267 |
| 2021 | .726 | 94.1 | 266 |
An uptick of 1.2 mph? That’s almost certainly a change caused by a new baseball and not home run-happy hitters, who’ve hit line drives with roughly the same might until now. If we compare line drives to fly balls (between 25 and 40 degrees) using the same metrics, more pieces of the puzzle begin to emerge:
| Year | wOBA | EV (mph) | Dist (ft) |
|---|---|---|---|
| 2018 | .522 | 90.1 | 327 |
| 2019 | .575 | 90.5 | 332 |
| 2020 | .589 | 90.8 | 334 |
| 2021 | .560 | 90.9 | 327 |
For fly balls, the increase in exit velocity is negligible. A hypothesis test would tell you that the difference is due to mere variance. Instead, what’s intriguing is the change in hit distance, which is down seven feet from last season. And unlike line drives, fly balls have become far less valuable per wOBA. What gives? It turns out that distance has a much greater impact on fly balls compared to line drives. Intuitively, this makes sense. 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.
Some statistical tomfoolery backs up our intuition. I downloaded two groups of hitters from Baseball Savant. Group one consisted of hitters who hit at least 25 line drives in the 2020 season, and group two consisted of hitters who hit at least 25 fly balls in the same season. I then plotted the relationship between hit distance and xwOBA for each group’s hitters. The results are straightforward. Hit distance explains 57% of the variance in xwOBA on fly balls (r-squared = 0.57), but only explains 18% of the variance in xwOBA (r-squared = 0.18) on line drives. Assuming their exit velocities are equal, an increase in the same hit distance will likely have a more meaningful impact on a fly ball compared to a line drive. Is using the averages of hitters too simplistic of an idea? Possibly, but it’s enough for our purposes.
Based on what we’ve learned so far, it’s reasonable to conclude that the higher speed and shorter distances of higher-drag balls don’t apply themselves equally to every batted ball. Line drives and grounders are primarily influenced by the former characteristic, while fly balls are primarily influenced by the latter characteristic. If so, the drop in home run rate makes a whole lot more sense. The long ball is mainly a product of the 25 to 40 degree ranges of launch angle, and while line-drive home runs do happen, they’re nowhere near as common. I’m still not sure why line drive wOBA has hardly budged despite a higher exit velocity – maybe this is one aspect where we need more time to draw a conclusion.
The pressing issue, though, is that inflated exit velocities on batted balls force us to view the improvements of hitters with a rather healthy dose of skepticism. So far into the season, many players have surpassed their previous max EV highs; to my knowledge, 19 of them did so by a margin of 2 or more mph. But how can we tell which of those power surges are genuine? Did they make mechanical adjustments, or are they beneficiaries of the new baseball? The league’s rate of 110 mph-plus EVs per batted ball events had been fairly consistent since 2015, but this season, that no longer is true:
This dilemma persists even when hitters’ adjustments are easy to spot, as it’s still difficult to distinguish between what’s signal and what’s noise. Sure, there are the changes, but how in the world do you isolate the effects of the ball? It’s maddening to even think about.
Indeed, when the status quo is disrupted, it forces us to make sudden, uncomfortable adaptations. There have been fluctuations in batted ball numbers and offensive environments before, but the fact that we’re alerted to a change this early into the season, one with implications for how we evaluate players, is somewhat alarming. Hitters are under scrutiny, and so are pitchers. In theory, those with a penchant for allowing fly balls are aided by a ball that travels less. The specs could change down the road – this is an early look, after all – but there’s mounting evidence that we’re correct about the gist. Connor Kurcon, for example, has illustrated that the Top 1% of exit velocities around the league are up by around 1 mph regardless of batted ball type. Debates about the official ball have become a sort of annual tradition. But really, they shouldn’t be.
Justin is an undergraduate student at Washington University in St. Louis studying statistics and writing.
Justin…this is a good, useful article that provides some perspective on the sss analysis that a few others on this site have been doing.
However, how does the new baseball impact the pitcher? There is another article posted on the site today that explores fastball velocities for pitchers, but does not comment on the ball. Why shouldn’t the new ball impact fastball velocity?
