Folks on the Romney side are upset with data coming from Nate Silver and his Fivethirtyeight blog. Over the last few days Obama has ticked up in his chances to be re-elected according to the models Silver uses to predict the probable outcome. Now there is now need to fully understand the make up of the methodology used to arrive at these projections but it is interesting how vitriolic the response has been to the blog's conclusions. Republicans loved Silver when he was predicting their prior victories in various races but have attacked his conclusions of late due to the nature of their conservative bias.
Silver is not a partisan but merely revealing the probability of Obama defeating Romney using the information he gleans from data mining certain criteria. He is not telling you what is going to happen but what the models say is most likely to happen. Vegas does not tell you if the Chicago Bears will win the Superbowl but only the chance of them winning; between 1/10 and 1/12 right now. Silver says there is roughly an 8 in 10 chance Obama will win the electoral college and a 50/50 chance he will win the popular vote.
Now Vegas is a bit different in the fact that it does not try to predict the winner as much as putting up a line which will encourage the most betting on both sides. Predicting the path of where the electorate, bettors, and investors with travel are correspondingly determined by votes, bets, and price movement. Building models on how to capture a probable path is what many of us have successfully done. But everyday is different and probabilities are not realities.
