The pitcher peers in for the sign. The runner takes his lead off first – slowly inching away so as not to get picked off. The pitcher kicks and deals and in a cloud of dust the runner takes off for second base, every muscle in his body straining for traction as he pulls a fade away slide at the bag just barely avoiding the tag of the second basemen. In the fantasy game all we care about is the steal. That's it. Still, should we be satisfied with simply looking at a box score and counting stolen bases in our attempt to analyze base runners? The simple answer to that question is of course not. Over the past few years more information has been uncovered that speaks to the effect of the stolen base on the game, and in what follows I will take a look at one of those measures in great depth. 

NOTE: If you want to consider updating your scoring setup in fantasy you could always go with the simple yet elegant NET Stolen Bases (SB-CS).


I'm going old school.

Pete Palmer’s Linear Weights, akin to Bill James Win Shares, attempts to measure a players overall ability on the field by taking into account everything a player does in all phases of the game including pitching, hitting, fielding and base running (this measure is a precursor to measures like WAR that everyone loves to chat about these days). In essence, Linear Weights is an attempt to come up with a single measure to evaluate all players, regardless of position, on one scale. Stolen Base Runs is the base running component of the formula, and it is the focus of the current piece (to see a full description of the massive formula used for Linear Weights which includes pretty much everything that a player does on the field. By the way, Linear Weights has now been renamed Total Player Wins so the terms are interchangeable).

Contradicting what “the book” on baseball says about the value of the stolen base, Palmer’s work has shown that stolen base really isn’t as important in terms of it's ability to effect the outcomes of a game. The reason? The negative effects of the caught stealing (CS) far outweighs the positive effects of a successful stolen base attempt. Palmer’s analysis has led him to conclude that unless a player is extremely successful in stealing bases that they should not even attempt to steal a base (this “theory” is based on reconstructions and simulations of thousands of games that have taken place over the past 100 years). Studies have shown that a base runner must be successful on two out of every three steal attempts or he is actually hindering his team’s ability to score runs. Therefore, unless a players steals bags at a success rate of at least 67 percent the benefits to the team are negligible at best. 


As mentioned, Total Player Wins (or Linear Weight’s) is a complex system that takes into account almost everything that occurs on the ball field. For obvious reasons, I don’t want to spend the next nine hours explaining how the whole formula works, so I have decided to pull out the one part of the metric that deals with the stolen base, and that is called SB Runs.

Historically the numbers used in Palmer’s formula dealing with stolen bases, based on the 2/3 break even point, were (.30) for SB and (-.60) for CS, but subsequent research has determined that (.22) for SB and (-.45) for CS are slightly more accurate, so they are the numbers that I will use in this study. With that, here is the simple formula for SB Runs.

SB Runs = ([.22*SB] – [.45*CS]) 

Basically, a caught stealing is twice as damaging to a team’s ability to score runs as a successful stolen base is helpful, so therefore a CS is more heavily weighed in this formula. That’s pretty much the gist of what this measure is trying to point to.

One last note before we get into the analysis. It takes about 10 SB Runs to account for one team win, so a player who earns five SB Runs in a season contributes about ½ a win to his teams total over the course of an entire season. Overall, the steal doesn’t have that much of an effect on the outcome of a game except in very specific situations.