Why Does Firing a Coach Usually Improve the Next Season?

Jared Diamond says coaches get fired after a losing season, and CEOs get fired when a company has a poor year and low returns. Fire a CEO after a losing season and the new CEO usually has a better season the next year. 70% of the variance in a sports outcome is due to factors that are random with respect to the CEO.

He compares the firing to dice. Throw a one, call the thrower lousy, and fire that person. Five out of six times the next thrower does better than one, because the other faces are two, three, four, five, and six. In sports, the new coach usually has a better record for the same random reasons. Jared Diamond says his university, UCLA, now has an athletic department $150 million in debt because UCLA fired every basketball coach since John Wooden and had to buy out each contract for $5 million, $10 million, or $15 million. Fire a CEO and the next CEO is likely to do better for random reasons, as with dice.

The dice pattern is why business school professors have gotten lots of their insights by analyzing sports.