Harris’s rise in the polls has stalled, while Trump holds steady

Yet in the days before the second presidential debate, Harris’s rise seems to have stalled while Trump’s polling numbers have mostly remained steady.

In the past week, a number of polls show that Harris has not improved further against Trump — some even show a small drop.

This includes the New York Times/Siena College poll released over the weekend in which Trump leads by one percentage point nationally; a Pew Research Center poll released on Monday showing a tie; and the CBS News/YouGov polls in Pennsylvania, Michigan and Wisconsin, which all show tighter races than what our polling averages currently suggest.

What’s changed in Harris’s fortunes isn’t entirely clear, but here are some possible explanations for why her poll numbers have begun to plateau and how we should think about the race heading into Tuesday’s debate:

Battleground states like Pennsylvania are just really close

In general, one should keep in mind that all of our state averages are currently within a normal-size polling error, but Pennsylvania, in particular, deserves a closer look considering it’s the battleground state with the most electoral votes at stake. It’s also one of the two states most likely to decide this year’s election. Put another way, Pennsylvania is something of a must-win state, because the candidate who loses it will have a very hard time otherwise putting together a path to 270 electoral votes, which is what’s needed to win.

Our average currently has Harris ahead in Pennsylvania by two percentage points, but the most recent polls are in Trump’s favor. The last three high-quality polls in the state, by CBS News/YouGov, the Times/SAY/YouGov and CNN, all show a tie or a smaller lead for Harris than our average.

To underscore the importance of winning the state: Even if Trump were to lose the popular vote — as he did in 2016 — if he wins Pennsylvania, that is likely to be enough for him to win the presidency. Harris would still have a path to victory, but it would be a tall order. For example, in one scenario, Harris would need to win Nevada, Arizona, Georgia and North Carolina — all states where Trump is currently either ahead or tied — to defeat him.

Harris’s ‘convention bounce’ may have already happened

According to data collected by the American Presidency Project at the University of California at Santa Barbara, political candidates have received an average convention bounce of five percentage points since 1948. But in recent presidential cycles, post-convention bounces have been smaller. In the week after the Democratic National Convention, Harris’s national polls have risen by about half a percentage point, according to our average. Our current levels of polarization and political engagement may mean that it is simply harder to change public opinion.

Notably, Trump didn’t really receive a convention bounce either — despite having survived an assassination attempt two days before the start of the Republican National Convention. The American Presidency Project shows just a one percentage-point increase in Trump’s poll numbers after the convention. Of course, Biden’s decision to withdraw from the race came just a few days later, so it’s possible that ate into any bump Trump may have received. Also consider that in the 2020 presidential race, Trump’s national polling numbers were mostly flat.

Another possibility to consider, however, is that Harris may have received her equivalent of the convention bounce when she became the nominee; polling shows support among Democratic-leaning independents ticked up for Harris before the convention. National polls have improved for Democrats about four percentage points on average since Harris replaced Biden as the nominee.

The race has changed. It’s now a toss-up.

Harris’s current polling numbers point to a plateau in her rise, but she’s still polling much better than Biden at any point this year.

Biden consistently trailed Trump in our national polling average, and except for Wisconsin, he was behind Trump in every battleground state we’re tracking. (He did hold the lead in Pennsylvania for the first half of the year, but Trump overtook him in May.) Moreover, Trump’s lead against Biden was outside a normal-size polling error in the Sun Belt, meaning in some states, Trump had put quite a bit of distance between him and Biden.

That’s not the case now. Since becoming the nominee, Harris has opened new paths for Democrats, making it less essential that they win the Rust Belt states of Pennsylvania, Wisconsin and Michigan. This, along with the likely decline of the Republicans’ electoral college advantage has put Harris in a much better position to win the presidency.

The race is now a toss-up — and that is a big change from where we were earlier this year. But a toss-up means that both Harris and Trump have plausible paths to the White House, and as we’ve seen from polls in recent weeks, it might take a lot to move the needle. We’ll see whether Tuesday’s debate can do that.