US midterm elections: A ‘blue wave’ may be building, but how high?

As my colleague Carl Hulse recently pointed out, there are a lot of similarities between Trump’s predicament today and Bush’s back in 2006. Bush was in a quagmire in Iraq and dogged by high gas prices. Today, Trump has just a 31% approval rating on the war in Iran, and a 28% rating on the cost of living.

Whether Democrats can translate these enormous national advantages into control of the Senate and a large House majority is a more complicated question. To win the Senate, Democrats will need to win at least two seats where Trump won by at least 11 points in the last election. That’s really tough, even in a wave year.

Here again, 2006 offers a useful comparison: to flip the Senate, Democrats needed to win at least a few seats in states where Bush had won by at least 8 points in the previous election. In the end, they pulled it off – but only barely, including two extremely narrow victories in Montana and Virginia. Without a few fortunate breaks in those states, those races could have gone the other way – along with the Senate.

The national environment may have been great for Democrats, but a Senate win was far from inevitable.

Of course, the Times/Siena poll is just one survey, and polls are imprecise and subject to many sources of potential error. But if you’re the kind of person who follows every latest poll, you know this is far from the only data point showing a major Democratic advantage.

The Times/Siena results are almost exactly in line with the average of other polls. And it’s not just the polls. There are the results of recent special elections, the Democratic turnout advantage in the primaries, or simply the long history of midterm elections going back to the dawn of party politics.

With Trump’s approval ratings so far underwater, Republicans may have only one realistic card to play: try to rally their base and hold on to Republican-leaning independents by casting the Democrats as too far to the left.

Republicans didn’t have a similar opportunity in 2006, when Democrats assiduously nominated moderate candidates in key red states and districts. While Democrats have nominated broadly acceptable candidates in most states this year, there are at least a few competitive races – most notably the Senate contest in Michigan – where Democrats have departed from their usual playbook, nominating candidates far to the left of anything the party would have contemplated 20 years ago. They’ve also nominated left-wing candidates in a number of non-competitive races, which has driven considerable national media coverage of the Democratic Party’s lurch to the left.

The poll does contain hints that primary victories this summer by progressive and democratic socialist candidates have made the Democratic Party less palatable to some voters. About half of voters – 50% of likely voters – say the party is “too far” to the left. That’s up from 46% in the last Times/Siena poll of registered voters in May, which was conducted before the democratic socialist surge in New York, Colorado and beyond.

But it’s hard to see any sign it’s damaging Democrats in the survey results – and in particular, in any question that pits Democrats against Republicans. Voters say, for instance, that Republicans are more extreme than Democrats by an 8-point margin, 48-40, even though a smaller share of voters – 48% – say the Republicans are “too far” to the right.

Now, “extreme” evokes more than mere left-right ideology; someone who shouts views about an issue will be seen as more extreme than someone who has the same views but stays quiet. But at a time when Republicans hold the White House and Congress, voters will tend to be much more concerned about Republican excesses than Democratic ones (much as Democratic excesses around “woke” seemed so much more salient during the Biden era). That’s part of what makes midterm elections so hard for the party in power.

There’s only so much you can say about contest-by-contest results based on a national poll, but there are two additional points beyond the national numbers worth considering.

First is demographic. The poll shows, yet again, that Democrats are making huge gains among young and non-white voters, particularly Hispanic ones. This suggests that Democrats could be poised to make outsize gains in relatively young, non-white and particularly Hispanic areas – first and foremost Texas. Based on the demographic trends in this poll, Texas would be expected to be highly competitive – even if Republicans had not nominated someone who has been investigated, indicted and impeached.

Second, we did something unusual: we asked people whom they would support in the race for governor and Senate in their state using the names of the actual candidates on the ballot. Overall in key states, Democrats did nearly 3 points better on the named Senate ballots than they did on the generic congressional ballot in the same states.

This is not necessarily a surprise: it’s pretty similar to what we found in our June polls of the key Senate races in Republican-leaning states. But it does provide a rare clue from a national poll that the Democrats might be poised to fare better in the key Senate races than one might expect from the national figures alone.

This article originally appeared in The New York Times.

Written by: Nate Cohn

Photographs by: Kenny Holston and AFP

©2026 THE NEW YORK TIMES