[Above: No Kings Day Saturday in my old hometown of Greensburg, Pa., which made the Rachel Maddow Show on Monday.]
The preference polls for the ongoing election that ends November 3 are still consistently favoring Democrats. So naturally there have recently been a spate of articles questioning their accuracy. To be sure, most of these articles quote the same Republican pollster, who at first gave statistical answers but now is claiming that they are purposely misleading.
Of course there are reasons to doubt their accuracy. Polls have been in trouble since the Internet era began, because nobody knew how to reliably reach people anymore, when they no longer just grabbed the phone off the wall and answered it when it rang. Add to that the more recent trend of Republicans disproportionately refusing to answer pollster questions, or even just lie.
So polls using various mathematical operations to weight the polls, usually adding to Republican numbers. Which makes the polls inherently inaccurate, except I suppose when they are accidentally correct. But in which direction are they inaccurate? NBC's Chuck Todd thinks they may be underestimating the Democratic wave as easily as overestimating it. Others point out that the weighting of Texas responses in two big recent polls was highly favorable to Republicans.
We were badly burned by polls and the rhetoric around them in 2016 and 2024, and even in 2020 they didn't accurately reflect Republican strength. So of course we don't want to get fooled again.
But even if not entirely accurate, polls can be indicative. And they aren't the only indicators. There's reporting (on voter enthusiasm, for example), and there's follow the money. "If Democrats don't believe the polls," said Democratic operative Dan Pfeiffer, "clearly Republicans do. Follow the money." The Rs are spending money where they shouldn't have to.
Polls can be compared with polls from other midterm elections. The two key indicators usually cited are the President's approval--particularly the disapproval numbers, and especially the strongly disapprove-- and the generic ballot of Ds vs. Rs. Several of the most recent major polls put disapproval at well over 60%, and strong disapproval above 50%. Historically, those suggest a bloodbath. The generic ballot favors Ds by 8 to 12 points in these polls. The Gallup Poll question, "Which party would do a better job?" called the "trust question" has "an almost perfect track record" of predicting outcomes. The latest Gallup poll has Dems up by 11 points, 44% to 33%.
Then there are normal trends, the ups and downs in the polls. According to those, the Rs should be gaining in at least the early autumn. Instead their numbers are getting worse, and more and more races are moving towards the Ds. Historically, there is one move left--in the last week, and in recent years, the last weekend, before election day. There seems to be no reason for it to move towards the Rs, except perhaps in very deep red states where the trans ads can sway MAGA types. And there seem to be reasons for it to move the other way.
For example in South Carolina, where Chaos handpicked the Senate nominee, the sister of the late Senator Lindsay Graham. One of the few pieces of legislation Congress passed recently--just before they all left town, many heading for end-of-the-world expensive partying in Europe--was a package of sanctions against Russia in honor of L. Graham. Chaos, after apparently getting pieces of Russian fossil fuel companies for his family and cronies, is trying to stop sanctions, including those he just signed into law. The reaction to this in South Carolina has not yet shown up in polls.
There's plenty of publicity about how Republicans are far outspending Dems in campaign ads, particularly in newspapers and TV outlets where all that money is going. But how many ads get any attention when there are too many? How many voters are still persuadable? The polls haven't changed much in months, except to get worse for Rs.
Then there are the candidates. Especially in places where they expected no real opposition, Rs nominated lazy and corrupt candidates at best, and at worst criminal and psychopathic ones, though loyal to Boss Chaos. Ds not only more often nominated superior candidates but some who can change the party from within, and perhaps the country.
Then there is the electorate. The No Kings demonstrations over the past few years, plus hundreds if not thousands of other focused local protests, show a large galvanized number. The polls and reporting showed much greater enthusiasm for voting against Boss Chaos, which is already reflected in early voting numbers. Participation everywhere is higher than ever. This past Saturday No Kings day was focused on people actually voting that day, where possible. (And what a contrast with demonstrations in France the same day.)
Polls matched against past polls show more enthusiasm and more opposition to Chaos in categories like Latins, young people and Independents. James Carville predicts a huge increase in black turnout, especially in the South. Moreover, the attempts by Rs to stack the deck aren't working so well. Re-districting actually weakened many of their safe districts, so they have even more vulnerable House seats. Some states--notably Texas--tried to discourage turnout by severely cutting down the number of polling places in selected areas. In Greensburo, North Carolina, which closed polling places near college campuses, students from several colleges marched together to the nearest polling place, 45 minutes away, and cast their votes. In fact the rise in early voting by 40 to 60+ percent may be partly in response to talk of Boss Chaos trying to disrupt voting on election day.
The most obvious reason for believing that Ds are winning is the condition of the country, and the number of voters who are being really really hurt by Chaos actions and policies. (In a recent survey 82% of respondents said their personal finances have become worse in the past year.) There has never been an election in which the in-party is defending high inflation, soaring federal spending and debt, economic damage from high tariffs, alarm at blatant prejudice and cruelty, widespread revulsion in response to widespread corruption, and a very expensive war that nobody wants. Nor has there ever been a president more obviously demented, but not kept in check by alarmed officials--instead empowered by them.
Yes, we are only two years past an election where an apparent majority of voters drank the same old Kool Aid again. But Norman Mailer used to say something about some plastic stuff--that it looks perfectly okay until it suddenly shatters.
If the polls are underestimating the D Wave, or there is a last minute move towards Democratic candidates (and a marked tendency for R voters to stay home), then close races go to the D side. (Sarah Longwell looks at percentage numbers above 50%--those candidates usually win. A recent poll put James Talarico at 52% in Texas, with almost no undecideds.) In particular there are a number of states electing a governor and a Senator, with one or the other the polls say is more likely to win. A last minute movement could mean they both win. The US Senate is the most important, with the House next, but state houses and state legislatures can also be very important to a deeper and longer change.
