The Race for the White House: It Isn’t a Tie
Media outlets have a vested interest in making it seem as though the race between President Obama and Governor Romney is … well, a race. I mean, blowouts aren’t interesting. Who wants to follow an election where one party leads wire to wire and the other never manages to catch up? Both campaigns have a similar interest. It’s awfully hard to drive turnout if American voters, famously apathetic to begin with, think the outcome is already predetermined.
It’s certainly true that the race is close. Nationwide, the President has never held a consistent lead of more than 2 or 3 points in the polls since the close of the Republican primaries. But, partially due to everybody’s incentive in keeping the race interesting, everybody’s been conflating a 2 point Obama advantage with a tie. It’s not a tie. It’s a two point advantage. And the closer we get to election day, the harder it becomes for Romney to make up that deficit. The Obama campaign won’t play it this way, because they are genuinely nervous that something might go wrong, but if he wanted to, Obama could win this thing by running out the clock. In other words, if the election were held today - or tomorrow, or the day after - Obama would win. And handily.
“Statistical tie.” “Within the margin of error.” “Could go either way.” You hear these phrases bandied about a lot on cable and on the Sunday morning chat shows. While they’re all technically true, they all bury the lede: Obama is winning this race, and he’s been ahead the entire time. Check RealClearPolitics’ polling average: Obama has held a consistent lead all year. In all of 2012, there has not been a single day where Romney has held a net average lead in the nationwide polls. Not one. He drew into a tie for the first time with the President last week - for one day - but that’s the best he’s ever managed. And Nate Silver’s FiveThirtyEight forecasting model at the New York Times has rated Obama as no worse than a 2-to-1 favorite to win reelection since May (it rates him at roughly a 5-to-1 favorite as of today). The lead has been narrow, but it has been incredibly stubborn. Romney’s close, but hey: horseshoes and hand grenades, right?
Is there such a thing as a margin of error? Should we take individual polls and individual data points with a grain of salt? Are some polls just totally wrong? Absolutely. But a pattern like this - an aggregated trendline reflecting the results of hundreds of polls across months and months - is impossible to ignore. Romney has never held a nationwide popular vote lead in aggregate while Obama has led every single day in 2012, and yet we act like Romney is running neck and neck with the President. He’s not, and it’s time to stop thinking of the race like that. Don’t get me wrong: Romney could still win. But his chances are, as of right now, probably somewhere between one in five and one in six. Obama is the clear favorite at this point. And, if the race keeps going like it’s going, he will win a second term.
That’s the trick with polls of Presidential elections: margins of error are usually 4 or 5 percentage points - anything within that range is a “tie” as far as a single poll is concerned - but a 5 point popular vote win on election day would qualify as a blowout. You reconcile this tension by aggregating the polls - much as RCP and 538 have done - and you find that this race isn’t tied at all. In fact, somebody is winning. And he has been winning from the very beginning.
Tags: Obama Romney Election 201
This work is licensed under a Creative Commons Attribution 3.0 License.