To my knowledge mass has no impact on velocity, just acceleration, and even if it did I’m inclined to think the effect would be minuscule.
“To my knowledge mass has no impact on velocity, just acceleration…”
And acceleration applied over time leads to a change in ?…
Pitch velocity is measured instantaneously as soon as it leaves the pitcher’s hand, so measured pitch velocities would indeed not be impacted by a change in the ball’s mass!
The ball accelerates based on the force applied by the pitcher, which ends as soon as the pitch is released. Lower mass results in higher acceleration given equivalent force. The point of release is the maximum velocity, which is why we measure it there.
A heavier object is thrown with a lower velocity given equivalent force applied. One can throw a baseball faster than a softball faster than a shot put.
Fortunately baseball exists in conditions for classical mechanics so we don’t need to worry about things like:
https://what-if.xkcd.com/1/
Bingo!
Just take it the the extreme and imagine a baseball sized shot put. You wouldn’t be able to toss that at 90MPH…
Likewise, I can throw a blitzball a lot faster than I can throw a baseball.
It’s math site, so here’s the basic equation. V = Ft/m. Embiggen m and V goes down.
This is what’s happening with the new ball. The mass evidently is down (higher velocity) but the drag is higher (less distance)
That’s a perfectly cromulent equation
That covers mass, but what about drag? Couldn’t it have an effect on breaking balls?
EDIT: it seems that Rob Arthur just wrote about this over at BP today
Good thing the pitcher doesn’t need acceleration or that pesky mass might become an issue.
The pitcher is the one accelerating the ball; *that*, if anywhere, is where the mass would become an issue. The ball does not accelerate on its own after leaving the pitcher’s hand, after all. The only question would be how the change in mass affects the ball’s *deceleration*.
Deceleration does not exist. There is only acceleration. Numbers can be positive or negative.
That’s why he put deceleration in quotes
Once the ball is released mass doesn’t matter. Drag, Magnus effect, and earth become the key players. If a pitcher generates max Force a decrease in mass should create higher acceleration.
So a team with a lineup of a lot of high contact hitters with average EVs previously and a short porch in right field should do better with this ball.
A team of 9 DJ LeMahieus and 5 Matthew Boyds and a bullpen of Emilio Pagans should be good this year.
How does the switch to hawkeye affect this? Would the higher ev effect be explained by hawkeye being a little hotter than the previous system?
Sorry if I missed it, but are you comparing 2021 to *Aprils* of other years, or to the entirety of the other years?
Not sure, but I think entirety, if only because 2020 would be a null datapoint otherwise.
Quick googles found me this: https://www.forbes.com/sites/owenpoindexter/2019/06/28/baseballs-historic-home-run-season-in-one-chart/?sh=49bde33f3399
April is definitely worth splitting out as it’s consistently lower than other months (chart is in HR/FB, as opposed to HR/PA, unfortunately)
Yes, this is a great call out. Would love to see that first graph of HR per BBE that compares 2021 to only April of past seasons. Slugging/power is always lower in April so not a fair comparison to prior full seasons.
I separated April from the full season for 2015-2019. First number is April, second number is full season. There’s no 2020 because there was no baseball in April last year. Also, the 2021 season is up to 4.9% as of this morning.
Formula: HR/(AB-K-SH):
4.9%–2021
5.3%, 5.5%–2019
4.3%, 4.5%–2018
4.6%, 4.9%–2017
4.1%, 4.5%–2016
3.5%, 3.9%–2015
That 2015 season was when the ball really started to fly and it seemed to happen around mid-year. In the first half, the rate was 3.6%. In the second half, it was 4.2%.
Thanks for this. 👍
Very nice, it may be as easy as remembering the ball weighs less and that the power of the players hasn’t changed. The ball’s increased drag coefficient, as discussed, doesn’t stop the ball from reaching higher speeds but allows it to lose its velocity quicker. Think of a badminton shuttlecock. They can reach up to 140MPH and only go up to 30 feet. Drag limits distance, not speed. If the changes are true, the league has used a trick in physics to increases the ever-fascinating Exit velocities while taming the offense a little.
