You need to read the second article. That number wasn't adjusted for some important factors; defensive alignment, the jump in batter performance with any runner on first, the relative quality of pitchers with runners on vs. with the bases empty, etc. The real benefit is much smaller, somewhere in the order of 0-3 runs per season. The author's numbers are extremely flawed since he compares a select few players that are known as base stealers compared to an overall group that also includes those same base stealers and other leaser known base stealers. It should make a seperate comparison of good base stealers versus poor base stealers. He is cherry picking numbers to try to support his argument when the basis of comparison is flawed. The overlap between the group of all runners and first base and the ones considered stealers is negligible. (About 3% of the 100,000+ sample size.) Removing the stealers from the overall group would have had no significant imapct on the outcome. The results are also verified by his independent method of adjusting for defensive shits and pitcher quality. (Though that method showed a slightly larger disruptive effect.) The analysis is still supportive of my conclusion that a base stealer can disrupt a pitcher's focus on the following batter. Batting average is a much better measure of this effect than runs scored. Runs scored also takes into account that base stealers get caught stealing, thereby reducing the overall number of runs scored. I'm not talking about runs scored... I'm talking about distracting the pitcher. In any event, the analysis is nevertheless flawed as a theoretical matter because the author is comparing a barrel of apples (base stealers) to a mixed barrel of oranges (non-base stealers) and apples (base stealers). The assumption that this error has only a 3% negligible effect is incorrect because it assumes, ipso facto, that the other 97% of runners are non-base stealers. Without doing a statistical analysis, I'd venture to say that more than 3% of baseball players are capable base stealers. I agree with katway - I'm less than convinced by that Hardball Times analysis. In Part 2, the author used a "complicated" (author's word) method, which I found difficult to follow, and whose statistical significance seems suspect at any rate, and a 2nd "simpler" method of separating out a golden group of annointed basestealers against a control group of ALL players. Why separate out the fastest runners? The fact is, a pitcher must pay attention to MOST runners on first -- all but the SLOWEST guys in the league would logically disrupt the pitcher to some degree. But all but the slowest guys will also force a change in defensive alignment (1B holding the runner). The 2nd method is a specious argument IMO, inherently biased in favor of the null hypothesis that a speedy guy on first does not disrupt the pitcher. The disruption effect is extremely difficult to isolate statistically because defensive alignment changes are so positively correlated. Also, I didn't see the author use any data involving errant pickoff attempts, which give speedy runners one or two extra bases - did I miss that? I really think you need to watch the games to see the disruption effect and hear from ex-professional pitchers (anyone?) - I believe it's a real effect.