In most elections, there is a telling shift at the end that gives us an idea of where things are headed. That shift has happened in this year’s election. From a variety of sources, and from the actions of the candidates over the last few days, it now appears clear that there has been a decisive shift in favor of Kamala Harris.

However, for Harris supporters, I must inject a note of caution. So many votes have been cast early, either in-person or by mail, that there is a real possibility that the early vote could blunt the late shift in her direction that seems to be occurring.

As I have noted many times on this site and on my X.com feed, I tend to discount polling. However, some notable polling results in recent days, particularly in the Midwest, are all moving in Harris’s direction, including the shock poll by Ann Selzer in Iowa showing Harris leading Donald Trump by 3% in that deeply red state.

I do not really believe that Harris is going to win Iowa. But even if she comes close in Iowa—say she loses the state by 5% or less, a number that would represent Selzer’s biggest polling miss ever—that indicates a sweep of the Midwestern battlegrounds and a Harris win regardless of what happens elsewhere. We are also seeing record-setting early voting in Georgia.

I now believe that some notable number of traditional Republicans are crossing the aisle and voting for Harris in this election. I cannot guess at how many, but that could be decisive in states like Arizona and Nevada, where there has been a significant early vote by registered Republicans.

The bottom line is this: I now believe that Harris will sweep the seven battleground states, and I am no longer absolutely certain that Texas or Florida are out of reach, though I would still tend to favor Trump slightly in both. That would result in a 319-219 Electoral College victory for Harris.

As far as Congress is concerned, I expect Democrats to flip the House of Representatives with a net gain of 17 seats to arrive at a 231-204 majority. The Senate is a very hard call, but I am reversing my ratings change in Ohio from two weeks ago and am now predicting Democratic incumbent Sherrod Brown will hold his seat, even if Harris loses the state as expected. This cuts against the usual rule of presidential and Senate races moving in lockstep and would be only the second exception in the last three presidential election years. I also, once again, am not completely confident of Republican victories in either Texas or Florida. I tend to think the momentum of Democrat Colin Allred in Texas is real, and if this were a midterm, I would probably pick him to defeat incumbent Republican Senator Ted Cruz. But I just cannot imagine there will be a lot of Trump-Allred voters. As such, I still believe Republicans narrowly flip control of the Senate, picking up Democratic-held seats in West Virginia and Montana to win the chamber, 51-49.

My predictions here are conservative. If Ann Selzer is really right about Iowa—and she has been right most of the time—then something much bigger is happening here, and the presidential contests in Texas, Florida, Ohio and Iowa are all in play, as are the three Senate seats up in those four states. If Selzer is correct, that would also mean Democratic Senator Jon Tester may still have a chance to survive in Montana, and my prediction of Democrats +17 in the House would be low, possibly by several seats.

Whether there is really a wave, as a correct Selzer Iowa poll would indicate, it now seems likely to me that Kamala Harris will be elected the 47th president on Tuesday and that Democrats will win the House. The Senate may be in play, and a Democratic trifecta is no longer out of the question.

To view my final ratings of the presidential and Senate races in each state, click here.