This is really unbelievable. Gallup posted an analysis on their site yesterday about estimating election probabilities based on labor day poll data that is almost completely worthless. The reason is that they focus on Kerry's 7 point deficit among LVs on labor day (can he overcome it?), while basing their analysis almost entirely on data about RVs on labor day.

How do I know their labor day poll data is almost entirely (prior to 1996) based on RVs? Because they published these data, clearly marked as being from RVs prior to that date, in an analysis on their own site just 6 days ago (September 3)!

Don't they read their own stuff? Clearly it makes no sense to analyze a lead among LVs this labor day, and its possible relation to the final outcome this year, on the basis of historical data about RV leads on labor day and how much they changed by election day.

Thus, the question Gallup should have been asking is: can Kerry overcome his 1 point deficit among RVs by election day, based on historical patterns? Turns out the answer to this question--really, the only question that their data can properly answer--looks pretty darn good for Kerry.

In 17 of 17 cases, going back to 1936, the labor day margin between the candidates changed enough for Kerry to tie or surpass Bush in the popular vote and, in 12 of 17 of those cases, the change was in Kerry's direction (i.e., that is, in the direction of the candidate who was behind among RVs on labor day).

Moreover, if you compare Bush's position to the position of incumbent presidents who won their campaigns for re-election, it doesn't look auspicious. In 9 cases, going back to 1936, winning incumbent presidents on labor day had an average lead of 12 points and a median lead of 11 points among RVs. The only winning incumbent president who was in a worse position than Bush is this year was Harry Truman in 1948.

Maybe I'm biased, but I have a really hard time seeing George W. Bush as Harry Truman.