by Clifford F. Thies

"It’s a good thing Barack Obama won’t be running against 'Generic Republican,' says one blogger. "There’s little chance of another Mr. Smith going to Washington," says another. "There’s just one small problem," says a third, "any time an actual candidate gets entered into the mix, the number drops."

Dream on!

At this juncture in the election cycle, it is not surprising that Generic Republican is ahead of even the best-performing named Republican. Let's look back to the 2004 election cycle. President Bush was presumed to be the Republican nominee; and, the Democratic nomination was up for grabs. Among the possible nominees paired against Bush in "match-ups" were Al Gore, Joe Lieberman, Hillary Clinton, Richard Gephardt, Welsley Clark, Al Sharpton, John Kerry and Jonathan Edwards.

I have compared the performance of Generic Democrat versus the best-performing named Democrat from the mid-term elections of 2002 to the end of 2003. Sometimes, the same poll-taking organization soliticited preferences in match-ups involving both the Generic Democrat and named Democrats. More often, certain pollsters used the Generic Democrat match-up and other pollsters the named Democrat match-ups.

For each month, I averaged all the available match-ups. As the accompanying chart shows, Generic Democrat consistently did about five points better in match-ups against President Bush in the monthly averages.

As to why Generic Democrat did better than the best-performing Democrat back in 2003, and why Generic Republican is doing better than the best-performing Republican this year should be obvious. It is not because the better-performing candidates are weak. It is because not all those who are inclined to support whomever will be the nominee of the challenging party is equally supportive of all possible contenders in advance of the nomination being decided. Some are more supportive of one candidate or another, and and reluctant to indicate to pollsters that, in the end, they will support whomever their party nominates.

This kind of result is typical in polling in Governor or U.S. Senate when one party, and not the other, has a contested primary. The candidate who doesn't have a contested primary will look artificially strong until the primary is concluded and the winner has had an opportunity to unite his party. One obvious way a nominee for President can do this is by naming a candidate for Vice President that helps to unify his party. Almost always, those who win their party's nomination are able to unite their party; although, from time to time, example 1964, the nominee is unable to accomplish this task.

One more thing: some pollsters ask a third form of match-ups, the incumdent versus "somebody else." Logically, "somebody else" includes potential independent and third-party candidates and even challengers for the nomination of the incumbent's party. Accordingly, "somebody else" tends to do even better than the Mr. Generic. In the 2004 election cycle, I found that "somebody else" outperformed Generic Democrat by an average of three points.