I don’t disagree with Ezra Klein often, so it is all the more pleasant to answer his rhetorical question. He wrote
As a general point, I think “making people take semi-embarrassing votes” is vastly overrated in American politics. Can anyone think of a campaign that even partly turned on one of these gambits?
It should go without saying that if anyone can think of such a campaign then that person is Nate Silver. In fact he has thought of many such campaigns (which appear as dots in a scatter).
I know this is at that other newspaper which must not be mentioned here, but it absolutely answers your rhetorical question
Votes against ACA and TARP are associated with better than otherwise expected performance by Democrats running for congress.
I know you appreciate math, but your question suggests that you think it is useful only for policy analysis and not for political strategy. Even if there is no single case in which a vote on a bill clearly made a difference, it is possible to use information from many races to test and apparently reject the null that roll call votes don’t matter.
I suppose that Klein will argue that those were serious votes and not gambits. I don’t think that the argument that the whole business on which they insisted was a charade will be helpful to embarrassed Republican candidates.
In any case, my argument is that to inist on one case which is convincing all by itself is to reject useful statistical tools.