there are multiple ways to estimate runs scored by stats.
i believe markov chains are the best method.
http://www.pankin.com/markov/btn1191.htm
there's 1 linkn i quickly found in google... way, way, way better info than you will find in a forum... unless that forum is filled with math PhD's.
the biggest hurdle for managers is their own insecurity about somethign they may not understand. fear is what held baseball back. same reason why 40% of americans still do not believe in evolution.
i actually heard someone use the 'incredulous monkey' argument. "i aint come from no monkey" lol. reeks of fear. similar arguments made by people that don't understand what math tells us beyond the black and white #'s it spits out.
msupoke -- that's always something to consider with any mlb-cultural change (the general "how"). e.g. shifting... if they are not adjusting to the new data as stategies change, they aren't doing it right.
in this case we can see that even with common use it's still effective... we will see if players put their ego aside and start slapping it the other way for a single... then we'll need suitable sample to see if that is actually a net negative or positive etc etc. as far as i know, very few players do this... most stubbornly maintain their normal batting strategy and ignore the changes to the defense -- which is retarded.
it's ever-changing. not always logical in nature, but ever-changing.