It's human nature to hate losing.
Unfortunately, it's also human nature to overreact to a loss, potentially abandoning a solid strategy and thus increasing your chances of losing the next time around.
That's one conclusion of a Brigham Young University study published this week by the journal Management Science. The finding is based on an analysis of two decades of data on NBA coaching decisions.
The researchers focused on whether coaches adjusted their personnel following games where the margin of victory or defeat was small. After narrow wins, coaches changed their starting lineup one-fourth of the time. But after narrow losses, they changed their starting lineup one-third of the time.
"A blowout win is informative about the team's future success, compared to a crushing defeat," said Brennan Platt, a professor of economics at BYU and co-author on the study. "But losing by a point versus winning by a point, most of that is just noise. To say 'a win is a win' ignores important information in the intensity of the win, causing narrow losers to overreact and narrow winners to be complacent."
Platt and fellow BYU economists Lars Lefgren and Joe Price modeled the result of these attempts to right the ship. They found that hasty adjustments actually backfire, resulting in about one extra loss per season per team.
None of the coaches in the sample demonstrated immunity to overreacting to close losses. The study authors note that coaches also underreact to close wins -- particularly games that were expected to win by large margins. That's why they titled their study "Sticking with What Barely Worked."
Platt said the findings are particularly relevant to bosses' evaluations of employee performance.
"A lot of the goals businesses set are related to zero/one outcomes -- did you meet your sales quota, did the patient recover, did the plane arrive on time," Platt said. "You need to be careful to process all of the information. Things that are out of your control should be accounted for before you start evaluating staff. The intensity of the outcome -- by how far they missed the goal -- should give you an indication of whether it was just bad luck."
- Lars Lefgren, Brennan Platt, Joseph Price. Sticking with What (Barely) Worked: A Test of Outcome Bias. Management Science, 2014; 140721074632007 DOI: 10.1287/mnsc.2014.1966
Cite This Page: