The Bradley Effect, Obama...and Hillary?!?

First a little history lesson: The Bradley Effect is the theory that an African-American candidate for political office can't win because in the end, while opinion polls can show the candidate ahead, when it comes to the actual vote, the candidate gets 10-15% less at the ballot box. In other words, there's a whole lot of white voters who say in opinion polls they will vote for the African-American candidate, but in the secrecy of the voting booth they pick the white guy.

It's named for Tom Bradley, the black mayor of Los Angeles who ran for Governor of California against Republican Robert Deukmejian back in 1982. The exit polls on election day showed Bradley had a comfortable lead in the polls...but the election showed Bradley lost by 2 points. Last minute undecided voters in particular chose Deukmejian in huge numbers.

Deukmejian's campaign manager, Bill Roberts, said a month before the election that Bradley's numbers were at least five points too high because he believed a large number of whites polled were lying about preferring Bradley for governor so as "not to appear racist". Roberts was forced to resign for his comments...but his predictions came true.

The Bradley Effect has been bugging me for months now that Obama has been the presumptive nominee, and I've often wondered about just how bad it will be in November for Obama.

Over at HuffPo this morning, Rebecca Curtis argues not only that the Bradley Effect will cost Barack Obama the election in November, but that there's only one possible way he can win: his Veep must be Hillary Clinton. Curtis argues history backs her up not just on Bradley but other African-American candidates.

In 1989, Douglas Wilder, the Democratic black Lieutenant-Governor of Virginia, ran for Governor, and stayed nine points ahead of white Republican Marshall Coleman all through the race. Yet on election-day, Wilder won by just half a point.

Also in 1989, African-American Democrat David Dinkins kept an eighteen-point lead over his rival for mayor of New York, white Republican Rudy Giuliani; until final tally. Dinkins squeaked by with two points.

In 1990, African-American Democrat Harvey Gantt ran against white Republican Jesse Helms for a North Carolina Senate seat. Throughout the contest, Gantt (like Obama) was predicted to win by 4-6 points. He lost to Helms by six.

Why the reversals? Some white voters lie about whom they support, so as not to seem racist. But most probably intend to vote for the black candidate, and simply, on the day of election, freak out. They feel suddenly nervous about the black candidate's "competence," or "experience," and pick the "known quantity,"--the white guy.

Summer-long, white liberals proclaimed we're "beyond race." In "The Myth of a Toss Up Election," analysts Alan Abramowitz, Thomas Mann, and Larry Sabato used voting patterns from presidential elections-past to conclude that--based on a 6-point lead--Obama would tromp McCain. By using (all-white) elections as their evidence, these upbeat boy-wonders assume race matters not at all. In June, Frank Rich of the New York Times reprimanded "doubters," noting that Obama had held on to "Hillary's" constituencies: blue-collar workers, Catholics, and Hispanics. (Obama's lead with those groups has since diminished.) Rich pointed out that Obama's June lead of six points was higher than Bush's over Kerry's in 2004, and concluded Obama would win in November. Rich (who's white) acts as if ignoring race were the only gentlemanly option: his suggestion that Obama will win because his June lead this year beats Bush's in 2004 implies--with country-club-style largess--the two men are comparably electable. But George Bush was a white, dynastic, Republican whose father was President; Obama's a black newbie Democrat. And Black candidates going for historically-white top governing positions always score nine to sixteen points lower than pre-election polls say they will.

With the polls as tight as they are, Obama's doomed if his support is really 10 points lower than it actually is being reported in the polls, right? So how can he win? Obama, Curtis argues, must not make the Democratic ticket about race and himself, but gender and Hillary Clinton.

To do as Wilder and Patrick did, Obama must partner with a figure who conveys tradition, competence, familiarity. Even if they're disliked by many--and by Obama--the Clintons convey that. More than her 18 million votes, Obama needs Clinton's household name. The Clintons ushered prosperity into America. One need not like them--or Hillary--to feel she's authoritative and familiar. Her presence on the ticket--like a well-known name-brand on an unknown product--would reassure swing voters.

An August Fox/Opinion poll found that Clinton's name--(and that of no other mate)--gives Obama an 8-point boost. Obama needs the boost.

In other words, the boost Clinton gives Obama will cancel out the Bradley effect...and it is the only thing that will help him.

Obama may loathe Hillary. But he says he wants to be president. No one knows who Tim Kaine is. No one will be reassured by his presence. And with him beside Obama, Obama's still what he is now: a diffident, perplexing, cosmopolitan, slightly arrogant black man. With Hillary beside him, Obama's the new guy on the team, and a hot ticket.

So here's the question: Is the Bradley Effect in play in 2008? Is eight to ten points -- or more -- of Obama's points in the polls going to McSame? If this holds true, this would give McSame a landslide victory in November, he would win by roughly 12-15 points and take any state where he's within 8-10 right now: Florida, Ohio, Michigan, Pennsylvania, VA and NC, and Missouri, and garner 350+ Electoral votes, handily winning the Presidency.

Do I believe Curtis? No. It's 2008 and Obama DID win the primaries. The Bradley Effect should have sunk him if it applied across the board. The surge of new voters flocking to him weren't even alive in 1982. The GOP has screwed up too badly, and McSame's campaign is too awful for even the Bradley Effect to bail him out. If the Bradley Effect was that powerful, Obama would have lost to Hillary...period. He would have never gotten this far.

I choose to see Obama's win in the primaries in states like Iowa and NC as proof that the Bradley Effect is dead. Curtis chooses instead to see it as the only thing more powerful than racism in politics is sexism...and yet there are a number of women in prominent political positions across America.

The Bradley Effect is a nasty little excuse to put Hillary on the ticket. It would be an admission that applied sexism and cynicism is the only way to beat McSame, not real hope and real change. It's an admission that America will always be racist in my lifetime and cannot change. Furthermore, it's an admission that Only Hillary Can Save Us...and she had her chance.

I don't buy it. Not at all. Even if you believe in the Bradley Effect after Obama won the primaries, I honestly believe it's already been factored in. Remember that consistantly in polls this year, generic Democrats are beating generic Republicans by 15-20 points. Obama is only ahead by 5. If you factor in the Bradley Effect, that would explain Obama's supposedly weak showing...but it would still give him a win.

Only if you believe that in a year where Democrats are winning by 20 points that racism is so bad in this country that McSame will win in a landslide does Curtis's ingenous theory hold water. it's much more likely that this is a scare tactic to get Hillary on the ticket, and it's pitting Democrat versus Democrat to try to accomplish that. By Curtis's own logic sexism cost Hillary the election, but as a subservient Veep she's a benefit?

If you believe Karl Rove is just the smartest political mind on Earth, go ahead and buy the line that Hillary is the only thing that can save Obama.