I have been wondering: how does attendance affect hitting performance?
By now we've grown used to seeing guys hit poorly once we acquire them, and seemingly hit well once they leave. Recall 2009: Matt Holliday performs okay-but-not-to-his-standards for us, then tears it up for St. Louis. Jason Giambi looked like he was completely washed up; but since then, even after adjusting for the Coors effect, is still doing all right for his age.
How much of this can be attributed causally to low attendance and lack of atmosphere at the Coliseum? Continued after the jump
Lest I give you false hope, I have not had time to carry out my own statistical analysis of this issue. So I don't know the answer to the question. But I will comment on a few things that perhaps can inspire someone else to do so.
My hypothesis is that: even though the Coliseum is a pitcher's park, at least part of its run-suppressing effect lately is due to sparse crowds and lack of buzz at games. I think this would hurt both teams' hitting, since players just wouldn't feel amped for the game. I even suspect that this hurts the A's more than visitors, since they see it (demoralizingly) day in and day out - though not to the extent of eliminating home field advantage.
Econometrically, it takes some care to discern the causal effect. Simple park factor stats lump together the objective pitcher-friendliness of the park and the "lack of buzz" effect. And we can't simply compare high-attendance games vs. low-attendance games either, since certain teams systematically outdraw others (think Yankees vs. Marlins), and that introduces other factors. Ideally, we would need a controlled experiment where we change nothing else about a game except the attendance, and see how that affects hitting performance. That is, "Marlins with small crowd" vs. "Marlins with big crowd"; "Yankees with small crowd" vs. "Yankees with big crowd."
EDIT: Thanks to those bloggers who have taken an initial look at the stats. A simple correlation between hitting and attendance is a place to start, though unfortunately not fully informative, since it can't distinguish the direction of causation (better hitting --> higher attendance or vice versa). mendelbob's observation that the year-to-year correlation is stronger than the in-season correlation shows the not-too-surprising effect that fans respond to winning. Now the key question is how much of the in-season correlation is a causal result, and not just "in-season" fan response to team success.
In a perfect world, there are a million and one things to control for. But possibly there is a creative and simple way (insrumental variable, anyone?) to answer the question in "mostly" sound technique. Anyone feel up to the task?