I have taken the data from 33 game logs from the College Softball Division I 2009 season and conducted a preliminary analysis of the effectiveness of stolen bases. The data set is relatively small, so I caution the reader to take these outcomes with a grain of salt – eventually I intend to enter in the data from hundreds of game logs to get more comprehensive results. Still, the preliminary analysis on stolen bases is interesting enough to share.
In the 33 games examined, there were 69 stolen base attempts (which also included some successful pick-off attempts). 46 runners were successful (66.7%) and 23 (33.3%) were caught stealing. 48 of the 69 stolen base attempts were with a runner on first base, and 31 of those base runners successfully stole second (64.6%). A runner on first base occurred 386 times in the 33 games, so in this situation teams tried to steal about 12.4% of the time.
The other situation in which teams often tried to steal occurred with runners on first and third base. Teams tried stealing 15 times in this situation, and all 15 times were successful. Runners on first and third occurred 63 times in the 33 games, teams tried to steal 23.8% of those times, and so a team was about twice as likely to steal in the first and third situation compared to the runner on first situation.
Peeling back the onion a bit more, there were many times where a runner stole a base to which she would have been forced later in the inning anyway. For example, the runner on first successfully steals second base, but the next batter walks. Ignoring the arguments that the pitcher pitched more carefully and that the defense played differently with a runner in scoring position (unknown from the game logs), these cases are times when stealing a base is not worth the risk if one could look into the future. Of course we cannot do that, but we can take a look at the numbers to get an idea of how often a stolen base is rendered irrelevant by the following batters.
13 (28.3%) of the 46 successful stolen bases were later rendered irrelevant by the following batters. 6 (26.1%) of the 23 runners caught stealing would have advanced to the next base if they had not tried to steal. We can therefore guess that anytime a runner attempts to steal a base, around 3 in 10 (or 1 in 4 – take your pick) times the attempt is unnecessary.
A runner who successfully stole a base ended up scoring 11 times. Of those 11 runs, 4 would not have scored without the stolen base occurring (assuming that following base hits drove the runner ahead one base, doubles drove the runner ahead two bases, etc). Of the 23 times a runner was caught stealing, I estimate that 3 would have scored if the steal was not attempted.
From this preliminary analysis, it appears that stealing a base in College Softball is generally not worth the risk. 28% of the time, the runner would have advanced anyway because of what the batters behind her did. Additionally, while 4 runs did score which would not have if the runner did not steal her base, an estimated 3 runs were erased because a runner was thrown out stealing but probably would have scored based what following batters did. On the other hand, stealing second from a first and third situation seems relatively safe with the right runner.
On a final note, I caution the reader on a couple of things. First, this is a preliminary analysis with only a little bit of data. More data is needed before any firm conclusions are drawn. Second, all of the above talks about averages and average situations. Game situations are unique and should be managed accordingly. If my runner is blazing fast and their catcher is weak, I am sending my runner to second no matter what the averages say the smart play is.