UNC professor offers his sports-science research to Major League Baseball to help project the next great player
By Nate Haas ’04
Numbers crunchers rejoiced when bestselling-book-turned-Hollywood-movie Moneyball pulled back the curtain on a major-league baseball team’s use of undervalued analytics to identify players to covet.
Since being pioneered by Billy Beane — the Oakland A’s general manager who became a household name after Brad Pitt’s portrayal of him in the biopic — the practice of selecting players based on once-overlooked data is no longer a novelty.
UNC Professor Bob Brustad attests to this. A friend of his works with the Tampa Bay Rays, which now employs 11 full-time analytics gurus.
“Teams see the value in this approach,” he says. “It drives every part
of the decision-making process.”
In a way, Brustad is working on the next chapter of Moneyball. His sports-science research is largely an untapped resource among teams.
With a background in sport and exercise science, Brustad intends to fill that void. In 2016, he presented his initial findings at Major League Baseball’s annual analytics conference, attended by over 300 front-office personnel and scouts.
His presentation, focusing on talent-projection errors based on players’ ages when they were drafted, led to at least three teams expressing interest in working with Brustad.
“I believe the prediction equation we use in projecting talent is wrong — not fatally wrong, but it’s distorted,” Brustad said during his presentation.
An aficionado of the game, Brustad wants to help teams project the success of players they’re considering, but he also seeks to improve sports at all levels for athletes, coaches and families.
Student of the Game
Brustad learned math from his father, a mathematician and a sabermetrics disciple long before Beane and Moneyball made it de rigueur. He recalls his father’s love of statistics and projecting outcomes of games, especially for college football’s venerable Rose Bowl.
“He would calculate score-point differentials based on common opponents of common opponents of common opponents,” Brustad says.
It will come as no surprise then that Brustad’s first lessons involved learning how to figure batting averages.
That kindled what has become a lifelong passion for baseball.
His research background in how player development and sports psychology affect success puts him squarely on the path of his current focus.
Brustad reviewed widely available draft data of high school players from the first 20 rounds from 2005-12 to look at how age influences the probability of being selected in the draft. He set out with the expectation that younger high school players might provide greater value in the long run.
In his presentation at the MLB conference, two players drafted in 2007 served as Exhibit A and Exhibit B.
The first, a 19-year-old “can’t-miss prospect” who was drafted in the first round by the St. Louis Cardinals. And the second, an 18-year-old pitcher taken in the later rounds by the Colorado Rockies.
The anecdotes serve as a cautionary tale for teams and support the initial results of the study.
The Cardinals’ pick, Pete Kozma, showcased his dominance as an older high school player who had accumulated more practice time. A top 20 prospect, he played two years for St. Louis before being optioned to the minors. He’s now a backup infielder with the New York Yankees.
On the other hand, the Rockies’ pick, Chris Sale, received much less fanfare, ranking outside the top 1,000 prospects in the draft class. His fastball, a key measure teams use in draft analysis, reached 86 mph (over 8 mph below the average of a major-league pitcher). Brustad says that not only was Sale young for his age group, but he also had a less mature body type and he focused more on basketball in high school. After opting to go to college, he was re-drafted in 2010 by the Chicago White Sox as the 13th overall pick. Last year he made the All-Star team.
The “Prediction Equation” is Inappropriately Weighted
Current Prediction Equation
Current Ability (High School)
Overvalued (due to age effects)
Maturation (physical, psychological)
Learning (amount and quality of practice; feedback quality; self-regulation)
Projected Performance (MLB)
Which Players will Outperform their Talent Projection?
During his presentation at the Major League Baseball conference, Brustad said these are variables to consider based on the data he analyzed:
“Players who are going to grow the most from the time they get drafted until they play at their major-league age,” Brustad says. “When a player reaches maximum rate of growth tells you a lot about growth potential past that point.” Body type is also helpful in predicting ultimate growth.
“For pitchers what we want to see is... the physical development and change in limb length in particular,” Brustad says. “Players who have the greatest growth would be the ones who would seem to have the greatest potential for change over time.” As for position players, “we’re looking at power more than anything else. … we’re looking at strength.”
“Some players are simply going to learn more than others,” Brustad says. “Learning corresponds a lot to practice history so the more quality practice a person has had, the less learning potential they have. The less specialized the athlete was at a younger age, the steeper the learning trajectory would logically be.” Other individual differences that affect learning would include the player’s adaptability and receptiveness to feedback.
“Some athletes are more capable of psychological regulation, in terms of emotional control, self-regulation, etc. than others,” Brustad says. ”That’s going to play a role because baseball is a game of frustration … it’s a very psychologically demanding process.”
Source: “Hidden Gold on the Diamond? The Contribution of the Relative Age Effect to Talent Estimation Errors of High School Players in the June MLB Draft” by Robert Brustad
Sale clearly outperformed his talent projection from the 2007 draft, and Brustad says he’s an example of the importance of taking growth and maturational development into analysis.
“It’s remarkable how much growth can take place between ages 17½ and 18½,” Brustad says. “The more specialized player looks better now, but he’s closer to his ceiling. The players getting greater attention are more likely to be early maturers who have less growth potential.”
When he factored in a statistic known as WAR, a formula for projecting the number of wins a team has with a specific player in the lineup as opposed to a replacement, Brustad discovered that many of the top 20 players in today’s game were drafted as 17-year-olds.
In general, Brustad’s research shows that teams overvalue players’ current ability.
“It’s a tendency we see across all sports that we make some fundamental errors in the talent evaluation process because we neglect the importance of age, maturity and practice,” Brustad says. “We’re missing a lot in terms of bringing in the maturational, learning characteristics into our projection system.”
Some players, like Washington Nationals’ ace pitcher Stephen Strasburg, considered to be one of the game’s greatest pitching prospects, aren’t even drafted in high school. Worse yet, the risk remains for teams to completely miss major-league talent because a player can be near the beginning of the development curve at age 17.
More work needs to be done, but Brustad allows this: all things being equal, if a team has a choice between a younger draft pick or an older one, the choice is clear.
Brustad wants to expand the draft analysis another 5-10 years and explore other related factors in a player’s development. Specifically, he intends to focus on sport specialization and maturity. If he can get his hands on retrospective data — showing the age when a player reaches physical maturity — he’ll be able to make even more accurate projections for their growth potential.
“Equations can be developed to predict who has the greatest physical growth potential based on height and weight and somatotype (body type) way back when,” Brustad says.
As for specialization, his hypothesis is that the more a player focuses on one sport growing up, the more likely that player is closer to maxing out at the growth ceiling.
“There’s some evidence that a multisport background pays big dividends in learning and injury prevention,” Brustad says, noting that research around that would help all athletes, families and coaches.
Already a consultant with the U.S. Olympic Committee and the Real Sociedad professional soccer team in Spain — where he travels to advise on player development for the team’s 13- to 16-year-old division — Brustad could be helping an MLB team in the near future with his Moneyball 2.0 approach.