When political analysts make it up

One of the great problems of having political analysts on the payroll is that they often have to say something even when there’s nothing to say. That’s when they make it up and hope you won’t notice.

We noticed.

Political analysts can be fairly bad at math, especially when they’re trying to jam their reality into an equation that doesn’t add up. Last night’s New Hampshire results provide the examples.

William Kristol got the ball rolling last night by declaring Romney’s vote total to be “worrisome,” noting that turnout was down.

The problem here appears to be math. Kristol said Romney’s vote total was about the same as, or a little less than, what he got four years ago. And that’s true, if by “the same as or a little less than” you actually mean 26 percent more. Romney garnered 97,532 votes last night, compared to 75,546 four years ago. It was also 10 percent more than the amount that John McCain received in New Hampshire four years ago.

Kristol was working for FoxNews, but today, NPR didn’t fare much better in solving the math problem.

In their weekly Political Junkie chat, political editor Ken Rudin and NPR Talk of the Nation host Neal Conan wanted to make a similar point to Kristol’s (the Republicans are in trouble) and if he had to slaughter the rules of statistics, that’s just what he had to do.

“If you take out all the new voters that Ron Paul brought to the caucuses in Iowa, and take out the new voters that Ron Paul brought to the primary in New Hampshire, turnout was actually lower than 2008 in both places,” Conan said, before Rudin agreed with him without question. Wrong.

Let’s think about Conan’s qualifier for a second. Why would you remove a piece of reality under the assumption that it would reveal an underlying reality? And why would you remove “new” voters from the comparison and not remove all the “new” voters from the 2008 primary to — inelegantly and inaccurately — achieve a more fair comparison?

“If you take out the killings, Washington actually has a very very low crime rate,” then Mayor Marion Barry once said of his famously crime-ridden city.

Similarly, one could say, if you took out 19 losses from last year’s Minnesota Twins season, they actually turned in a winning season. The problem, of course, is you can’t remove a piece of reality in order to create a clearer reality. The Twins were pretty bad and as noted political analyst Bill Parcells once said, “you are what your record says you are.”

The last time there was a Republican primary in New Hampshire without a Democratic contest of any note was 1996. In that election, 282,697 Republican votes were cast. Last night 248,485 votes were cast for the major candidates, that’s about 7,000 more than in 2008.

Is that significant? Sure, because as Linda Fowler, professor of government at Dartmouth College in Hanover, New Hampshire, said on Midday yesterday, New Hampshire has been turning more purple than its traditional red. So an increase in turnout is pretty impressive.