MSM Musings: Stolen Base Efficiency

For my money, the New York Sun has the best baseball coverage of any New York daily. No disrespect to my friend Jim, but as a whole the Sun's baseball writers are an innovative lot with great ideas, whose daily efforts are more insightful and more useful to baseball fans than the predictable offerings you get from the print columnists in the established dailies.

You would never see a piece like Caleb Peiffer's excellent offering about stolen bases in one of these papers, because the old-school mentality of sports writing and reporting is too deeply ingrained in the editors there. (It should be noted that Peiffer is not a regular writer for the Sun, but a contributor from Baseball Prospectus.) Pieffer's column today, however, is far more illuminating about a topic that's always good for a discussion among baseball fans - the efficacy of the stolen base and the effect that speed has on the game.

For my money, the stolen base is one of the most exciting plays in baseball. When a good base stealer gets on, everyone sits up and takes notice. The pitcher alters his natural motion to pitch out of the stretch and to decrease the time it takes for the ball to leave his hands and land in the catcher's mitt. The catcher has to consider calling more fastballs and fewer breaking balls to give himself a better chance if the runner takes off. The first baseman moves out of position to hold the runner on. The middle infielders exchange signs about who will cover second if the runner decides to try his luck. What other play in baseball changes the roles of five out of the nine players on the field?

Speed has such an underrated effect on the game, mainly because it is not easy to conclusively prove the effect it has. You can measure stolen base efficiency, and plug it into a run expectancy matrix to see if the attempts had a positive or negative effect on the overall number of potential runs scored. Doing so has led to the conclusion that stolen base attempts need to be successful 75 percent of the time to have a positive effect on run scoring. Since most base stealers are rarely so successful over the course of the season, a new school of thought has emerged that stolen bases are not only overrated, but they are also generally harmful to a team's chances of winning.

Disruption, however, is very hard to quantify. If a fast runner causes a pitcher to alter his motion and his pitch selection when facing the next batter, does that have a positive effect on the offensive team's chances to score? With one fielder always out of position and a second ready to be at a moment's notice, do more groundballs find their way into the outfield with a fast runner on first? What about the pitcher's overall mental state? If the pitcher has a lapse in concentration because he's focusing too much on the runner, couldn't that lead to more runs?

The act of stealing a base can be quantified and objectively determined to be positive or negative on the team's chances of scoring. If Jose Reyes ends up swiping 80 bases this year, he can only be caught 26 times before his successful attempts are outweighed by the unsuccessful ones. However, there is no way to determine how the threat of Jose Reyes stealing a base affects the Mets' chances of scoring. I am willing to guess that the threat alone makes a positive impact, which is where the true value of speed lies.

That does not mean, of course, that fast baserunners should simply threaten to steal a base but never actually attempt to do so. Peiffer's column notes that teams in general have become much more efficient when attempting to steal, a compelling sign that sabermetric thought is having a positive effect on the game and strategic planning.

An efficient base stealer, then, may be one of the most dangerous weapons in the game, even if his stolen base totals do not immediately reflect that.

EDIT: One more thought: I've been playing Strat-o-Matic baseball for many years, and Strat actually quantifies the effect that a fast baserunner has on the game.

In the league I'm in now, we play a hybrid of advanced and super advanced rules. Say the aforementioned Reyes singles to lead off the game. The manager of the team in the field has to decide if he's holding Reyes on first, taking Jose's running rating, the pitcher's hold and the catcher's throwing arm into effect. This creates a series of probabilities. The first two are Reyes's chances of stealing second base if he "gets a good lead" or if he does not.

The more interesting probability, as it relates to this post, is the batter's chances of getting a hit. In our league, the probability of the batter getting a hit with a runner held on base increases about 3 percent. Put another way, a .250 hitter becomes a .280 hitter and a .300 hitter becomes a .330 hitter. It's easier to believe in the power of disruption when I play a game that quantifies it for me.