Huffpost Politics

Featuring fresh takes and real-time analysis from HuffPost's signature lineup of contributors

Mark Blumenthal Headshot

Then and Now, Take Two

Posted: Updated:

If you read about today's Washington Post/ABC News poll online, you probably missed the "Act II, Scene I" sidebar graphic from the Post's print edition that is also available online. The gist if the graphic is that while the McCain-Obama vote looks now looks identical to Bush vs. Kerry preferences four years ago, the larger political terrain as defined by the Bush job rating, the perceived direction of the nation and views about the Iraq War are very different.

2008-06-17_Act II, Scene I -

One piece of context worth considering, however, is that the Post/ABC poll of June 2004 had Kerry doing slightly better than other polls taken at about the same time. The following list -- which comes from the RealClearPolitics listing from 2004 -- shows that the Post/ABC survey was the only one conducted during the latter half of the month showing Kerry with even a "numeric" lead. The average for the month had Bush ahead by a single percentage point (45% to 44%).


Regardless of whether the current national horse-race is exactly the same as at this point in 2004, or a few points better for the Democrats, the larger point of the Post graphic still holds: We have seen far bigger changes in the percentage of Americans that disapprove of George Bush's performance as president (+17 percentage points since June 2004 in the ABC/Post poll), that say things in the country "have gotten pretty seriously off on the wrong track" (+27) and that conclude the war in Iraq was "not worth fighting for" (+11).

And as long as we are on the subject, we cannot repeat it often enough: Polls are, at best, a measure of where the race stands "if the election were held today. " It isn't held today. The 2004 race aside, polls in June are historically poor predictors of the ultimate outcome of the presidential election in November.

Nate Silver did a nice round-up over the weekend of how June polls since 1988 compare to November outcomes and concluded:

So in four out of the last five elections, an average of June polls would have incorrectly picked the winner of the popular vote. That's kind of a problem for anybody who is overly confident about how this election is going to turn out.