Pollstradamus - US Election Forecasting

90 Days

Where the Race Stands with Harris' Rise

August 07, 2024

“The board is set, and the pieces are moving.”

Harris has been officially coronated as the nominee following a virtual vote, and she has her number two.

90 days remain until polls close, and everything is on the table: the Presidency, House, and Senate.

In a race with more twists and turns than a scoliosis patient, one should not feign about knowing how this race will truly all go down. Just over a month ago, Biden had ninety-percent odds to be the nominee; now Kamala Harris is. Three weeks ago, former President Trump survived the United States’ first serious assassination attempt in forty years. Josh Shapiro was the odds-on favorite to be Harris’ sidekick going into the weekend; now we know it’s Tim Walz.

All that is to say, whichever candidate you are supporting: Do not get complacent. Act where you can and encourage others to do likewise.

But here’s the state of the race as it stands today.

Following the post-Biden recalibration, states like Arizona and Georgia moved back into serious contention while states like Minnesota, Virginia, New Hampshire, and Maine shifted out of focus. Harris now boasts anywhere between a 3 and 7% lead in each of those latter four states.

It’s become clear now that the only states anyone should be caring about at all going forward are the original six battleground states we started with in the first place: the aforementioned Arizona and Georgia; fellow Sun Belt state Nevada; and the three Rust Belt states of Michigan, Pennsylvania, and Wisconsin.

Arguably the seventh battleground state, North Carolina, will be close, but no polling has suggested that Biden or even Harris can realistically win the state, even though the margin will be narrow. For it to stick with Trump in 2020 amidst all of the political headwinds means there is little chance of it turning its back on him now. Democrats will certainly fight hard for the governorship, and it is anyone’s guess who comes out on top there. The scattered polling has generally been more favorable towards Stein, but Robinson has kept it close and down ballot Republicans have tended to outperform their polling numbers when Trump is on the ballot (See Senators Thom Tillis and Ted Budd, neither of whom are supposed to have their jobs).

With that said, the Sun Belt states experienced a larger shift in Democrat support following the Harris rise than did the Rust Belt states. According to Nate Silver, each of the three battleground Sun Belts states shifted by an average of 3.5% towards the Democrat compared to the Rust Belt’s 1.9% shift. However, consider also that the Rust Belt shift includes a horrendous outlier poll from Morning Consult that showed Harris beating Trump in Michigan by 11%. That would top Barack Obama’s 2012 performance over Mitt Romney! They can believe that if they want to; Trump consistently outperforms their polls anyway…

What this means for the race is that Trump now needs to play more defense than before. My electoral model has Trump at a cool 72% chance of winning; the only problem with that right now is that I believe my averages are slightly overestimating his support in Arizona and Georgia. There’s been very limited polling out of those two states, and even less polls that publicly released their crosstab data (one of two requirements for a poll to be included here).

Trump is still favored to win in those two states, but having to push more resources to those states means that those resources are not being put in places where Trump could be on the offensive.1 That of course being the Rust Belt. Those three states have firmly shifted back into the “purple” column, with Michigan having a tinge of blue. Michigan is, afterall, the bluest of the three. With Kamala’s VP selection blunder, Trump’s long term prospects in Pennsylvania ticked up. While the Shapiro pick might have only accounted for a half a percent boost to her numbers, that amount proves to be a big deal in elections like these.2

The absence of Shapiro on the Harris ticket is especially important because the road to 1600 Pennsylvania Avenue goes through Pennsylvania itself. Whichever candidate wins that state wins the election 90% of the time. What’s truly baffling about the pick is that Pennsylvania is more important for Harris than it is for Trump.

You see, Trump can win without Pennsylvania, but Harris realistically cannot. Let’s say Trump loses Pennsylvania. Okay, well he can still win the election by taking Georgia, Arizona, and Wisconsin (272 EVs) ‒ all states he lost by less than a percentage point in 2020! Or, Nevada, Michigan, and Georgia (272 EVs); all states he held sizable leads in throughout the spring. Or, any combination of three of the following: Georgia, Arizona, Wisconsin, and Michigan.

Now let’s flip it. Trump wins Pennsylvania. Kamala Harris could win Arizona, Michigan, Wisconsin, Nevada, and Nebraska’s 2nd Congressional District and still come up short 268 to 270 if Trump wins Pennsylvania and Georgia. The math is so bad for Harris in this scenario that a loss in Pennsylvania means that Harris must win in both Michigan and Georgia as well as two of Arizona, Wisconsin, and Nevada in order to achieve victory. Which really begs the question why Harris did not pick Shapiro. If some of the reporting is to be believed, it is because the Walz pick was a “chemistry” pick. Not buying it. Even commentators on CNN are coming out to say the obvious: Shapiro was not picked because that would have really upset the far left anti-Semites in their party. But for that reason, picking someone other than Shapiro may have been an electorally sound pick in its own way.

