Democrats have spent months telling anyone who will listen that the midterms are theirs to lose. With President Donald Trump's approval rating in the basement and conditions supposedly ripe for a blue wave, the party is feeling confident it can win both the House and the Senate in November. Polls do seem to suggest they have momentum, but a closer look at the numbers tells a very different story, and the wave that Democrats are banking on may not be coming after all.

The latest NBC News poll does show Democrats with a lead in the national congressional race. On the surface, that sounds like good news for the party. But CNN chief data analyst Harry Enten put that lead in context, and the context is the problem.

"Democrats have an edge, but compared to what we see historically, it's really not that big of a lead," Enten said.

The NBC poll puts Democrats up by five points in the national congressional vote. Five points may sound meaningful until you stack them against what those numbers looked like in actual wave years. In 2018, when Democrats flipped the House, they led by ten points at this stage. In 2006, another Democrat wave year, the lead sat at 11 points.

"It's this type of lead where you say, you know what, Democrats are ahead, but don't count your chickens just quite yet," Enten said.

And this is not a one-poll problem. Enten pointed to a pattern emerging across multiple pollsters, including NBC, Marquette University Law School, and Ipsos. All three tracked the Democrat lead from January and February into May and June, and none of them shows momentum building. NBC dropped a point. Marquette University Law School dropped three points. Ipsos held roughly flat.

"There is this group of pollsters that are out there that are just not showing the wave you might expect, given where the president's approval rating is," Enten said. "The Democrats hold these leads. But again, you have this group of pollsters that are out there where this lead, simply put, is not matching what you might expect if you were expecting a Democratic wave."

The redistricting wars did not help Democrats at all. Republicans are locked in structural advantages through the redistricting process, which means a five-point national lead may not translate into actually flipping the House.

"Democrats need between a three and four point advantage in the national vote," he said. "You average those polls together. It's right on the border. It is no guarantee. It is far from a guarantee at this point if you believe these pollsters."

That is not exactly the kind of language that inspires confidence in a Democrat war room. And then there are the prediction markets. The Kalshi prediction market gives Democrats a 78% chance of winning the House. That sounds strong, but it also means Republicans have a better-than-one-in-five shot at holding on. In the Senate, the GOP still holds the edge.

"Democrats probably take back the House, but it's far from a guarantee," Enten said. "We are not going to get caught, like in past years. We ignore those polls that had those warning signs."

The Dems midterm warning sign is pretty clear:

There are high quality polls (Ipsos, MU Law, & NBC), which show Dems shy of their generic ballot benchmark in past wave years.

Their Dems avg. lead is just 3 pts... & Dems need a 3-4 pt win to take the House given redistricting. pic.twitter.com/qPy5CsxFZS

— (((Harry Enten))) (@ForecasterEnten) June 15, 2026

Political analysts burned themselves in past cycles by dismissing data that did not fit the expected narrative. The warning signs were there, and they got ignored. Enten is clearly trying not to repeat that mistake, and the data he is looking at right now says the same thing it has been saying for months: this is not shaping up to be the blowout Democrats want... or need.