A forecast of the 2024 US presidential election -- are the assumptions and results reasonable?

After January 6, my father watched the Arnold Schwarzegnegger response I sent him and wrote me back, “I’m never voting for Trump again.”

He and Mom are two-time voters for him.

I told them to feel free to put a vote on an actual Christian candidate out there. Evan McMullin served this role in 2016 and that is where my brother-in-law put his vote.

Not sure they would be comfortable with McMullin. I know they voted Romney rather than Obama but being Southern Baptists they have serious issues with the LDS church.

RFK is wackadoodle but a vote for him in TX is a vote not for DJT.

Yeah, I meant to point that out. The fact is, they were OK with Trump, so a Mormon seems downright completely acceptable. I think McMullin is like an actual American, too. Military, didn’t betray his nation, the whole thing.

Realize my parents voted Mitt Romney, another Mormon. And another real American who actually, you know, would have just done the job as president like anyone else. And also called out Russia as the top threat to our nation, also correctly.

Edit: My parents are like many, though. Pette Buttgieg is also a strong Christian with military background who should by all means be an obvious choice as a possible president, But since he is gay, they’d never support him. Trump? Sure. Buttgieg? Nope.

Big picture, vote percentages for the Presidency depend upon incumbency and economic performance during an election year. Using forecasts of the latter, we can apply Ray Fair’s model, first designed in 1978 and subsequently modified once or twice.

Assuming a growth rate in per capita GDP of 1.8% during the first 3 quarters of 2024, the Dems should earn 51.46% of the 2 party popular vote. During the past 2 elections, Trump has underperformed the model substantially. If that pattern holds, the Dems should get more than that.

https://fairmodel.econ.yale.edu/vote2020/indeane2.htm

Do things other than the economy and incumbency matter? Sure they do. They make up the error term. The average error is about 2.5 percentage points (RMSE). And that’s based on actual economic growth rates, not predicted ones. So the signal at this juncture is weak but happens to favor Biden.

This is the most rational reason I have seen for Democratic optimism. AFAIK all the econometric models have Biden ahead, at least in the popular vote.

However, as a Democratic pessimist, I note that these models usually give a boost for incumbency. And of course they consider there to be only one incumbent, Biden. But in some psychological sense, Trump and Biden are perhaps both incumbents. Due to lack of precedents, I think this is impossible to answer.

An argument against the last paragraph is that Hillary Clinton could be seen as a psychological incumbent because Bill would have returned to the White House. I thought she would win because of swing voters remembering the good economy during her husband’s presidency and thinking he would do his magic again. Obviously I was wrong.

I think polling averages give the best inkling of what voters are thinking, and even far out are slightly predictive. But econometric models mean more to me than betting markets.

Polls are useless until about June. By the fall, polling models outperform fundamentally-based models. At least according to Nate Silver.

Biden could win the popular vote and lose the electoral vote. So that needs to be factored in as well. Given that we’re within 1 standard deviation, the fundamentals put Biden at a pretty narrow lead. But Trump underformed by a whopping 3.4 and 5.4 points during the past 2 elections. So borrow 2 of those points, add it to the current 1.5 points and we have a more substantial 3.5 point lead. It’s still within the 95% confidence interval, but it exceeds the 2.5 point standard deviation.

I’m cautiously optimistic, but we still have a fight on our hands.


I have a fairly standard message for conservatives: Thatcher’s Britain was more liberal than Carter’s America. Remove the most liberal 30 members of Congress -either House- and you have a Democratic Party that would be center-right in any other mature democracy in the world. So I wouldn’t worry too much a socialist takeover.

Things will be fine as long as everyone plays by the rules and respects the Lord’s creation. As a liberal, they won’t necessarily be the way I’d prefer – I think Wall Street portfolio managers should pay the same taxes as anyone else with their income – but it will be fine.

withdrawn

True.

I would modify that slightly - things have changed recently.

