SoSH Glossary: RE24

Baseball is filled with statistics, rules, and archaic terms that can often form what sounds like a foreign language.The Sons of Sam Horn glossary provides a better understanding of these terms through straightforward definitions, clear explanations, and examples pulled straight from the baseball world. If there is anything you would like us to add to our glossary, please contact us.

RE24

RE24, or run expectancy based on the 24 base-out states, quantifies the effect a player has on the runs that are expected to score from the start of a play to the end of it. The quantity is found using a run-expectancy matrix, derived from empirical data lifted from MLB games, after adjusting for league- and park-effects. An example matrix can be found below:

MLB 2016

Runners(1st,2nd,3rd) RE 0 Outs RE 1 Out RE 2 Outs
000 0.5075 0.2749 0.103
100 0.8746 0.528 0.2197
020 1.0942 0.6675 0.3178
003 1.3123 0.9318 0.3627
120 1.4613 0.9206 0.4345
023 1.899 1.3244 0.5775
103 1.6794 1.1849 0.4794
123 2.2661 1.5775 0.6942

The 24 come from the 24 possible base-out states: three out situations and eight combinations of baserunners. The table above is courtesy of Baseball Prospectus and can be found here.

How is RE24 Calculated?

The calculation for RE24 is rather simple. You just need to have the proper matrix, then know the base-out states at the beginning of the play and the end of the play. Take the difference between the end value and the beginning value, then add any runs scored, to get the RE24 for that play. All of the offensive credit, or blame, goes to the batter while the pitcher receives any credit or blame for his team’s circumstances.The net RE24 is always zero on any given play, so that anything that the batter gains, the pitcher loses and vice versa. In the event of a running play (stolen base, caught stealing, wild pitch, passed ball, balk, etc.), the lead baserunner, if there are more than one, and the pitcher are the only two players that will have their RE24s affected.

For example, suppose the leadoff hitter in a game strokes a solid single to left field. At the start of this play, there were no runners on base (000) and no outs, so his team was expected to score 0.5075 runs by the end of the inning. As he stands on first (100), there are still no outs, so his team is now expected to score 0.8746 runs. The difference, 0.8746 – 0.5075 = 0.3671 runs, is the leadoff man’s RE24 score.

What’s the Point?

At the end of a game, a player’s RE24 can be accumulated to see what his impact on that game was. A player with a score above zero is considered above average while anything below zero is below average. Further, RE24 is kept track of throughout the season and we can then see how many runs above or below average a player contributed contextually to his team. The sum can further be divided by a figure usually between nine and ten (the value of a win depending on the year) to determine how many wins that player was worth based on RE24.

Relievers and RE24

The players that RE24 benefits the most are relievers who enter the game with men on base. Let’s use as an example Kansas City Royals relief ace Kelvin Herrera entering a game with the bases loaded and no one out against Boston Red Sox right fielder Mookie Betts. We’ll use the above unadjusted table for simplicity’s sake:

MLB 2016

Runners(1st,2nd,3rd) RE 0 Outs RE 1 Out RE 2 Outs
000 0.5075 0.2749 0.103
100 0.8746 0.528 0.2197
020 1.0942 0.6675 0.3178
003 1.3123 0.9318 0.3627
120 1.4613 0.9206 0.4345
023 1.899 1.3244 0.5775
103 1.6794 1.1849 0.4794
123 2.2661 1.5775 0.6942

If Herrera forces Mookie Betts into a 1-2-3 double play, and is then lifted so that lefty Matt Strahm can face David Ortiz, his traditional box-score line for the night would read: ⅔ of an inning pitched… and that’s it. However, RE24 will capture the magnitude of what he accomplished more accurately. When he entered the game, the Red Sox were expected to score 2.2661 runs according to the run expectancy matrix. When he left, there were two outs and men on second and third: a run expectancy of 0.5775, or a swing of 1.6886 runs. This is certainly an extreme example, however, it illustrates why having multiple tools at one’s disposal to evaluate players is so valuable.

For further reading on the run expectancy matrix, check out this article by Lee Gregory.


Follow Pete on Twitter @PeterWHodges.

LEAVE A REPLY