News at Smith

Three Reasons Baseball Defies Prediction Models

Oct 26, 2016


PDF: Predicting World Series WinnersSMITH BRAIN TRUST — Baseball prognosticators love statistics, but performance over the course of 162 games does not necessarily predict postseason success. Since the wildcard era began in 1994, the best regular season team has won the World Series just 19 percent of the time — the same rate as the postseason team with the lowest regular season win percentage. If the Chicago Cubs can survive the Cleveland Indians, they would be the fifth club in 22 tries to convert the regular season’s best record into a World Series championship. Sean Barnes, a professor at the University of Maryland's Robert H. Smith School of Business with a background in applied mathematics, shares three reasons why six months’ worth of baseball data means so little in October.

1. Condensed sample: Poker players who study cards can tilt the odds in their favor over the course of an extended tournament, but anything can happen in a single hand. The longer they sit at the table, the greater chance they have to separate themselves from inferior opponents. Something similar happens in baseball. The best teams emerge over the course of 162 games, but anything can happen in one short series. “You’ve now condensed a 162-game season into just a few games, and that significantly increases the possibility of unexpected outcomes,” Barnes says.

2. Narrow margins: Elite football and basketball teams routinely win more than 70 percent of their games. This rarely happens in baseball. Most baseball teams win between 40 percent and 60 percent of their games, and playoffs teams are typically separated by thin margins. In the wildcard era, the gap in terms of win percentage among postseason qualifiers has averaged only 8.7 percentage points. The NFL average over the same period was 29.9 percentage points, while the NBA average was 29.4 percentage points. "Playoff teams are more closely matched in baseball than they are in other sports, and this contributes to the uncertainty in outcomes," Barnes says. With such little separation from top to bottom, one or two players can get hot on any given night and determine the outcome of a game. Pitcher Corey Kluber and catcher Roberto Perez did this for the Indians in the World Series opener on Tuesday.

3. Objective function: Business analysts sometimes talk about an "objective function," which quantifies an organization's primary goal. Barnes says baseball’s objective function changes in the postseason, and some teams adjust better than others. “During the regular season, your objective function is to qualify for the postseason,” he says. "This means you want to win as many games as possible out of 162 — at least until you've clinched your spot." Teams with this goal don’t sacrifice the future to win a single game or series. But once they get to the postseason, they adopt short-term tactics that they wouldn't be able to sustain during a longer period of time. "They modify their objective function," Barnes says. "They focus on winning the current game, with minimal regard toward long-term consequences." For example, he says, top starting pitchers will often pitch on short rest, and relief pitchers are often brought in at the first sign of trouble.

Baseball has no shortage of experts willing to make predictions, but Barnes doesn’t listen. “The chance of unexpected outcomes is so large that it’s almost pointless,” he says.




About the University of Maryland's Robert H. Smith School of Business

The Robert H. Smith School of Business is an internationally recognized leader in management education and research. One of 12 colleges and schools at the University of Maryland, College Park, the Smith School offers undergraduate, full-time and part-time MBA, executive MBA, online MBA, specialty master's, PhD and executive education programs, as well as outreach services to the corporate community. The school offers its degree, custom and certification programs in learning locations in North America and Asia.