A strange air of unreality has hung over the presidential campaign, in which a candidate who poses a mortal risk to the sanctity of American democracy and world peace has a disconcertingly high chance of winning, yet the prospect somehow has not felt real. Intelligent people continue to debate the exact size of this risk. Nate Silver has called the race close to a tie.
Economist Justin Wolfers replies that betting markets continue to favor Hillary Clinton.
Should the wisdom of the markets comfort Democrats, non-authoritarian Republicans, and other people who are fond of life on Earth as we know it? No, it probably shouldn’t. Betting markets do not appear to have any special knowledge. The prediction markets badly missed the outcome of the Brexit vote. Markets also gave Trump less than a 50 percent chance of securing his party’s nomination in February, and less than a 60 percent chance as recently as April. In both of these cases, prediction markets diverged from what the polling suggested, and the polling proved correct. The other commonality between these events is that conventional wisdom reflected the preferences of social and economic elites, who refused to take seriously beliefs held by very few people in their own circles.
So what do the polls say? FiveThirtyEight, Silver’s site, gives Clinton a 51.5 percent chance of winning. The Upshot, the New York Times calculator, gives her a 69 percent chance. (Both forecasts are based mostly on polling results.) Silver’s forecast makes Clinton the equivalent of a football team that is a 1-point favorite. The Upshot’s forecast makes her the equivalent of a 5.5-point favorite.
If your football team is either a 1-point favorite or a 5.5-point favorite, then you should be deeply concerned about the chance of losing. If the outcome is not a football game but the chance that the Executive branch falls under the control of a bigoted, uninformed, dictator-admiring man-child, you should be more than concerned. You should be freaked out.