Great for keeping balls from travelling 300, 400, 500 feet.
Perhaps a little terrifying if you’re 60.5 feet away?
next thing we know the pitchers will have full-face cages
I think that was the intent but the reality is inconclusive. Does that mean that Mercedes 485′ shot would have gone 510′ or more with last year’s ball? The number of true tape measure shots seems to have gone up if anything. Stanton, Ohtani, Acuna and even Yermin have put a couple of extra bags of powder in their bats already.
There seemed to be a trade-off that was missed or the original reporting had incomplete info on intentions. An increase in EVs may not have been accounted for with the focus being on limiting distance based on the EVs they had recorded. Or they knew they could increase EV with a lighter ball.
That last graph shows the best evidence. A near doubling in 110+MPH BBE’s!
It will be interesting to see how this new ball sails through the warmer air this summer. One would expect fly ball distance will tick upwards through the year which would negate most of the decline we have seen in the HR rate.
Her name is Meredith Wills, not Willis.
This sounds like an improvement to me. If the change in the ball affects fly ball outcomes more than line drives and grounders (due to the higher EV and greater drag described in the article), it creates relatively higher rewards for hard hits in play over long fly balls. To me, that change reinforces what’s most entertaining and what the game is supposed to be.
What the game is supposed to be is underhand pitching, no gloves, no strike zone, no fences, 100% BIP.
You’re confusing “supposed to be” with “originally.” Nobody would argue that we’re supposed to be keeping track of the years and seasons via the phases of the moon.
I don’t know about “supposed to be,” but otherwise this was my thought as well: this change could work to diversify the game, which I do think would be an improvement.
So my high level take is that the new ball is beneficial for groundball hitters (~ +2mph EV groundball is more difficult to field) and neutral for line drives and fly balls. This should put even more importance on infield defense. I feel bad for Reds pitchers.
I love reading this sort of observational analysis. I think both ball parks and temperature are also drivers into the total umbrella of review. I mean in comparing year to year which is fine, it doesn’t include any of the many variables in addition to temp and where the opening week games were played. So I would think a more micro look would be to compare for example just the teams that perhaps all played home games in the first week. Is there a way for you guys who do such valuable servic3 to drill down on this so we have apples and apples? There is absolutely no criticism intended here by the way. FG is the best.
Citing a stat I posted elsewhere yesterday (& updating it) – there 303 players who had at least 50 PA last year and at least 1 batted ball this year. A full 64 guys (21.12%) already have a higher maxEV this year than last.
What is baseball thinking? The offense is disappearing at an alarming rate with the Three True Outcomes dominating the game so The Lords of Baseball decide that they will make it even more difficult put runs on the board. Until some balance between contact and strikeouts can be reached, a difficult task, then the ball needs every bit of juice it can find. I can say from experience that baseball does not want to see a return to the 1968 style of hitless and runless games. It is interesting, not good but interesting, that through the first week of the season, batting averages are below 1968, .237 in 1968 and .232 so far in 2021.
You beat me to the punch on the TTO and 1968 information. But, I think they were actually trying to liven up the game with the new ball. If offense is defined by the number of runs scored/game, then offense has not been an issue – only how the runs are scored.
My assumption has been that they want to disincentivize HR (more drag) while incentivizing other balls in play like line drives and grounders (higher EV).
I agree based on the behavior of the new ball.
This is in response to both DDD and Anon. Somehow, in some dark back room somewhere, some people came up with that idea but it can’t possibly work as long as it very, very difficult to string a number of hits together. The excessive percentage of outs by strikeout, and the simple difficulty of making contact with 95-100 mph. heat is the real problem and if what would have been home runs last year are only deep fly outs this year I don’t see how that helps the game at all. Hitters can’t magically change their focus from hitting HR’s to hitting line drives towards the gaps. The game is going to be very reluctant to take the only step that might work and that is lengthen the distance from the rubber to the plate. Jay Jaffe, in today’s chat wanted to shrink the zone but that would only lead to more walks and wouldn’t necessarily improve contact rates.
