Company News · November 20, 2009

The Baseball Award You Don’t Want To Win

By jlenski

Recently, Major League Baseball announced their Managers of the Year – Mike Scioscia of the Angels and Jim Tracy of the Rockies. Congrats to both.
However, based upon the history of the award, it is probably an award they would rather do without.
Of the previous 53 winners since 1983, almost half had lost or left their managerial job within the next three years. In addition, only four of the 53 teams with a Manager of the Year improved their record in the next season – the last being Joe Torre’s 1996 New York Yankees.
The win-loss stats of Manager of the Year teams follow a specific pattern. The average number of wins for a Manager of the Year is 95*. The average number of wins for the same teams the year before is 80. Obviously the baseball writers are rewarding a manager whose team shows a marked improvement. However, these same teams the year after their manager wins the Manager of the Year average only 83 wins – only 2 wins better than .500 – an average drop of 12 wins.
So what explains how a manager can be good enough to be honored as the best one year and then becomes spectacularly average the next year? This is just another example of regression to the mean. Many of the teams whose managers receive the award are teams with relatively average talent on the field that over-achieved for one season. The baseball writers observe this over-achievement and figure that the manager must be the reason behind it. The next year the team still has the same relatively average talent and, surprise, they revert to being a .500 team.
I haven’t yet done the analysis for coaches of the years in the other professional sports but I would bet that the pattern is similar. In fact, just this week the Hornets fired their coach Byron Scott just 19 months after he had been NBA Coach of the Year.
So beware Angels and Rockies fans. The odds are strongly stacked against your teams repeating their strong performances next year.
* Note: In my analysis I scaled the number of wins for teams in the strike years of 1994 and 1995 to what they would have projected to over a full 162-game season.

Get our latest insights delivered to your inbox.