Election Summary

I'm intensely interested in the election. There's boundless information on the internet, but I have difficulty determining what it means in terms of who is winning the election. For example, whether McCain or Obama is ahead in a national poll is irrelevant to determining the electoral college winner. Here's a summary, therefore, of the bottom line of the election as I've distilled it from various expert sites.

A summary:

• Obama needs 269 electoral college votes to win. McCain needs 270. Ties are broken by the house which will be Democratic.
• The polls currently suggest McCain would win the election if held today.
• Obama's major victory scenarios are to win Ohio, Florida, OR Nevada in addition to the states he is currently projected to win. He is currently project to (barely) win Michigan and Pennsylvania.
  
• Thus, Obama could win by simply winning the Kerry states plus New Mexico, Iowa and Nevada (which would create a 269-269 tie in the electoral college and the house Democrats would select Obama.

The polls currently show McCain with a 1-2% edge nationally. The prediction markets currently favor McCain by 52% to 46% - roughly translated as odds of victory. Yet, the major underlying fundamentals are strongly Democratic, so how reliable are these predictors? Or, perhaps, more to the point, can we reasonably hope that Obama will easily win and the numbers are somewhow wrong?

Prediction markets are persuasive to me in general. The pollling numbers vary according to who's doing the polling. And given that Democrats have dramatically outregistered Republicans this year, one might think that polls are wrong since some of them seem to oversample Republicans: that is, there are alot more registered democrats now yet many polls sample equivalent or near-equivalent amounts of democrats and republicans. Well, this leads to issues that mean if the race is close in the polls, nothing will be decided until Election day, which is a good thing I suppose.

The two big unknowns are the massive voter suppression efforts of the Republicans and the Get-out-the-vote (GOTV) effort of the Democrats. I personally take this as roughly off-setting and place my faith in the prediction markets to judge the state of the race. I mention them here because if you read about just the one you might be either unduly pessimstic or optimistic about the results.

Republicans are much better at winning elections than Democrats. Even though the money advantage is with the democrats this year, the fundamentals of the money issue are with the republicans, by which I mean that they seem to have unlimited money for voter suppression activities. In Wisconsin, Michigan, Ohio, Florida, Virginia, and presumably all other swing states the Republicans are undertaking systemic efforts to disqualify valid democratic voters from voting in the election. If you peruse the links above you'll find the tactics vary from state to state. There is simply no way to reliably predict their effect.

On the other side of the coin is the Obama GOTV effort which seems to be remarkably strong relative to past presidential races. Yet, here too, there seems to be no reliable way to predict the effect of these. It also seems intuitive that the fundamentals of this election - which is also with the democrats - should spur on better than average democratic turnout, but this advantage, if true, also eludes reliabel quantification.

The bottom line: unless the polls show a 5% advantage for one candidate or the other before election day hold your breath and pray.