Have you ever wondered how managers chose when the right time to steal a base is? There's actually a strategy to it, and it may be a little more challenging than you think.
I am currently reading a book called Smart Baseball, by Keith Law, where he goes really deep into all the different baseball stats, and one that really stood out to me was stealing. The stolen base was first counted as a stat in 1886, but then later updated in1898.
The hardest thing for an offense to do is put someone on base. So why would someone steal because they're risking a baserunner, and they only have three outs to score. As Keith Law says, "Once you get a guy on base, the last thing you want to do is lose him to an entirely preventable out like a failed stolen base attempt." He then presented the "Run Expectancy Matrix," which is a table that shows what the expectancy is to score, given the baserunners and how many outs there are. The table from the 2018 season looks like this:
RUNNERS 0 OUTS 1 0UT 2 OUTS
000 (none on) 0.4998 0.273 0.102
003 (man on third) 1.3081 0.9098 0.3617
020 (man on second) 1.0785 0.6638 0.3207
023 (men on second and third) 1.8868 1.3006 0.5804
100 (man on first) 0.8721 0.527 0.224
103 (men on first and third) 1.6804 1.1638 0.4837
120 (men on first and second) 1.4508 0.9178 0.4427
123 (bases loaded) 2.2591 1.5546 0.7024
This table is also on the Baseball Prospectus website:
What this table shows is the averages for all of these situations, throughout the season. For instance, you can see that when a team has a man on first and 1 out, they would score about 0.52 runs for the rest of that inning. If the next batter gets a double, so now there are men on second and third and still 1 out, then the average run expectancy would now be 1.30.
Now imagine the stealing scenarios: Let's stick with the same situation as before, with a runner on first and 1 out, and a run expectancy of 0.52. If there is a successful steal attempt, then the runner moves to second, increasing the run expectancy to 0.66. But if the attempt is failed, then there are no runners on with 2 outs, moving the expectancy down all the way to 0.10. So you can see how a stolen base can cost a lot, sometimes less and sometimes more, deepening on the situation. Of course there are other factors of stealing, like speed, and who the batter is, etc.
MLB teams are realizing the costs of stealing, and so far in the 2018 season the stolen base leader, Mookie Bets, only has 27 stolen bases.
Another thing that plays a huge part is how often the runner could be caught stealing. An example given in the book is,"If I tell you a player stole 100 bases, is that good? Well, if he was caught ten times,yes, It's great. If he was caught 40 times, it's probably fine. If he was caught 90 times, what the hell is the manager doing?" The point that Law is bringing up is that no matter how many stolen bases you have, if you also have a high number of how many times you've been caught stealing, then something is off.
The bottom line, as said in the book, is that "teams shouldn't completely ignore the stolen base entirely- but should use it more wisely."