If in two weeks the DNC and the surrounding Chicago metro had erupted into protests, demonstrations, and violence over a pro-Israel Jewish man being selected to the Democratic ticket, it would have caused a national meltdown that the media wouldn’t be able to spin. As we saw with the debate, their ability to spin reality only goes so far. A full-scale display of that sort would be undeniable and put into the public mind the idea that the Democratic Party and its ticket have become mired in political extremism and anti-Semitism. And the public would vote accordingly. An event like that would stick with voters and would prove to be something that the Harris campaign simply couldn't shake off. Now, with a pick that is uncontroversial to its radical base,3 the Harris camp can move forward into the DNC with optimism, thinking that it has successfully appeased the radical excesses of its party. That remains to be seen.

But that may be the only thing in between now and the end of the month that can derail the Harris honeymoon, dubbed the “Kamala-nomenon” by her supporters online. Harris’ bump in the polls following her ascension to the top of the ticket eighteen days ago seems to be holding steady at this point. Thanks in no small part to two and half weeks of glowing coverage from the mainstream media, from whom she has yet to answer a single question. Not one! In eighteen days. And she’s running to be the president.

It remains to be seen when such an event will take place. If Harris pursues a full-scale Joe Biden 2020 basement strategy, Trump could be in trouble. Afterall, it does not seem entirely implausible that the media would allow such a strategy to be undertaken if it meant their preferred political candidate wins out. This is all the more reason why Trump should end up debating Kamala Harris, even if it is not on home turf. Harris should debate Trump on Fox, but that does not seem likely to happen at this point. The incentives are on Trump, not Harris at this point. If he doesn’t, the media will play cover for her and make him out to be the one dodging the debate, even if that is only half true.

So, who do I think is going to win? My brain says Trump. Despite the cover that the media can and will run for Harris, that jig can only last for so long. People will notice, Republicans will campaign on it, and the rumblings will reach the broader public. If Kamala is forced to do routine interviews and press conferences, she will doubtlessly offer up some more word salad fodder. Even still, Republicans have a treasure trove of clips from her from over the past eight years of her national political life to pull from. Anything from her advocacy of far left policies to her incoherent ramblings to her general weirdness. While many Americans will indeed cast their ballots for Harris, I believe the sensible middle will side with Trump in a narrow but decisive election.

My gut is less sure. The board is indeed set, but there are some powerful pieces on the other side. The most dangerous one ‒ frankly, for both sides ‒ is the upcoming sentencing hearing for New York’s absurd criminal case against Donald Trump. Jail and house arrest are on the table. That date, September 18th, is also very close to the election, which means it has the potential to be much more impactful on the election than his May 30th conviction. Still, the potential blowback of forcing Trump into jail or into house arrest could be a case of cutting off one’s nose to spite one’s face.

A lot remains to be seen. But think back to four years ago, even eight years ago. Trump, like today, was down in the polls. But do you remember how badly he was down in those elections? Trump is now running neck-and-neck with the Democrat when in past elections he was always playing catchup, down by two or three scores going into the fourth. He never led polls in those elections. He’s led most of them this election cycle and still leads some today. Trump, like today, was hated by the media. But has Trump ever received as much open public support as he has this time around? Football players, rappers, tech moguls, influencers, business leaders, and other big name cultural icons now openly support Trump and don the Red Hat.4 Trump, like today, was losing with minority voters. But they were never even up for discussion before. Now, Democrats legitimately fear losing their Hispanic majority, and their eroding support among Black (male) voters also concerns them. States like Minnesota, Virginia, Maine, and New Mexico were never previously even a part of the conversation. Much less, New York.5

So the dynamic has certainly shifted, in a macro sense, in Trump’s favor; now we wait to see if can capitalize on it.


  1. A bit weird to consider how Trump is considered to be playing “defense” in two states he lost in the last go round.
  2. Georgia and Arizona were decided by less than 0.4% in 2020; Michigan by less than 0.3% in 2016.
  3. Its preferred candidate actually!
  4. No pun intended.
  5. No serious person believes he can flip it, but it is looking a lot closer than it has been in a long time!