Don’t forget that home runs are one of those three true outcomes, and they’re the only part of the offense that these new baseballs are limiting. If anything, offense on balls in play is being increased, not decreased.
I no more than mention hitless and runless games and Joe Musgrove goes out that same night and pitches the Padres first no-hitter in their existence. Watch out baseball, there is danger lurking behind every 102 mph heater.
If this effect was intentional– and I have to think it is– then I actually do like the trend to favor line drive hitters, because they’re often a joy to watch hit. It might be too early to see yet, but wouldn’t we expect the average launch angle of home runs to decline a little?
Also, which five teams added humidors? I hadn’t seen anything about that before this.
Justin, great writeup! In comparing the SSS of 2021 with previous full seasons on Baseball Reference, there are some interesting comparisons. Thus far in 2021, there are 1.16 HR/G. This most closely matches the 2018 season of 1.15 HR/G and the 2016 of 1.16 HR/G. I’m guessing the rate will likely increase as the season progresses to more closely match the 2017 (1.26 HR/G) and 2020 (1.28 HR/G) seasons. The current ERA (4.14) and R/G (4.55) also most closely match 2018 (4.14 and 4.45) and 2016 (4.18 and 4.48). But, where it gets very interesting is (and I realize we’re looking at a SSS) the current season has 9.7 K/9 (by far the most ever), 3.6 BB/9 (the most since 2000), and 7.9 H/9 (the least since… the infamous 1968 season). It appears TTO is still alive and well.
If line drives go further/faster and fly balls don’t go farther, this achieves the league’s goal of fewer home runs and incentives more doubles and triples, no? Probably not enough to move the needle yet, but on the right track.
Looking forward to seeing how things shape up once we get into the summer months and the dataset gets bigger.
Is it possible that players are hitting the ball harder and throwing the ball harder because they’ve only played 60 games in the past 365 days? That an abbreviated system gave more rest time and players returned to camp stronger than ever before?
The SABR Science and Baseball Committee (which I chair) has been talking about the game’s aesthetic problems and possible fixes, including alterations to the ball, for several years.
First of all, reducing the weight of the ball increases EV off the bat, while reducing the COR decreases it. Since the two changes are working at cross-purposes, they pretty clearly came necessarily bundled in the change to the winding. The net result was an increase in EV, and we have to assume that was intended. That could be expected to increase BABIP on liners and grounders, which all things being equal would increase the number of hits in play. That would increase offense further, so the drag on the ball was decreased to reduce home runs. That seems to shift the game in the desired direction; more multi-hit rallies, fewer skill infielders hitting opposite-field homers. (BABIP is extremely dependent on game temperature, it seems, so we won’t get a good handle on whether it has actually been increased until the end of the month.)
It’s all wrong, however.
Making the ball lighter can be expected to increase pitch velocities, which means more strikeouts, which reduces the number of balls on play of all sorts. (The rise in strikeouts in the game can be explained entirely by higher fastball velocity and lower fastball frequency.) And pitch velocity is indeed up, very markedly. And SO% is 25.3 % so far, versus 23.5% in April 2019. The increase may well more than offset any gain in BABIP.
Furthermore, MLB didn’t try to figure out what the actual problems with the game are. At the last SABR conference (2019), in our committee meeting, i presented a regression model for attendance from 2004 to 2018 with an adjusted r^2 of .92. Competitive imbalance (measured by the kurtosis of team winning percentages, 57% previous year, 43% current year) accounted for about 40% of the model. The other three factors were contact percentage, home runs per contact, and SO as a percentage of SO + BB, all for the previous year. Dick Cramer gave it a thumbs up.
The thing that’s killing baseball is the decline in contact percentage. The only way to get back to the game we grew up with is to cut down strikeouts. This change is going to increase them.
You can’t shrink the strike zone because walks are even worse than strikeouts; i.e., the SO / (SO + BB) factor actually is a positive one. A strikeout can be exciting for the fans of the pitching team. You don’t see anyone putting up “W” cards in the stands.
