To see what I could see, for myself, I downloaded retrosheet data for Oakland 1982. Rickey stole 130 bases that year. What I wanted to know is if having Rickey on first with 2nd open bothered pitchers enough to make a difference with the batter.
I've never worked with retrosheet data before, so additionally this gave me a way to start to familiarize myself with the files.
Since I was trying to make this as simple as possible, I really really cut down on the sample size. So my conclusion may be totally worthless, I may have done things totally wrong. Maybe someone could quickly run some simple numbers of their own and see if what I found holds up, or is way off base (haha!).
Basically I looked at times when Rickey was on first and second was open. I ignored times that Rickey actually stole 2nd or was caught stealing second. I ignored times he was picked off of first. I decided, arbitrarily, that these events took pressure off the pitcher.
In the remaining events, I checked for a positive or negative outcome. Positives were hits, walks, wild pitches and passed balls. Negatives were, well, anything else (outs).
With Rickey on first and second open (n=81) there were 27 positive outcomes - .333 average.
With someone not Rickey on first and second open (n=2632) there were 928 positive outcomes - .352 average.
The p-value here is 0.72, no significant difference.
I'm a little surprised by this, since intuition tells me Rickey at first would make pitchers worry and maybe get a little rattled. But maybe not, maybe its just anecdotal when he rattled a pitcher here or there, but for the most part guys didn't really care. I dunno - but this back of the envelope (of sorts) rough estimate says that pitcher harrassment by having Rickey on first, threatening to steal, didn't exist.
Has anyone looked at this before? Any thoughts? Again, this is the first time I've used retrosheet data and just thought I'd throw this out there to see if I'm doing something way wrong.