Nov 03 2008
At RCP the state polls are definitely showing a last minute surge to McCain. The country is going gray as more and more states head to the toss-up category – too close to call. Some folks just don’t want to admit this is happening, but it is and in Â some surprising states. But I count 10 states with 128 EC votes in the toss-up category, and another 8 states in the ‘lean’ category just waiting to tip into the ‘toss up’ group. That is a total of 18 states and 172 EC votes on the cusp right now. Who really knows which way these will go?
Let’s look at MN first. Survey USA just dropped a poll that shows a statistical tie with Obama +3% and at 49%. But to understand what this means one needs to look at the trend in this poll. Â In mid October Survey USA had Obama up +6%. That is a shift folks. Same trend is in the Rasmussen numbers. The RCP average of more stable (and democrat optimistic) polls show a much larger lead, but if the Survey USA poll is as accurate has it has been in the past, Minnesota is in play. Recall these late polls can only sample a bit of any last minute surge, not all of it.
Let’s look at PA. The Morning Call daily tracking poll for PA had Obama up +14% in early October, today that lead is now down to +6%, and one has to assume still falling as of today. And Morning Call is one of those heavily tilted polls which allow for unprecedented advantages for the Dems in their turnout model.
Look at VA, OH, FL, CO, NV at RCP and look at the trend lines – all of them are showing a striking movement towards McCain-Palin in the last two weeks. There is a wave out there building. How big is it? Don’t have a clue. What I do know is this election will test the two camps of pollsters – those who let irrational exuberance filter into their turnout models, and those who stayed anchored in historical trends.
At NASA we have seen people so enamored with their math or perspectives they never check to see if their assumptions are correct. We saw bad math models crash a satellite into mars. We saw people ignore historical trends and data, which led to the destruction of two shuttles. NASA performs millions of detailed and complex calculations a day, and we still need to make sure we aren’t seeing what we want to see in our math models. We assess them against independent data which should indicate concurrence.
Right now the data on polls is mixed and diverging. One group will be right tomorrow night. I cannot tell you which one, but I can tell you there is a battle of theories going on no less complex and emotional as the battle over what caused Global Warming.
I tend to think we are not breaking history to the point of wild speculation. It happens, but rarely. And the fact both camps are mobilized and voting makes me think the traditional ratios of voters will not be all that different. We will make history on turnout in general, but I do not see one side gaining a history shattering edge. In fact, I do not think the two can happen simultaneously. To get a historic edge means a large number of voters from the center to the right stay home, which means turnout could not be at historic levels as well.
We shall see tomorrow, but I my guess is when I do the last Bi-Poller report this afternoon we will have a final estimate for the two families of polls which will be quite divergent from each other. And only one family will have been right come tomorrow night.