Helmut Norpoth’s “prediction model” shows that President Donald Trump has a 91% chance of being reelected and also shows he will get 362 electoral votes [“Trump not bad bet to win in 2020,” Opinion, April 17]. The model is no doubt based on a statistical and historical theory that the professor outlines. Popularized by Mark Twain said there are three types of lies: “lies, damn lies and statistics.” The professor never mentions two factors that seriously undermine his conclusion.
One is that we are entering a historic recession/depression, the likes of which the country perhaps has never seen. In the past two such financial calamities, 1932 and 2008, which were nowhere as severe as this one, the party in power was defeated.
Second, while the professor doesn’t lay out which states Trump must win to get to 362 electoral votes, he would have to carry all of the states he won in 2016 (three of which were blue states he won by a whisker) plus New Mexico, Nevada, Colorado, Vermont, Virginia, Minnesota, Maine and New Hampshire. I’d say there’s a better than 91% chance that won’t happen. So do most present polls.
Helmut Norpoth defended his forecasting model using past examples, and cited a recent Gallup poll to prove Trump’s approval rating would be helped by the coronavirus crisis, as he’d be cast as a wartime president. I wonder if the professor was red-faced to discover that the Gallup poll he cited was updated that very day, giving Trump his biggest drop ever in approval ratings.
And another factor will wreak havoc with Norpoth’s model. Once in office, other presidents soften their partisan rhetoric to prove they are working hard for all Americans, not just their voters. Trump has done the opposite, doubling down on his hostility and divisiveness. This president, who won swing states by slim margins, has done nothing to cast a wider net among voters, and everything to drive away all but the most hard-core supporters. He flaunts his hatred, mocks his detractors, and wields partisanship with the cruel edge of power. Yet there are simply not enough in his loyal base to push him over the finish line. Not this time. Not when he’s had four years to show us who he really is.
After reading Helmut Norpoth’s op-ed, I wonder whether Norpoth’s algorithm considered President Donald Trump’s having been impeached. It is a question that he and, more important, voters must factor in any preelection prognostications and in the upcoming election.
William R. Kearns,