Silver vs Academic Election Forecasters
Courtesy of Robert Waldmann at Angry Bear
In a recent post, I mentioned a blog post by Nate Silver which I hadn’t managed to google. It is here.
Models Based on ‘Fundamentals’ Have Failed at Predicting Presidential Elections
I think this is a serious effort worthy of publication in a top political science journal (not that anyone asked me). Silver attempted to find all election forecasts in a well defined population (there is a long boring definition in the post) and see how they performed. Notably all of the models showed excellent fit within the sample used to estimate parameters. He is particularly hard on models which use only (mostly economic) fundamentals not supplemented with data from polls.
The “fundamentals” models, in fact, have had almost no predictive power at all. Over this 16-year period, there has been no relationship between the vote they forecast for the incumbent candidate and how well he actually did — even though some of them claimed to explain as much as 90 percent of voting results.
The R-squared of the regression of outcome on forecast is 0.0371.
It is certainly true that Silver has limited respect for prominent academics who attempt to forecast presidential elections using fundamentals. But it seems to me that his lack of deference is justified by solid data analysis.
See Also:
When FiveThirtyEight last issued a U.S. Senate forecast — way back in July — we concluded the race for Senate control was a toss-up. That was a little ahead of the conventional wisdom at the time, which characterized the Democrats as vulnerable but more likely than not to retain the chamber.
Our new forecast goes a half-step further: We think the Republicans are now slight favorites to win at least six seats and capture the chamber. The Democrats’ position has deteriorated somewhat since last summer, with President Obama’s approval ratings down to 42 or 43 percent from an average of about 45 percent before. Furthermore, as compared with 2010 or 2012, the GOP has done a better job of recruiting credible candidates, with some exceptions.
As always, we encourage you to read this analysis with some caution. Republicans have great opportunities in a number of states, but only in West Virginia, South Dakota, Montana and Arkansas do we rate the races as clearly leaning their way. Republicans will also have to win at least two toss-up races, perhaps in Alaska, North Carolina or Michigan, or to convert states such as New Hampshire into that category. And they’ll have to avoid taking losses of their own in Georgia and Kentucky, where the fundamentals favor them but recent polls show extremely competitive races.
Keep reading: FiveThirtyEight Senate Forecast: GOP Is Slight Favorite in Race for Senate Control
