Last night was not an example of the Bradley Effect!

The talk of the day is how Barack Obama was a victim of the Bradley Effect last night, meaning that his poll numbers were inflated because he was black (people tend to say they'll vote for a black candidate because they feel guilty otherwise, but then when they actually vote, they go with the white candidate).

Here's the thing: his poll numbers were not inflated. He got 37% of the vote in New Hampshire. He was projected to get 38.3% of the vote, based on the RCP average. A 1.3% decrease isn't indicative of the Bradley Effect: if anything, the discrepancy is due to the massive--and frankly, unrealistic--surge he got after Iowa.

Hillary Clinton led in the state up until Iowa. This isn't a major upset, and this isn't indicative of the Bradley Effect. This is the country righting itself after fueling Obama's hype for the past week. Clinton was expected to win New Hampshire.

Ponder this, though: Maybe Iowa was the reverse of the Bradley Effect. Undecided voters supporters of second-tier candidates had to publicly choose who they'd caucus with. Did white guilt play a role? If Iowa was a normal primary, would Obama have won by such a wide margin?

Food for thought.

--Wyndam