NEW YORK -- Baseballs really have been getting extra lift since 2015, and it's not from the exaggerated uppercuts batters are taking, according to a 10-person committee of researchers hired by the commissioner's office.
"The aerodynamic properties of the ball have changed, allowing it to carry farther," said committee chairman Alan Nathan, professor emeritus of physics at the University of Illinois at Urbana-Champaign.
But the panel, which includes professors specializing in physics, mechanical engineering, statistics and mathematics, struck out trying to pinpoint the cause.
The committee's 84-page report was released Thursday by Major League Baseball. There was no evidence of meaningful change in the bounciness of the balls, formally called coefficient of restitution (COR), or alteration in batters' swings, such as uppercutting.
As for what caused of the change in aerodynamic properties, it remains baseball's great mystery, the sport's equivalent of the search for the Loch Ness monster.
"We have to admit, and we do admit, that we do not understand it. We know the primary cause is the change in the drag but we just simply cannot pinpoint what feature of the ball would lead to it," Nathan said during a conference call Wednesday, ahead of the report's release. "Therefore it was probably something very, very subtle in the manufacturing process, but again it has to be pretty subtle, because if it weren't, we would have found it."
Physicist Leonard Mlodinow, in an executive summary accompanying the report, speculated "manufacturing advances that result in a more spherically symmetric ball could have the unintended side effect of reducing the ball's drag."
The major league average of home runs per game for both teams combined climbed from 1.90 before the 2015 All-Star break to 2.17 in the second half, then rose to 2.31 in 2016 and a record 2.51 last season. The percentage of batted balls resulting in home runs rose from 3.2 percent in 2014 to 3.8 percent in 2015 to 4.4 percent in 2016 and 4.8 percent in 2017.
"We found a consistent picture that the drag coefficient is a little bit smaller as we progressed through 2015 into '16 [and] into '17," Nathan said. "Finally, we used our physics expertise to conclude that the small change we found in the average drag coefficient going throughout the period 2015 to 2017 was completely consistent with the change in the number of home runs per batted ball.
"So you're using partly pure physics, partly a model, partly statistical data about home run distances and things like that, but it all hung together very, very well. So all four of those things point to the fact that it's the aerodynamic properties of the ball that have changed. So that much we know. What we do we not know? Well, what we do not know is what specific measurable property of the ball has led to this change."
The committee inspected the Rawlings factory that manufactures the balls in Turrialba, Costa Rica, and analyzed test data from 2010-17 compiled by Rawlings and the University of Massachusetts Lowell, which has analyzed balls for MLB. The group tested 15 dozen unused balls from 2013-17 and 22 dozen game-used balls from 2012-17. The committee devised new tests conducted at Washington State and examined MLB Statcast data from 2015-17 that included pitch type, exit velocity, launch angle, spray angle, spin rate, spin axis and distance.
MLB announced five steps in response to the report:
Monitor temperature and humidity of ball storage areas at all 30 ballparks this year and work with the committee to determine whether to mandate humidors throughout the major leagues in 2019;
Update production specifications with Rawlings and add specs for aerodynamic properties;
Develop aerodynamic tests;
Create standards for mud rubbing, to be enforced by the umpires;
Form a scientific advisory council.
Balls have been stored in a temperature- and humidity-controlled environment at Denver's mile-high Coors Field since 2002 and in the desert at Phoenix's Chase Field, starting this season.
The Official Baseball Rules state balls must be 5-5.25 ounces and 9-9.25 inches in circumference. Major league balls have rubber pills at the center, wound over by three layers of yarn that is 85 percent wool and 15 percent synthetic, and then a thin layer of cotton. The cover of hide from Tennessee dairy cows is hand-sewn with 108 stitches.
"Rawlings makes baseballs with a much, much, much tighter spec than they are required to do by the actual spec itself," Nathan said. "So we recommended altering that and tightening up the spec, and so that when you say the ball is within spec, it has some meaning to it, and they followed that recommendation."
Application of the Lena Blackburne Original Baseball Rubbing Mud, which comes from the New Jersey side of the Delaware River, was not examined. The mud is used by clubhouse attendants to make the balls less slippery.
"There could be some non-uniformity there," Nathan said. "One of the things that is known to affect the flight of the ball, the carry of the ball, is the roughness of the surface of the ball. That's why the seams matter, but also the leather part, the white part matters, too, and differences in how that mud is applied could possibly provide a clue to it."
Nathan would like additional tests on surface roughness and whether pills are off-center.
"There are some smart people who are looking into this drag business, and the hope is that it will be uncovered and we will understand things better," he said.
In addition to Nathan the committee included Bowling Green statistics professor Jim Albert; Southern California mathematics professors Jay Bartroff and Larry Goldstein; Stanford school of humanities professor Roger Blandford; MIT mechanical engineering and mathematics professor Anette (Peko) Hosoi; CalTech mathematics professor emeritus Gary Lorden; Washington State mechanical and materials engineering professor Lloyd Smith; Dan Brooks of the Brooks Baseball website; and Southern Cal Ph.D. student Josh Derenski.