Re: John Sickels' Rankings
I don't have time to do the math right now, but I strongly believe having Hamilton on second and Votto on first would yield a significantly higher expected runs than having Hamilton on first and Votto at the plate. You might be a little less likely to score one run, but you would be a lot more likely to score multiple runs. However, the risk of Hamilton getting busted in his attempt to steal second base is a major impediment even with Hamilton's blazing speed. The negative impact on run expectancy of a Caught Stealing is much, much higher than the positive impact of a successful steal. If you are going to have Hamilton steal second you should do it while Phillips is at the plate, and ideally have him steal third too.
In general, stolen base attempts have a negligible affect on run scoring. The Steals, Pickoffs and Caught Stealings cancel each other out over the long haul unless the runner has an 80% or better success rate (70% if you ignore pickoffs). The stolen base is an exciting play, but it is not a good idea to build your offense around it.
As mth123 stated, the run expectancy charts are generated by recording every single event in MLB and calculating the average number of runs that scored after that event. So it basically assumes that the batter and runners are league average in every respect. When you have a runner as fast as Hamilton and a batter as awesome as Votto the real run expectancy in that situation would be higher than the run expectancy matrix indicates. But those numbers should be higher across the board in pretty much the same ratios, so as a tool for picking the best strategy to employ in a particular situation the matrix should still be your guide.
If I were the manager I would have Hamilton bat leadoff and have Joey Votto and his .450+ OBP bat second. Mathematically that is the best way to do it to maximize run production.
Last edited by AtomicDumpling; 01-26-2013 at 08:01 PM.