A Florida Poll That Should Change the Way You Look at the Election

A big Trump lead in the state paradoxically adds to evidence of a smaller Electoral College edge for him. And a choice by pollsters may be causing them to miss state shifts.

We have three new New York Times/Siena College polls this morning. Only two look “normal.”

The national poll is one of the normal-looking results: It finds Kamala Harris ahead by four percentage points nationwide, 49 percent to 46 percent (these are rounded figures), compared with her three-point lead in The New York Times’s polling average. It’s her best national Times/Siena poll of the cycle.

Our first Texas poll of the cycle also looks “normal.” Donald J. Trump leads by six points, 50 percent to 44 percent, another tally that’s right in line with the polling average and close to the 2020 result.

And then there’s Florida.

Our first poll of the state this cycle finds Mr. Trump ahead by a staggering 13 points, 55 percent to 41 percent (again, rounded figures). This looks nothing like the other polls of the state. Heading into today, Mr. Trump led Florida in the Times average of all polls by just four points. None of the 11 polls fielded in September or October put him ahead by more than six points.

If you’re a longtime reader of this newsletter, you might be expecting me to hem and haw about whether this Florida result might be an outlier — a statistical fluke. Often, the smartest thing to do with the result of any one poll is simply to add it to the average and move on.

I’m not going to do that this time.

We’ve interviewed nearly 1,000 Floridians as part of our national polls over the last year, and Mr. Trump has had a considerable lead among Florida’s voters all along. As we’ve noted before, our national polling shows Mr. Trump excelling in states where Republicans performed well in the 2022 midterm elections — as they did in Florida (and New York, too).

As a result, this poll is not the usual outlier. It’s certainly an outlier compared with other polls, but it probably isn’t a fluke simply attributable to random chance. If we polled Florida again tomorrow, it’s certainly possible that Mr. Trump wouldn’t lead by 13 points (the accumulated national poll subsamples, for instance, have a larger sample and show him up by nine). But it’s hard to imagine his lead would finish at a mere four points, like yesterday’s polling average.

In a key respect, a big lead for Mr. Trump in Florida doesn’t affect the presidential election much. The state had already drifted off most analysts’ lists of core battleground states; Florida voted for Mr. Trump by more than three percentage points in 2020. And in the winner-take-all Electoral College, it doesn’t matter if you win Florida by four points or 13 points — either way, you get all the state’s 30 electoral votes.

So if a big lead in Florida doesn’t affect the election much, why should it change the way you think about it, as suggested in the headline?

It’s because this Florida result helps clarify both what’s happening in the race — and what’s happening in the polls.

The pandemic’s lasting effect

If Florida becomes more solidly Republican in 2024, it suggests that the upheaval during and after the pandemic has had a lasting effect on American politics.

This poll, after all, is far from the first indication of Republican strength in the Sunshine State. Republicans won a landslide victory here in the 2022 midterms, as the state was ground zero for the conservative reaction against lockdowns, vaccine mandates and “woke.”

These same issues didn’t do nearly as much to help Republicans elsewhere in the country. In fact, there were other states — like Michigan, Kansas and Pennsylvania — where the backlash against Mr. Trump’s stop-the-steal campaign and the Supreme Court’s decision to overturn Roe v. Wade seemed to outweigh them and propelled decisive Democratic victories.

What’s telling, though, is that the basic political pattern from the midterms still seems evident in the polling today. If the poll is right, Florida really has gone on a different path from Pennsylvania and Michigan. And since the paths diverged several years ago, the most straightforward explanation is that the fights over the pandemic, “woke,” abortion rights, crime, Mr. Trump’s effort to overturn the election and so on left a lasting mark on the electoral map.

Importantly, the pattern is consistent with the idea that Mr. Trump’s edge in the Electoral College relative to the popular vote has shrunk somewhat since 2020.

A 10-point gain for Mr. Trump in Florida and New York — where Siena College also shows enduring Republican strength, though the state remains safely Democratic — would be enough to shave about one (inefficient) point off Vice President Harris’s lead in the popular vote.

The state of the polls, and polling

There’s another reason the Florida result is important: It says a lot about the state of the polls.

As mentioned earlier, there have been 11 polls of Florida fielded in September or October, and they show Mr. Trump with leads between one and six percentage points — nothing resembling a 13-point lead.

Obviously, it’s possible that the Times/Siena poll will differ wildly from the result in November and that the other 11 polls will come closer. That said, these 11 recent polls aren’t traditional “gold standard” surveys. And there’s a very plausible explanation for why they’re producing such different results: They usually employ a methodological choice called “weighting on recall vote” and we do not.

If you missed our detailed analysis on this choice this weekend, “weighting on recall vote” is a technique in which pollsters ask respondents how they voted in the last election, and then weight the number of Biden ’20 or Trump ’20 voters to match the outcome of that election. In this case, the pollster would ensure that there were about three percentage points more Trump ’20 voters than Biden ’20 voters, because Trump won the state by about three points in 2020. Many pollsters believe the technique might help them avoid underestimating Mr. Trump yet again.

As I explained, there are a lot of challenges with this approach. But the Florida poll reveals an entirely separate problem: What happens if the makeup of the electorate changes?

Over the last four years, Florida’s party registration has shifted significantly, going from D+ 0.7 to R+ 7.5. According to L2, a nonpartisan political data vendor, people who moved to Florida since January 2021 have registered as Republicans by a margin of more than two to one, 49 percent to 22 percent. The broader group of new registrants, which includes young people who probably didn’t vote in the last election, registered Republican by a smaller but still significant margin, 39 percent to 22 percent.

The bottom line: This is probably not a Trump +3 electorate anymore, even if the recall vote measure itself was accurate. That doesn’t mean the Times/Siena poll is “right” — it’s an inherently imprecise measurement — but it does make it harder to discount it based on the other polls.

In the past, almost every poll would have had a chance to capture this kind of shift. In fact, that’s the whole point of polling. Now, many polls are designed to ensure they don’t show it at all.