Our forecast puts Kamala Harris and Donald Trump neck and neck

THE MOOD among Democrats, as they prepare to gather for their national convention in Chicago next week, is ebullient. Kamalamentum has transformed the presidential contest. Meanwhile, the Republicans are finding it hard to adjust to the new dynamics: the glitch that delayed the start of Donald Trump’s conversation with Elon Musk, carried live on X on August 12th, was emblematic of a campaign that is no longer going to plan. Yet if there is one lesson from this wild election year, it is that things can change with remarkable speed.

It was not so long ago that some Democrats were deluding themselves that, even with a mumbling 81-year-old Joe Biden as their candidate, the election was in essence a coin flip. In mid-June, when The Economist launched its election model, its findings should have been a wake-up call. The forecast—which takes into account information from opinion polls and “fundamental” factors such as the state of the economy, presidential approval ratings and the share of swing voters—was that Mr Biden’s chance of winning was in fact well below 50:50. By the time Mr Biden dropped out of the race on July 21st, a few weeks after a calamitous debate with Mr Trump, the model showed that his likelihood of victory had plunged still further.

Since then Kamala Harris has made a stunning start to her campaign. Our nationwide poll tracker now gives her a three-point lead over Mr Trump, who had been ahead of his Democratic rival since last October. Ms Harris has sharply improved the position in swing states, too. The latest New York Times/Siena College poll in Michigan, Pennsylvania and Wisconsin puts her four points ahead in all three states. And now that enough data are available on the new matchup, we can relaunch our forecast model (with a few tweaks) for the Trump-Harris election. It, too, shows that Ms Harris has moved ahead.

Reality check

But only just. Our model gives her about a 52% chance of winning. It shows that her chances have improved by about ten points since the beginning of the month. These findings are in line with those of another forecasting model, run by Nate Silver, an election-statistics guru. They mean that, with less than three months to go to polling day, the race is wide open.

So small shifts between now and November could swing it one way or the other. Of late the shifting has been all in Ms Harris’s favour, helped by the novelty and energy of her candidacy as well as by a number of unforced errors from Mr Trump. The former president did not help his cause by, for example, gratuitously attacking the popular Republican governor of Georgia, or questioning Ms Harris’s sense of her own ethnicity.

But it would be naive to assume that Ms Harris’s current momentum, impressive as it has been, will carry her all the way to the White House. Republicans will surely hone their attacks on her, especially on issues such as immigration and inflation, where she is particularly vulnerable. Mr Trump will have an opportunity to confront her face to face in at least one debate next month. Above all, Ms Harris can expect more scrutiny than she has had so far, and rightly so: voters need to know what she truly believes and what policies they can expect from her on everything from defending America to tackling crime and handling the economy.

A feel-good convention—casting the election as a choice between the future and the past, and between hope and gloom—is all very well. But it risks giving Democrats a false sense of inevitability. An antidote to such delusion is data. Our forecast model, updated daily, will capture an objective assessment of the likely outcome of the presidential election. And right now, it points to a coin flip. ■