Economic performance does not matter, in the sense of the actual performance as measured by normal indicators.

What matters is economic performance as stated by who can lie the most egregiously about how terrible the economy is - despite the factual evidence. If a party can convince a substantial number of the public that white is black, sunshine is rain, and a good economy is actually bad, then that is what will win an election.

That’s the elephant in the room. The rules don’t matter anymore, and if Trump gets in, the rules will be nonexistent.

Everyone is not playing by the rules. Things have the potential to not be fine at all.

There’s no evidence for this. Except, the last 2 cycles have had large modeling errors in the Fair model. Large enough to affect the residuals in previous elections. So you might be correct. We’ll have a clearer take in another decade.

My point: yeah it seems that way. But there’s just no getting around a core of uncertainty, when big elections come every 2 or 4 years, as opposed to eg every month or quarter for the general economy.

Thanks for posting this — right up my alley.

Trump is polling at high rates among Latino voters, rates that haven’t been seen by a Republican candidate in 20 years.

The problem (as you surely know) is that, this time around, many voters are saying “Well, MY financial situation is fine – but the ECONOMY AS A WHOLE is terrible!,” because that’s what Fox News tells them to think.

ETA: Euphonious_Polemic explained this well, a few posts back. I suggest that this has become more of an issue lately – that, in previous elections, economic indicators were a better (if still imperfect) predictor.

Nate Silver/538’s past models typically used the economics for about 20% of their weight (IIRC), aggregate polling trends being the bulk of the rest. Sometimes they would run several models simultaneously, and one would weight economics more than the others.

If I may be indulged to be a bit unserious to make a point …

The recent hand-wringing over Latino voters reminds me of a favorite Star Trek quote. From the original series episode “The Arena”:

Mr. Spock: A sustained warp 7 speed will be dangerous, Captain.
Captain James T. Kirk: Thank you, Mr. Spock. I mean to catch them.
Scott: We’ll either catch them or we’ll blow up, Captain. They may be faster than we are.
Captain James T. Kirk: They’ll have to prove it.

I’d like to think people who are straight-up told what to think by Fox News are largely unreachable to the Biden campaign, anyway.

For the rest of the country, there is hope that the “everyday life” economic indicators (e.g. gas prices, a plateauing of grocery prices) will improve enough for much of the electorate to shrug off 2023’s economic concerns. Stations convert from summer-blend gasoline to winter-blend by mid-September with price decreases typically following soon after. Sure, nothing at all to do with any Biden policy – but fortuitously-timed pump relief will make some (many?) voters feel better during mail-in/early-voting season.

Small change, but worth noting: Since you posted the item quoted, Statista has re-run the same battleground-state polling. Their March polling data is already a bit dated at three weeks old, but it still shows that Biden pulled even in Pennsylvania and Michigan and to within the margin of error in Wisconsin (-1%).

Wisconsin is leaning Democrat. In statewide elections the Democrat has the edge. Ron Johnson at 50.4% of vote for senator is barely holding on. As Republican. By 2028 he will be history.

I’m cautiously optimistic. Seeing as this turns on the (IMO braindead) middle, I fear current polls are of very limited value. It does reassure me to see things seeming to trend towards Biden/sanity. Trying to avoid thinking too poorly of my fellow citizens, I hope there are enough non-insane folk in the middle who will respond to the constant news of Trump’s legal issues to decide maybe he isn’t the sort of person they wish to have the most powerful job in the world. Combined with abotion. And either vote Dem or stay home.

Meanwhile, I’m hoping the Dems get charged up enough to show up in force. So I’m cautiously optimistic for stronger Dem turnout than Repub.

This thread feels like a patient trying to calm himself with the fact that the doctor told him there was only a 40% chance he has cancer. Whether it’s 25, 40 or 45 percent, it should be frightening to contemplate.

That’s a great analogy!

Me,l I have a gut feeling that in November, the patient is going to get a real shock when the final results of the lab tests come, in November…