However, social science is not like physics--human behavior is not that predictable

rickyp wrote:freeman

However, social science is not like physics--human behavior is not that predictable

really? Maybe its not as predictable as figuring a ballistic path but its predicitable enough that enormous industries (marketing, advertsing, packaging etc) make companies profitable based on that predictability. Think of any others? (Safety, security, investment etc.) Its really all based on the fact that past behaviours tend to provide guidance to how people behave in the future...
You're seling Siver short. His model incorporates a lot more complexitiy than any other. He's not predicting an election national poll performance or indicators like "presidential approval". he's looking at 51 seperate electons, and analyzing polls as they go through. In the end it will depend upon the final results... And he's a true stats man. he doesn't say Obama is going to win., He says there's an 81% chance that Obama will win. (The elctoral college).
Since April, when I made my prediction based on Silver, only one state seems to be likely to change. (Florida). All the fury and noise and so little has changed.

The most current Public Policy Polling survey, released Saturday, has Obama up only one point, 49–48. That poll is given a weighting under Silver’s model of .95201. The PPP poll taken last weekend had Obama up five, 51–46. This poll is a week older but has a weighting of 1.15569.

The NBC/Marist Ohio poll conducted twelve days ago has a higher weighting attached to it (1.31395) than eight of the nine polls taken since. The poll from twelve days ago also, coincidentally enough, is Obama’s best recent poll in Ohio, because of a Democratic party-identification advantage of eleven points. By contrast, the Rasmussen poll from eight days later, which has a larger sample size, more recent field dates, but has an even party-identification split between Democrats and Republicans, has a weighting of .88826, lower than any other poll taken in the last nine days.

In other words, the results of "The Model" are dependent on Silver's subjective decisions on how each poll should be "weighted," i.e. how "accurate" Silver believes the poll is. After two decades in politics, I'm shocked that someone would think an almost two-week old poll is a more accurate picture of the race than a current poll from the same pollster. This isn't a model; this is an elaborate mathematical equation designed to generate what Silver "thinks" the outcome ought to be.

This was evident in Silver's predictions for the 2010 elections. While he acknowledged that the GOP was favored to win control of the House, he pegged their chances of picking up more than 60 seats at only 25%. In other words, there was a 75% probability that they would pick up less than 60 seats. They picked up 64.

Another serious problem for "The Model" is that Silver operates from a premise that state polls are, prima facie, more accurate than national polls. It is only in the last decade or so that we've had robust polling at the state and national level, so I'm not really certain we have enough data-points to back up this assumption. Yes, the presidential election is fought state-by-state, but state and national polling tends, over time, tend to move in concert. The reason a state is considered a "swing" state is that it generally matches up with the country as a whole.

Doctor Fate wrote:You have blind faith.

geojanes wrote:

Doctor Fate wrote:You have blind faith.

Now isn't that the pot calling the kettle . . . oops, I thought I was just thinking that.

Is he biased? Does he favor one poll over another? Is it all a proven formula

Friday’s polling should make it easy to discern why Mr. Obama has the Electoral College advantage. There were 22 polls of swing states published Friday. Of these, Mr. Obama led in 19 polls, and two showed a tie. Mitt Romney led in just one of the surveys, a Mason-Dixon poll of Florida.