There’s been a little back and forth between the political press and the nerds about Mitt Romney’s “momentum” – what it means, if it exits, and the concept of momentum as it relates to politics in general.
The nerds, in this case both 538 and the Princeton Election Consortium, note that whatever momentum Mitt Romney had after the first debate had pretty much stopped by the second debate, and they also argue, essentially, that these races tend to regress toward the mean: if the entire year has shown Obama slightly ahead, then that’s probably where things will end up.
Sam Wang at Princeton has some bonus fun with people like David Brooks or John Dickerson who talk anecdotally about momentum, or energy, by looking at the candidates’ rallies. Or bluster. Or by sniffing the air.
In Brooks’ case, he talks about a “whiff” of momentum for Romney (this, after Romney was shellacked at Hofstra University on Monday night) while calling Obama a “bit more likely” to win. (Wang corrects this to “a lot more likely.”) John Dickerson of Slate wrote about a “fight over momentum,” referring to fired up crowds for Romney. Wang took that apart, too.
First, I’ve talked before about the danger of inferring anything from the enthusiasm of a crowd:
24 years ago, I was a volunteer for the Rhode Island Mondale campaign – the year, of course, of Geraldine Ferraro. Those of us who were true believers ignored the braying of the press about her qualifications and her husband’s real estate transactions, and felt she had held her own against Poppy Bush in their vice-Presidential debate. And, best of all, we got an in-state event with her: on the Sunday before election day*, Geraldine Ferraro came to Rhode Island Community College, and speaking to an overflow crowd of 3,500, predicted an upset victory.
Well, so much for that.
My point is this: when surrounded by thousands of people who agree with you, it’s easy to believe that there’s no one who disagrees. Those who get sucked into theSarah Palin – megastar hype would do well to remember that – even as she headlines one raucus rally after another.
Second. Jon Chait noted earlier this week that a lot of the Romney bluster was a con:
Romney is carefully attempting to project an atmosphere of momentum, in the hopes of winning positive media coverage and, thus, creating a self-fulfilling prophecy.
Over the last week, Romney’s campaign has orchestrated a series of high-profile gambits in order to feed its momentum narrative. Last week, for instance, Romney’s campaign blared out the news that it was pulling resources out of North Carolina. The battleground was shifting! Romney on the offensive! On closer inspection, it turned out that Romney was shifting exactly one staffer. It is true that Romney leads in North Carolina, and it is probably his most favorable battleground state. But the decision to have a staffer move out of state, with a marching band and sound trucks in tow to spread the news far and wide, signals a deliberate strategy to create a narrative….
Karl Rove employed exactly this strategy in 2000. As we now know, the race was excruciatingly close, and Al Gore won the national vote by half a percentage point. But at the time, Bush projected a jaunty air of confidence. Rove publicly predicted Bush would win 320 electoral votes. Bush even spent the final days stumping in California, supposedly because he was so sure of victory he wanted an icing-on-the-cake win in a deep blue state. Campaign reporters generally fell for Bush’s spin, portraying him as riding the winds of momentum and likewise presenting Al Gore as desperate.
Bonus fun (in the “Media likes a horserace” category): Yesterday, CNN released a poll that showed Obama up by 4% in Ohio, and called it a tie. See for yourself:
This is not correct. The AP Style guide suggests that this result be referred to as “about even.” If you want a better way of evaluating polling leads vs. Margin of error, look here (from Kevin Drum):
The percentages in the boxes are the probability that a candidate is ahead given a combination of polling lead and margin of error. The CNN poll actually has a 3% margin of error, so CNN’s poll really says that’s it’s 90% likely that Barack Obama is ahead in Ohio.
But I digress.
In 2008, (which was, granted, pretty much over after McCain “suspended” his campaign), they way I re-grounded myself was to visit Nate Silver’s blog (which was also, as in this case, doing some extraordinary field reporting), to get a look at straight data.
This is how it will be, I imagine, for the next ten days. Lots of cowpies flying around, and me in my bunker, waiting for the poll nerds to refresh their pages.