Palin and the Polls

We all know that Obama did not get a bounce from selecting Joe Biden as his running mate. The McCain camp hopes that after tonight, Palin will deliver an uptick in his poll numbers, and given the enthusiasm the GOP is showing, it is all but certain.

Given that we live in a poll-driven 24/7 news cycle, the media will surely over-analyze Palin's convention speech and tomorrow, will inundate its consumers with new flash polls and talking-head speech impact predictions.

Should McCain receive a bounce, the media is likely to report it as a dead-heat horse race; should Palin not deliver, the media will likely cast her as a pariah. (Media bias has become more definable and the average news consumer is now telling pollesters just that.)

Any political anaylist worth his salt will tell you when it comes to polls, the devil is in the details. Campaigns look at their internals and toss aside things such as registered voter polls, flash and exit polls, frankly because respondents prior to election day lie about their intention to turn-out, or are reacting to some news event, or on election day, who they actually voted for.

That's why likely voters are the best indicators prior to the election, however the day, time, and demographic are key to deciphering the accuracy.

Over the past few weeks, LV polls have shown time and again, Obama is in trouble. At this juncture, the Illinois Senator should have at least 12 to 15 point lead -- he does not. What's more his generic electoral count is mucher higher than McCain's, meaning he has little room to grow, whereas McCain can make bigger leaps.

McCain is leading in Florida and Ohio, and is within a point of the margian of error in Michigan, and nearly as close in PA. Save the partisan 527 sniping, any real momentum is likely to come from the debates and with the approach of the debates, it means voters are still unsure about Obama and he can little afford a misstep.