The one change that helped attendance? More homers. Relative to the modest but significant (p < .02) downward trend in HR/Contact from 2003 to mid-2015, extra homers increased attendance by about 1,200 fans per game in 2017 and 1,750 in 2018. As I noted in the handout to the presentation, "Even though home runs represent a very high percentage of online game highlights, I think it’s a mistake to think that the home run itself is what puts more fans in the seats. The possibility of home runs makes games more engaging and exciting because they make every game closer. If your team is down three runs late in a game but has two runners on base, your ratio of optimism to despair is entirely driven by how likely you think a home run might be." (There was, however, no interaction between this factor and the declining contact rate, which is to say, the home run "hope" factor did not get bigger as it became more difficult to string together a rally with multiple hits within the yard.)
So reducing homers, another result of the changed ball, is also bad for the game's appeal.
Our plan was to finish this work at last year's convention and then publish a paper with our recommendations. I have a plan B I'll be implementing soon.
Eric, thank you very much for this information. In your second paragraph, you said “so the drag on the ball was decreased to reduce home runs. ” Should that say “the drag on the ball was increased… ?”
Regarding why there has been a decline in attendance and why more HRs helped, I think the most significant reason is that the younger generations want continuous external stimulus and more immediate gratification (primarily due to what electronics provides them with gaming, chatting, etc.). Baseball doesn’t offer continuous stimulus as much as most of the other sports. Sadly, I’m not sure there is a good answer to fix that without fundamentally changing the game itself – and that would be tragic. I don’t want baseball to turn into beer league softball. Nor do I want to see it turn into wiffle ball.
This is what needs to be discussed. You are right in every way. The game needs more contact!!!!! There is no other solution. The game I grew up was full of contact, I believe Sandy Koufax and Nolan Ryan are the only pitchers, with more than1000 innings, whose career began before 1980 that averaged over 1K/inning. Sam McDowell was very close. Home runs were special, not the only way to score. Base-stealers electrified the crowd, hit and run plays filled the game with excitement. Sadly, after 120 years, the players have outgrown the dimensions that worked so brilliantly for all those years and baseball’s hierarchy has to recognize it soon. There appears to be a real possibility that batting averages crater this year. It is clear that very few human beings can consistently make any type of contact with a baseball thrown at 97 mph from 60’6″ away but that doesn’t even take into account the increasing height of the pitching fraternity. Tyler Glasnow releases the ball at a distance of 53’6″ from the plate. Hitters may be bigger and stronger but they still have no chance. As I write this, I just watched, Conforto, Alonso and Smith, the best the Mets have, not even hit a foul ball as they all fanned. The score is 1-0 in the 8th with 25 K’s and a total of 7 hits. There is no more dedicated baseball fan than me and this is %^&*%^!!! The time has passed for something to be done and pitchers are going to have to come to terms with the fact that velocity is destroying the game and accept an increased distance between the rubber and the plate. There is no other possible solution to velocity.
bosoxforlife, I strongly agree the game needs more contact to improve the quality of entertainment, and I think most baseball fans agree. However, I don’t think it will significantly grow the number of fans. The reason is as I stated above and it is, unfortunately, a sign of the times. Nonetheless, I’m sure it would help, especially if it were combined with a robust marketing strategy to encourage our youth to play baseball.
DDD. Baseball will never be the only game in town as it was when I was growing up but there will always be a place for The Summer Game, unless the Ivy League fools who run the game kill it. There is nothing that brings a city or even an entire region together more than a thrilling pennant race with the home team in the hunt. Soccer, despite all attempts to try to stuff it down America’s throats, is a poor excuse for entertainment. The only reason it is played throughout the world is that even in the poorest countries there are fields and a ball can be found anywhere. There is a lot of nothing for 99% of the time with virtually no scoring and very few opportunities to score. Yes, baseball has been shooting itself in the foot for quite awhile and needs a bold move now.
I worry that with the increased exit velocities, soon a pitcher will be hit by one of these line drives. Having pitched and been hit a couple of times(thigh fortunately), and one time almost hit, by a ball I never saw but felt go by my ear, it is something I worry about.