UPDATE (10:15 PM, 11/4): Zohran Mamdani has been declared the winner of the New York City mayoral race by basically every reliable news outlet. His margin will probably come in below the polling average — so once again, AtlasIntel gets to look smart. But it projects to be a solid+ victory.
Cuomo and his backers can’t really blame Curtis Sliwa, as Sliwa considerably underperformed his polls, with voters gravitating toward the two frontrunners. But it’s not as though that will prevent them from trying. They’re still deferring responsibility from the role they played in what will be remembered as a legendarily terrible campaign:
I’ve tweaked the subheadline but left the rest of the text unchanged from this article as originally published early Tuesday afternoon. Eli and I will see you at noon tomorrow for our Substack Live with Ross Barkan.
We’ve had a lot of content lately. But it’s Election Day here in New York, which means I figure you won’t mind getting another email from me. I wanted to share some data I collected, which helps explain why Zohran Mamdani will probably beat Andrew Cuomo tonight.
I say “probably” because even when we don’t have an official model of the race, we’re sensitive about the distinction between a remote chance and zero chance. Cuomo’s chances aren’t zero; his backers can hang just a little hope on a pair of AtlasIntel polls that show a much tighter race than others, with Cuomo closing to within 5 points in their final survey, along with a history of late shifts in NYC mayoral polls.
AtlasIntel is, in fact, our top-rated pollster, and I always appreciate a polling firm that puts its neck out there instead of herding toward the consensus. At the same time, the consensus — which shows a Mamdani lead in the low to mid teens — is correct more often than not. It was almost exactly a year ago when I found myself writing about an Ann Selzer poll that improbably showed Kamala Harris winning Iowa, only for her to lose the state by 13 points.
The other thing that stood out in the AtlasIntel poll is this: their chart of positive and negative ratings for a variety of local and national politicians.
New York City voters really dislike everybody, except for Barack Obama, AOC and — improbably because he has a negative rating in most other polls — the Republican nominee Curtis Sliwa. The past two mayors, Eric Adams and Bill de Blasio, are deeply unpopular in this survey. So are congressional leaders Chuck Schumer and Hakeem Jeffries, Governor Kathy Hochul, and to a lesser extent, Joe Biden and Kamala Harris. Any and all Republican figures apart from Sliwa are also unpopular in New York, of course, which is why Mamdani probably doesn’t mind President Trump’s endorsement of Cuomo.
Still, while Mamdani’s ratings are net-negative in the AtlasIntel poll whereas they’re positive in most others, take a look at Cuomo’s numbers. Only 33 percent of voters have a positive impression of him, while 58 percent have a negative one.
It’s hard to win an election on that basis, especially a three-way race where you aren’t the only option. Remember, Cuomo isn’t the only alternative to Zohran; in addition to Sliwa, incumbent mayor Eric Adams (who has discontinued his campaign) is listed on the ballot along with three other candidates. Voters also have the option of a write-in candidate, of undervoting the mayoral race (there are also ballot propositions and city council races on the ballot), or just not bothering to vote at all.
And nearly all evidence suggests that Mamdani’s voters are more enthusiastic than Cuomo’s. I don’t want to dive into crosstabs or turnout projections, but if the AtlasIntel poll is off, I suspect it will be because it’s describing the result you might get if everyone were forced to vote. But you can have reservations about Mamdani without wanting to cast an affirmative vote for Cuomo, and if you do, you might just stay home.
Here’s that data I collected: what Mamdani’s favorable/positive ratings look like in every survey since the primary:
Not really much going on here. In fact, Mamdani’s numbers had been slightly improving over the course of the general election until recently, and most of the late movement is because of the pair of AtlasIntel polls. Perhaps some of Cuomo’s relentlessly negative messaging is sinking in. There have been some new developments, like Mamdani’s comparison of the NYPD to the IDF. But it’s come at the cost of making Cuomo even more unpopular:
Cuomo simply doesn’t have any foundation of people voting for him for positive reasons. In last week’s Quinnipiac survey, for instance, which actually showed a slightly closer race than most, the only demographic group to give Cuomo a favorable rating above 50 percent was voters aged 65 and older, and then just barely at 51 percent (against 38 percent unfavorable).
