Kerry has slipped from even to down 3-2 in the Tradesports betting market.
But the latest poll data don’t seem to be confirming the huge convention bounce for Bush measured by Time and Newsweek. Rasmusen’s latest has Kerry down by just 1.1%, and that includes the numbers from Friday, when Bush was riding high. (UPDATE: Scott Rasmussen thinks Bush’s lead is actually four points, based on the average of Friday and Sunday,and that his Saturday sample was punk. He thinks Time and Newsweeks fouled up their “likely voter” models and overweighted Republicans.)
The latest CNN/USAT/Gallup has Kerry down by 7, but Gallup tends to lean Republican; Bush has been ahead by three or four points, according to Gallup, since the end of July. The survey was taken Friday/Saturday/Sunday, so it reflects mostly the convention bounce and almost not at all any return to earth.
Prof. Pollkatz has a state-level model, using fairly fancy statistics, which projects that if the election were held today Kerry would squeak by Bush 271-267, without taking either Ohio or Florida. Kerry would have to hold on to slim leads in West Virginia, Wisconsin, Arizona, New Mexico, and Oregon to win that way, but it’s not a discouraging number at all.
It’s not that I don’t want Kerry and his supporters angry and scared. Angry and scared is good. What I don’t want to see is a replay of 2000, where the media convinced Democrats that Gore was losing, which made them think of him as a loser and start to pick him apart. The time for post-mortems is after the patient is dead. This one seems pretty healthy.