Bradley and the Bradley Effect

I've heard a couple of commentators on the networks this weekend talking about "the Bradley effect"--the phenomenon that black candidates punch below their polling weight with white voters. Will it torpedo Obama? So far, I gather from the commentators that (a) everybody assumes there is a Bradley effect; and (b) nobody knows how to measure it.

I don't know how to measure it either but I was living in California and voted for Tom Bradley in the 1982 election for governor, where the term arose. Let me offer a few wonky footnotes:

• Most important--unless my memory has gone bollywackers, Bradley had the misfortune to share the ballot that fall with some sort of gun control initiative--I've forgotten the details, but it was precisely the sort of thing that was bound to bring the wingnuts to the polls on election day, in numbers far ahead of whatever might have been the conventional estimate.

• Bradley was no bozo: he was a five-term mayor of Los Angeles with indisputable stature as a pro. But he was also low-key to the point of near-invisibility, the kind of guy whose very presence could suck the oxygen sraight out of the room. This wasn't a bug, it was a design feature: his very impassivity is one of the things that made him effective in the sometimes raucous cross-currents of Los Angeles politics. But the corollary is, he was precisely the kind of candidate who would look puzzling or unfamiliar to voters from outside is this base.

• For what it is worth, his opponent, George Dukemejian--vapid and competent (he swamped Bradley in the rerun in 1986). Dullness can be a virtue in a state that churned up Ronald Reagan, Jerry Brown and the gentleman from Muscle Beach.
  

I haven't any idea what all this proves except that every election stands on its own feet, and there probably a great deal about the Bradley effect that is peculiar to Tom Bradley.