There’s still hours of voting yet to go before we begin to find out who our next president will be.
But this much we can be sure of: Nate Silver was right.
If Donald Trump wins, Silver, the number-crunching genius behind the popular website FiveThirtyEight, called it. He said on ABC News on Sunday that Hillary Clinton is “one state away from potentially losing the electoral college.”
And if Clinton wins the electoral college? Silver predicted that, too. His website on Tuesday gave her a 71 percent chance of winning the electoral vote.
Same with the popular vote. Silver has foreshadowed a Trump victory: He wrote on Friday that “Trump is about 3 points behind Clinton — and 3-point polling errors happen pretty often.” And he also predicted a Trump defeat, giving Clinton an 81 percent chance of winning the popular vote as of Tuesday.
This raises a pertinent question: How can Silver be predicting a healthy Clinton victory while also noting she is in danger of losing (and simultaneously making allowance for the possibility she’ll win in a landslide)? Well, this is the result of a complex statistical method known as covering your bases. Or your backside.
Silver became a statistical sensation with his highly accurate models predicting President Obama’s victories in 2008 and 2012. But this year, he completely missed Trump, pegging the businessman’s chance of winning the GOP nomination at “considerably” less than 20 percent.
And Silver, once bitten, is twice shy. His probability that Clinton will win the presidency has bounced all over the lot, from percentages in the 50s to the high 80s, as developments occur and opinion polls shift and the model reacts to the shifting polls. His chances of a Clinton win have been much lower than other forecasters’, such as the New York Times Upshot (85 percent), Daily Kos (92 percent), Huffington Post (98 percent), a Princeton University modeler (99 percent) and, my favorite, Charlie Cook (“it’s a really big number”).
But I don’t fault Silver for his caution. It’s honest. What it really says is he doesn’t know with much confidence what’s going to happen. That’s because there’s a lot of human caprice and whim in electoral behavior that can’t always be explained or predicted with scientific precision. Politics ain’t moneyball. Good-quality polls give an accurate sense of where a political race is at a point in time, but they don’t predict the future.
Predictive models, generally based on historical patterns, work until they don’t. American University Professor Allan J. Lichtman, whose model predicted the popular-vote winner in every election since 1984, is still convinced Trump will win.
Explaining why he missed Trump, Silver cited his unprecedented dominance of media coverage and the failure of the Republican Party to coordinate against Trump. Like Silver, I also predicted that Trump wouldn’t win the GOP nomination, but I didn’t dress up my reason with statistical justification. I had more confidence in Republican voters — that they wouldn’t nominate such a flawed and hateful candidate — than turned out to be warranted.
In his hedged forecasts this time, Silver appears to be acknowledging that polling and historical patterns don’t always capture what John Maynard Keynes, in his classic 1936 economic General Theory, described as “animal spirits.”
There is, Keynes wrote, “the instability due to the characteristic of human nature that a large proportion of our positive activities depend on spontaneous optimism rather than on a mathematical expectation, whether moral or hedonistic or economic. Most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as a result of animal spirits — of a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities.”
Big Data changes this, a little. Silver’s historical trends and the other items he stirs into his special sauce try to capture that. It works, sometimes. Then the animal spirits change, as they did with Trump.
The Huffington Post’s Ryan Grim took a shot at Silver in recent days, calling him an “outlier” who artificially elevated Trump’s prospects by tinkering with poll results in the form of “trend line adjustment.” Wrote Grim: “He may end up being right, but he’s just guessing.”
Silver tweeted that the criticism was “so f---ing idiotic and irresponsible” and said he makes his adjustments because “that’s what works best emperically [sic].”
Maybe it does. Maybe it will again. But Silver is just guessing. And, in the run-up to this election, he made so many guesses that at least one of them has to be correct.