Whither or Whether the Stolen Base?

"The Book" is the canon of conventional wisdom on how to be successful in Major League Baseball. The Book is subject to revision, sometimes in radical ways. Nowhere is this more evident than in the deployment of the stolen base strategy. In recent times, stolen base strategy has come under fire, most notably perhaps, with the publication of Michael Lewis's Moneyball. Lewis attributes the A's success despite a modest payroll to their ability to overcome various flawed, biased approaches to constructing and managing teams that are endemic to Major League Baseball. Among those flaws is overvaluing speed and overusing the stolen base strategy. Stolen base attempts are down by more than 25% from fifteen years ago in the AL and by more than one third in the NL. Apparently, managing by the Book has changed. Over the course of the last century, the stolen base strategy went from great prominence to near extinction, only to rise again before an apparent slow atrophy has once again set in. A brief look at this history and at some measures of stolen base efficiency lead us to what we think is a rather interesting and somewhat surprising conclusion.

A Brief History of the Stolen Base

As home run hitting skyrocketed in the 1920s, stolen bases plummeted. Stolen bases were uncommon until offenses once again needed a jump start in the 1960s.

From there, stolen bases took off once again, and, for a time in the 1970s and 1980s, they seemed to coexist rather nicely with the home run until the 1990s when home run-hitting reached a new high, sending stolen bases down once again to about where they were in the 1920s. Yet there is more to the story than this brief overview can provide. And conclusions that the game is either a power game or a speed game do not capture all that is important regarding base stealing.

Fig. 1.

Stolen bases and home runs per team per game, MLB, 1900-2006

From Cobb to Ruth: Whither the Stolen Base

The "inside game" of singles, base stealing and sacrifices dominated the first two decades of the twentieth century. Paced by the likes of Honus Wagner and Ty Cobb, nearly 20% of players with at least 1,000 at bats, the modern equivalent of about two full major league seasons, over the five year periods shown below stole at least 100 bases; and more players stole at least 30 bases than had at least 1,000 at bats. Quite simply, nearly everyone ran. That all changed in the 1920s, in no small part due to Babe Ruth.¹ The change was not gradual; it was swift and dramatic.

More players started hitting home runs in the 1920s and 1930s, at the direct expense of stolen bases. In the second half of the 1930s, not a single player stole 100 bases for a team, while 98 players belted at least 20 homers for a team. While it would be easy to blame Ruth for the death of the inside game (as Cobb apparently did), at least some of the blame should be placed on the architects of the inside game itself.

Since it was simply part of the fabric of the game, little thought had been given as to the wisdom of the inside game as baseball strategy. Even the rules and official league statistics indicate a rather uncritical approach to the game. The modern definition of a stolen base did not emerge until 1898. Prior to that time, a player who took an "extra base," such as going from first to third on a single, was credited with a stolen base. As for the official statistics of baseball, caught stealing was only sporadically recognized; not until in 1951 was it finally recognized as an official statistic in both leagues for good. With the rise in home runs came a more critical look at using up precious outs on the basepaths.

Table 1.

Stolen bases and home runs...