As of Friday at 10 a.m. Eastern, our average of national polls says Harris has the support of 45.0 percent of voters, while Trump garners 43.5 percent.
That 1.5-percentage-point lead is within our average’s uncertainty interval, which you can think of as a sort of margin of error for our polling averages.
It’s a little weird that they say Harris is “tied” with trump, even though she’s ahead by 1.5%. That seems like a big deal. Margin of error is important, but it’s just factually true that Vice President Harris is up by an average of 1.5%.
I looked back at how 538 treated polls when trump was up by a similar amount:
https://abcnews.go.com/538/polls-after-presidential-debate/story?id=111610497
In 538’s national polling average, Trump now leads by 1.4 percentage points over Biden, while the two candidates were just about tied on June 27, the day of the debate.
So Harris up by 1.5% is actually “tied”, but trump up on Biden by 1.4% is “leads” (and explicitly different from “tied”!). No mention of margin of error in that paragraph.
🤔🤔🤔
Polls are almost meaningless.
There’s some value versus knowing nothing. But until October early voting actually begins, not worth obsessing over. To be fair, their forecast was still one of the better ones for that infamous election, dropping around 60-40 Clinton during election day (NYT was still 90-10). And I’m sure many things were learned from it, maybe even over-corrected, based on '20-22. I’ll be following Nate’s Silver Bulletin this year (he left 538 and took his algorithms).