AILSA CHANG, HOST:
All right. Tens of millions of people have voted already with one day left until Election Day. But those votes, and the ones that will be cast tomorrow, are still a mystery, at least for now. Polls tell us a lot of the races are going to be very close. Prediction markets have their own take on this. But how do we sift through all of these different clues? Well, that is what NPR senior political editor and correspondent Domenico Montanaro's here for. Hi, Domenico.
DOMENICO MONTANARO, BYLINE: Hey. Good to be with you.
CHANG: Good to be with you. OK. So there are all kinds of numbers out there right now. Most of them are showing that Democrats have a sizable advantage at this point. Is that what you're seeing, as well?
MONTANARO: Yeah. I mean, there's tons of data out there and lots of nerves, of course, understandably. And that's with good reason on either side. You know, Democrats are the clear favorites to win the House. So putting that there, the Senate is a toss-up. Polling is showing things trending Democrats' way, but a lot of these races are still within the margin of error, and we really can't predict what's going to happen.
CHANG: Exactly. OK. So let's talk about those polls because they are driving so much of the storyline right now. Like you said, they are showing that Democrats are doing well, but there's always a but - right? - when it comes to polling.
MONTANARO: Of course. You know, it's important to remember polls are not supposed to be predictive. You know, they're showing us what's happening now, or rather, when they were conducted. They're are more estimates than exact because they have margins of error. If something is plus or minus three or four points, that's a range of six to eight points. It could be three or four points lower or three or four points higher. You know, horse race polls, especially at the state level, I have to say, are pretty flawed, though, and not even the best use of polling.
CHANG: OK. Well, with all of that in mind, what is the latest, with the election literally tomorrow?
MONTANARO: Sure. And look, there was a new round of New York Times/Siena polls out last night. They were eye-popping. They showed Democrats ahead by 11 in Michigan, seven in Texas, six in Maine, five in Kansas and Ohio. A couple caveats, though - this does not mean that that's what's going to happen. First, not all of these show Democrats at 50% in these polls, so there's a good number of undecideds. These polls also found that people who made up their minds in the last month sided with Republicans by nine points. And we should say that Michigan poll was before the feud between Democratic candidate Abdul El-Sayed and his primary opponent, Congresswoman Haley Stevens, who endorsed the Republican in the race, Mike Rogers, on Sunday.
CHANG: Yeah, drama. OK. What about prediction markets and all of this? Because they're getting a ton of attention right now. What do you make of them? Is there actually valuable information there?
MONTANARO: I mean, these are not polls, right? And we should say that again. These are not scientifically conducted, rigorous surveys. These are reflecting who people are putting their money on. And guess what drives their decisions on that? Polls.
CHANG: (Laughter).
MONTANARO: So even...
CHANG: There we are again.
MONTANARO: ...If they...
CHANG: Yeah.
MONTANARO: ...Wind up reflecting the results, they don't have some superpower. Someone showed me last night that one of the sites shows Democrat James Talarico ahead of Republican Ken Paxton in the Texas Senate race here by 18 points.
CHANG: Whoa.
MONTANARO: That's not a 59 to 41 lead, like you would see in polls. That's just, again, who people are putting their money on to win. And if you look at it like stock, right now, based on the polls, Talarico stock is high, and you can see people playing games and shorting the race and saying, well, maybe it's a good time to buy Paxton. So, look, these are not scientific, right? And they don't matter all that much to, like, what we do in our job. NPR journalists are not allowed to place bets in prediction markets on developments of news or events or anything else we might cover.
CHANG: All right. So it sounds like, everybody, just take the polls with a grain of salt. Do not rely on prediction markets to tell you what's going to happen. And that is the advice of NPR's Domenico Montanaro. Thank you so much again, Domenico.
MONTANARO: You got it. Transcript provided by NPR, Copyright NPR.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.
