FREE hit counter and Internet traffic statistics from freestats.com

Monday, November 08, 2004

Measuring Baserunning: Setting a Baseline

In the interests of furthering the state of sabermetric knowledge per my previous post I compiled the probabilities of runners advancing in various situations using play-by-play data from the 2003 season. The scenarios I looked at included:

  • Runner on 1st, Batter Singles
  • Runner on 1st, Batter Doubles
  • Runner on 2nd, Batter Singles
  • Runner on 2nd, Batter Doubles

For each situation I calculated the odds of advancing a "typical" number of bases (1 for a single, 2 for a double), the odds of going one better, the odds of going 2 better (where applicable), the odds of getting thrown out advancing, and the percentage of the time a runner occupied the base directly in front of the runner. I then create subtotals for each situation by the number of outs and the outfielder who fielded the hit. For example, for a runner on 1st when a batter singles the values are:



Outs To Typ +1 +2 OA Next Base Occ
All All 70.5% 27.2% 0.9% 1.4% 29.2%
0 All 73.4% 25.0% 0.5% 1.2% 21.1%
7 84.5% 14.1% 0.6% 0.7% 21.6%
8 68.6% 30.1% 0.3% 1.1% 25.0%
9 59.7% 38.3% 0.6% 1.4% 16.5%
1 All 72.4% 25.5% 0.7% 1.3% 30.6%
7 84.7% 13.4% 1.0% 0.9% 31.5%
8 70.3% 28.6% 0.4% 0.7% 34.0%
9 58.1% 39.0% 0.9% 2.0% 27.9%
2 All 66.3% 30.7% 1.4% 1.6% 33.9%
7 81.1% 15.8% 1.4% 1.6% 33.8%
8 60.0% 35.9% 1.8% 2.3% 33.1%
9 48.3% 49.7% 1.0% 1.1% 31.1%
All 7 83.4% 14.4% 1.0% 1.1% 29.6%
All 8 65.8% 31.9% 0.9% 1.4% 31.5%
All 9 55.3% 42.3% 0.8% 1.5% 25.8%

The value in breaking these numbers down can be illustrated by noticing how runners advance from first to third less than a third of the time (27.2%) overall but almost half the time (49.2%) with two outs when the ball is hit to right field. It's also interesting to notice that runners get thrown out most often in this situation with 2 outs when the ball is hit to center field and that runners only advance to third base 14.4% of the time when the ball is hit to left field.

Because I couldn't format the results as I wanted on this blog I've posted the complete results for all scenarios here.

More on how we can use these numbers to evaluate teams and players next time.

No comments: