Mon 05 Nov
2012
Tomorrow is Election Day, so get out there and vote. Barring any major polling malfunctions, by the end of the day we’ll finally have an answer to the question of who will reside in The White House for the next four years. Almost as interestingly, tomorrow could also mark a definitive change in the way we look at Presidential campaigns, potentially for decades. In particular: if Nate Silver’s ongoing, deeply statistical analysis of the race at Fivethirtyeight turns out to be an accurate predictor of the final outcome, it may alter political punditry for a long, long time.
If you’re not familiar with Silver’s work, it’s probably a reasonable if gross characterization to say that he is a kind of ‘meta-pollster.’ Each day, he surveys the most recent state and national polls, aggregating their results using a sophisticated — but proprietary — statistical model that accounts for such factors as polling methodology, past accuracy and tendency to favor one party or another. The result is what some believe to be an exceedingly accurate picture of who is ‘winning’ at any given stage of the campaign — and, of course, a prediction of who will actually win at the close of Election Day.
Silver began doing this work in the lead-up to November 2008, and produced eye-popping results. His model correctly predicted the winner of forty-nine of the fifty states in the presidential election, and all thirty-five of the senate races held that year.
Whether that was pure luck or not is the question that will be answered when the results of tomorrow’s election are in. If his predictions are largely accurate, it will go a long way towards validating Silver’s approach. It’s my feeling too that if that happens there’s no going back; in at least the next few election cycles, you can expect to see much more attention paid to this sort of statistical evaluation of a campaign’s progress.
You might also expect to experience a lot of what I’ve felt as I followed along with Fivethirtyeight throughout this year: a growing dissatisfaction with the largely unquantified nature of traditional punditry. Silver’s blog makes for gripping reading, day in and day out, both because he is a good writer and because his evaluations of current polling events are so grounded in numbers, so well-argued, so rich with detail, that they seem far more rewarding than what normally passes for political analysis. Take for example his recent explanations of what it will take for Mitt Romney to outperform his polling and why he felt confident in saying that Barack Obama is the favorite. Both are thoughtful, compelling defenses of his thinking, but they’re really remarkable in that they exist at all. What other political commentators can be bothered to regularly lay out their thought processes so extensively?
After months of reading Fivethirtyeight on a daily basis, traditional political commentary is looking more and more outdated, even analog, to me. Most of it seems more like bloviation or superstition, and not true explication. My tolerance for it has been markedly reduced, whether it’s of the blue chip opinion columnist variety, or the more free-wheeling blog sort. My sense — or, to be fair, my hope — is that Fivethirtyeight is effectively disrupting the punditry industry, that in the coming years commentators will be expected to be much more quantitative than they are today.
On the other hand, Silver could turn out to be disastrously wrong tomorrow, in which case never mind.
If he’s wrong, it doesn’t change the fact that he provided insightful analysis based on actual data, educating people on how informed predictions (that may or may not come true) are made.
If he’s wrong, it doesn’t mean the blowhards win; just that the model needs to be improved.
Well said. Let the data have a voice. I don’t know why the news outlets give outlets to pundits who’s only job is spin. It’s not news any more, if it ever was.
If US political reporting is anything like the “horse race” reporting we get in Australia then lets all hope Silver’s predictions are spot on.
The sooner the political pundits are forced to abandon talking about polls and predictions because a statistician repeatedly reveals them to be fools, the sooner they might get replaced with journalists who can discuss *policy* instead.
All of the poll aggregators are in rough agreement with Nate Silver, so if they’re all wrong, there’s a fundamental problem with how polling was conducted for the past year or two. That could be the case, but the problem would be much more significant than Nate Silver not understanding how to aggregate polls.
“Whether that was pure luck or not is the question that will be answered when the results of tomorrow’s election are in.”
I understand that you’re just turning a phrase here, but the odds of flipping coins correctly 13 times in a row are incredibly low. About 1/100th of a percent. His model has been shockingly accurate.
But what’s more important is not whether he is correct in the calls, but the degree to which is his correct. Each state is given a “projected vote share.” It will be most interesting to to compare those projections to actual shares of vote.
I love Nate Silver and his approach is thoughtful and refreshing, especially in a field of journalism that often has sacrificed real analysis for inside baseball clichés.
Your post reminded of the recent Mother Jones article “Inside the Obama Campaign’s Hard Drive”—Not only are are campaign analysis from the outside changing but so is how data is driving these campaigns from the inside.
I guess I’d like like to see 538 doing it’s own polling in the next election as well, and see how it does against the likes of Pew, Rasmussen (ahem), Zogby et alia.
I will say the data visualizations around the election and serious data/quant analysis by the @nytimes team has been awesome. Please keep up the good work in other vastly more important areas like the environment, sustainable agriculture, climate change, etc.
My concern over polls is that some people will believe, and thus start to rely on, such polling, and therefore see their candidate as either a “sure thing” and not bother voting, while the.other side could see the prediction as a challenge and turn out in higher numbers than thy might have… In this case, though, I suppose a sort of “magic hand” effect could then prove the prediction wrong, then returning people to their normal behavior… Bah!!
More precisely, he is taking away the pundits’ ability to spin recent events by cherry-picking polls that support their point-of-view. However, it’s clear that in the process, Silver is also rattling the NYTimes political reporters as well as the pundits. Some have already made unsubstantiated comments that he works from home and doesn’t interact with the staff. It would not surprise me if he is snatched away by AOL or Daily Beast before the next round of federal elections.