Republicans don’t really have much basis to be comfortable with Cuomo; he governed as a fairly moderate but nevertheless partisan Democrat in his three terms (one truncated) as governor. But Cuomo also lost the Democratic nomination to Mamdani and is attacking Zohran from the right. Independents, meanwhile, have no particular reason to like someone who is such an avatar of the political establishment. And every voter can object to the extensive list of sexual harassment accusations against Cuomo — although for whatever reason, a lot of polls show him doing better among women than among men.
Meanwhile, if you go to Cuomo’s website, the lead message is “Say No to Zo”:
Click on that, and you’ll see a lot of stuff like this:
I do think Mamdani has more vulnerabilities on substance than might be assumed — even relative to even the New York City electorate, which is liberal but not that liberal. He’s tried to walk back his previous support for defunding the police, which might be just about the most unpopular political position in the country. New Yorkers are very reluctant to make any changes to gifted-and-talented programs. Cuomo hasn’t really emphasized trans rights, but I’m not sure how Mamdani’s pledge of $65 million for gender transition surgeries (including for minors) would poll even with the NYC electorate. Even Mamdani proposals like free bus service that are popular on the surface can lose support depending on the question wording.
But the messaging from Cuomo is dripping with contempt — like “failed rapper”, “rich mom”, or a bizarre AI-produced video in which Fake Zohran slurredly sings “I’m just a shill” in what I’m not even sure you can call a Schoolhouse Rock! parody because parodies are supposed to be funny.
Meanwhile, Cuomo backers like Bill Ackman have accused Sliwa without evidence of just being in the race for the money. (Sliwa, also without evidence, has accused Cuomo backers of trying to bribe him out of the race.) This projects an incredible sense of entitlement for Cuomo, a candidate who had the chance to beat Mamdani fair and square in the Democratic primary, but actually lost further ground once ranked-choice ballots were tallied because Cuomo was so polarizing.
Meanwhile, Mamdani has run an affable, positive campaign, bouncing around between Knicks games and gay clubs, always in his trademark suit. He’s a talented candidate, and you can see why it isn’t an easy task for Cuomo to convince voters that Mamdani is more radical than his recent branding makes him seem.
Still, this has been a hell of a long race: Cuomo declared for the primary way back in March. Three-term governors ought to be able to walk and chew gum at the same time, i.e., soften their own negatives while attacking their opponents. However, among other problems, Cuomo has never convincingly persuaded voters that he cares about New York City instead of being an Upstate Guy. He also hasn’t shown much of an independent streak at a time when the Democratic establishment is profoundly unpopular, instead trying to convince voters that he’s the real Democrat in the race. (You can imagine a different Cuomo campaign where he ran against an establishment that overwhelmingly called for his resignation.) And he ought to be able to do all of this with a little more grace toward Mamdani instead of making it seem so personal.
If Cuomo pulls off the Hail Mary, we’ll have to ask what every pollster except AtlasIntel was missing (and even AtlasIntel does have Zohran ahead). At this point, such an outcome would be every bit as shocking in its own way as Mamdani’s rising from obscurity to win the primary. But I wanted to get this out there before tonight because I hate Monday morning quarterbacking and this certainly doesn’t look like a winning formula.
Last Election Day things:
We’ll have a Substack Live on the results tomorrow at noon with Ross Barkan.
Eli is going to boot up a subscriber chat at 6:30 tonight. Since I haven’t been following races apart from NYC all that closely, it’s mostly going to be him and you guys.
Finally, thanks to those of you who came out to see me, Galen Druke, Clare Malone — plus a guest appearance by Eli! — last night at the Comedy Cellar. For paid subscribers, there’s a video of the event down below.