Skip to main content
ABC News
Sept. 29: As Iowa Goes, So May Go Romney’s Chances

Saturday, not Sunday, is the news media’s traditional day of rest — and so it is the slowest day of the week for polling.

But the national tracking polls were published on Saturday, and continued to show President Obama in a fairly strong position. He held at a six-point lead in the Gallup national tracking poll, although his approval rating dipped. He also maintained a rough seven-point advantage in the RAND Corporation’s online tracking poll. Mr. Obama also pulled ahead to take a two-point lead in the Rasmussen Reports tracking poll, which had differed from other polling firms by previously showing a tie. (Another national tracking poll, from Ipsos, is not regularly published on the weekends.)

We’re getting to the point in the campaign where a day on which the polls are in line with expectations is a winning one for Mr. Obama, since Mr. Romney trails in the race and now has just five full weeks to make the deficit up. Mr. Obama’s forecast rose slightly, to an 83.8 percent chance of winning the Electoral College, from 82.7 percent on Friday.

The Des Moines Register also published its highly regarded Iowa Poll on Saturday, which showed Mr. Obama with a four-point lead, 49 to 45. This result is quite consistent with other polls of Iowa published since the conventions, which also have shown Mr. Obama ahead by four points on average.

The only prior Des Moines Register poll this year, which was conducted in February, showed Mitt Romney up by two points instead. So this represents a favorable trend for Mr. Obama.

On the other hand, the same polling firm, Selzer & Co., conducted a national poll for Bloomberg recently, which gave Mr. Obama a six-point advantage. So they have Mr. Obama polling slightly worse in Iowa than he is nationally.

The FiveThirtyEight forecast concurs: we have Mr. Obama projected to win Iowa by 3.6 percentage points on Nov. 6, smaller than his 4.1-point advantage in the national race.

These are marginal differences, obviously, but they matter some in terms of the electoral math since any hope that Mr. Romney has of winning the Electoral College without Ohio probably requires him to win Iowa. In the simulations that we ran on Saturday, Mr. Romney won the election only 2 percent of the time that he lost Iowa.

This isn’t a good poll for Mr. Romney, but it does suggest that Iowa hasn’t gotten out of hand, and that it could trend back toward him if the national race does.

Iowa ranks seventh on our list of tipping-point states, but it packs a lot of bang for the buck because its television markets are fairly small and cheap to advertise in. We estimate that a dollar spent there will do twice as much to sway the Electoral College outcome as one spent in Florida.

Nate Silver is the founder and editor in chief of FiveThirtyEight.