The Next Big Thing in Management By Stats? Antislumpism

In baseball, the most minute aspects of the game are captured, measured and used to adjust strategy and optimize a team’s chances of winning.  A few examples:

  • Pitches are selected based on hitter skills and tendencies.   

  • Hitters adjust their approaches at the plate based on pitcher tendencies.

  • Defenders are shifted according to where hitters hit the ball.  

These strategic changes are done game by game, matchup by matchup and even pitch by pitch.  And they have increased significantly in recent years (the increased use of defensive shifts is the most obvious example of this trend).

Yet one of the most important strategic decisions that a manager can make remains oddly resistant to day to day statistical analysis and that is lineup order.  

A 2014 study by Stanford’s Brett Green and Jeffrey Zwiebel concluded that baseball players do in fact have streaks which can significantly impact their performance in the short term.  Basically an average hitter can perform like a hitter at the top 75% mark when they are on a streak. These streaks can be measured with 25 at bats worth of data (5-7 games). (A link to the study is provided here for those who are statistically inclined:   

The Hot Hand Fallacy: Cognitive Mistakes or Equilibrium Adjustments? Evidence from Baseball

It seems that managers are still loath to move batting orders to give more at bats to hitters on a  hot streak and conversely, fewer at bats for hitters in a slump. This would seem to be a serious tactical error and something that teams should address.