Monday, October 22, 2012

Obama, Ohio and Nate Silver



Nate Silver, election polling expert at the NY Times, is still, stubbornly, talking about a 70 percent chance that Obama will win in Ohio. I have examined all the polls that I know of and I can’t call it anything but 50-50. Silver is a genius, there’s no doubt about that, but I don’t know what he’s reading (or drinking).
by Charlie Leck

If you don’t know about Nate Silver, you might want to read this
“character sketch” in the Boston Globe about him…
Nate Silver: Stats whiz remakes political punditry.

Nate Silver says, right now, there is a 65.7 percent chance that President Obama will be reelected. So what?

Just get this: In the 2008 election, Nate Silver called 49 states correctly in the presidential election (he only missed in Indiana). He also picked all 35 U.S. Senate races correctly. As a result, in 2009 Time Magazine listed him as one of the world’s most influential people. Silver, a baseball statistician at the time, had rejected a great deal of the conventional wisdom about polling. Now, at 34 years of age, Silver is followed closely. He explains which of the polling organizations have bad records and which have good ones – which are insider organizations with particular views and bents – and which ones can be relied quiet heavily.

It was because of Silver that I became something of a polling freak and have gotten overly attracted to watching the poll results come in – and I use the word “overly” very advisedly because I know it has reached the point of obsession with me. I go nuts when I can’t get up-to-date polling results on some races (as I haven’t been able to for the Michele Bachmann/Jim Graves congressional race here in Minnesota).

Silver has a blog, Five Thirty Eight, that is very popular during election campaigns. He also has a very recent book (September 2012) that I keep close at hand on my desk (The Signal and the Noise: Why So Many Predictions Fail – But Some Don’t).

Remember, Gallup came out last week with polling results that showed challenger Mitt Romney had cracked the 50 percent mark for the first time and is now leading Obama in the race. It caused a great stir on political blogs all around the country. Silver treated it calmly and analyzed the Gallup results reasonably in Gallup vs .the World.

And, still more recently he wrote…

“We’re now in the political equivalent of the eighth inning, however. A run scored in the eighth inning is potentially much more important than one in the first.
“The reason I say ‘potentially’ is that it makes a tremendous difference depending what the score is. In a blowout, the eighth inning won’t matter at all. A team down 9-1 is almost certainly going to lose; but so will one that gets a solo home run and trails 9-2 instead.
“…But if the score is tied, or it it’s a one-rune game, a run scored in the eighth will make a huge difference.
“That’s where we find ourselves right now in the presidential race.”

I don’t suggest you get involved in the Nate Silver models and approach unless you want to drive yourself a little bit whacky. Essentially, Silver weights the polling organizations and sometimes discounts their results by a certain percentage – and sometimes he gives them a percentage of a point more weight and upgrades them. He also gives a certain amount of weight to political tendencies within states. When there appears to be one renegade polling organization that comes up with a remarkably different set of results than the other polls, Silver doesn’t automatically dismiss that poll, but he wants to clearly understand just why there is a difference and how much weight he can give to that difference.

I’ve tried to hone my own approach by understanding Silver’s. I’m a long way from fully understanding him, but I’m have a whale of a lot of fun. Don’tcha know?


_________________________
Why not become a follower?
If you read my blog regularly, why not become a follower? All you have to do is click in the upper right hand corner and establish a simple means of communication. Then you'll be informed every time a new blog is posted here. If all that's confusing, here's Google's explanation of how to do it! If you don’t want to post comments on the blog, but would like to communicate with me about it, send me an email if you’d like.

No comments:

Post a Comment