Above: Who is driving the BMW? Oh who!?
I find it most amusing that the R Tent and D Tent are ‘absolutely certain’ of their own victory. It like watching pent up pressure awaiting to explode on Kermit Day.
I began asking myself, “Have I seen this before?” Actually, I have. It was in 2004. There were several people in the D Tent who made little homesites that showed the maps and said, “I have done the math. There is a 99% probability that Kerry wins the election.” I did a simple google search and found…
Poll-based Election Model Gives Kerry a 99% Probability of Winning
17-Oct-04
Campaign 2004 Predictions
The Election Model by TruthIsAll is a comprehensive and unique election analysis model which cuts through the fog of polling. Right now, the model shows that Kerry has a 99% probability of winning the election. That is the number derived from the National poll projection model and the State Poll simulation model. They confirm one another. The methodology is explained in summary and detail on the home page. The graphics are plentiful and illuminating. The polling data input (national and state) are presented in detail, and the analysis as well. This model has been back-tested with the 2000 final state voting numbers and was correct to within .04% of the Gore and Bush final percentages.
Kerry was always winning the election until he didn’t.
The HorseRace Blog, the Right equivalent of the 538 model back in 2004, had Jay Cost remark on his column about how much he had changed. He thought back to those statistical models of his youth and says how much he has grown up.
When I started making election predictions eight years ago, I had a very different perspective than I do today. I knew relatively little about the history of presidential elections or the geography of American politics. I had a good background in political science and statistics. So, unsurprisingly in retrospect, I focused on drawing confidence intervals from poll averages.
Since then, I have learned substantially more history, soured somewhat on political science as an academic discipline, and have become much more skeptical of public opinion polls. Both political science and the political polls too often imply a scientific precision that I no longer think actually exists in American politics. I have slowly learned that politics is a lot more art than science than I once believed.
What I don’t understand about the Nate Silvers, who place a high degree of self certainty in complex sociological phenomenon of elections, is why they don’t go into investing and finance. There are countless “polls” (or surveys) all over the investment community. Elections are more complicated than investing than markets can be. If I could predict elections with such assurance, I’d make a fortune in investing. Why apply that to just games like baseball?
And the question answers itself.
At first, I was thinking the election would be like 2004 with the Rs having a +1 or +2 better performance. This is what Karl Rove is thinking. But after looking at the Bickers and Berry model, it got me thinking of just how economic fundamentals defy campaigning and has a gravity that pulls people to earth. Michael Barone agrees and predicts an electoral map not unlike the Bickers and Berry forecast.
The D Tent seems to be a consensus of Nate Silver’s prediction of where the race will come down. The R Tent seems more divided on the degree of a Romney win.
However, there is a HUGE and DRAMATIC difference between the R and D tents. The R tent is relishing the electioneering and can’t stop talking about it. The D tent seems to want the election to be over, and their discussions don’t vary much from the 538 model or polls. If there was another week until Kermit Day, the R tent would eagerly take it while the D tent would groan.
