Attempting to build a model of what the 2024 presidential election will look like based on current data

 So, I know that it's very early. Way way WAY too early, but I wanted to try to build a model for what the 2024 election would look like with current polling data vs 2020. Now, I do want to say that what I come up with should be taken with a grain of salt. I expected, based on where we were in 2021 for the 2022 mid terms to be a complete and utter disaster for the democrats. But then roe v wade was overturned and the numbers went in a much different direction. I would probably say the best time to start SERIOUS projections would be June 2024 or so. By then we will know who the nominees probably will be with some level of certainty, and states do more in depth polling on the state level. 

Right now? There are too many variables to know for sure, and so much can change over the next year or so. Still, I know a lot of people are already doing doomer projections and arguing that not much will likely change, so I wanted to give my input of what the 2024 election will look like with current polling.

Republican & Democratic primaries

Right now, and I believe this to be the case barring health issues or legal troubles, but Biden and Trump are both heavily favored to be their respective parties nominees for president. It's not even competitive. 

Trump has 57% of the vote in his party, with the next nominee, Ron Desantis and having 14.5%. From there, you got Haley and Ramaswamy at around 5-6%, and it just gets worse from there.

I just don't see the republican base shifting away from Trump, and it's gonna take a heart attack (health issues) or literally being thrown in prison to stop that. I'm not saying this can't happen because Trump is, in fact, old, and he is, in fact, dealing with legal issues, but I wouldn't count on it from a normal polling perspective.

The same is true of the democrats and Biden. Biden is at 65%, with Kennedy at 14.9% (why?!) and Williamson at 4.9%. Once again, barring some sort of medical crisis, which could happen at Biden's age, I just DON'T see him being replaced.And if he was, we would probably get vice president Harris being the obvious shoe in whether we like it or not. So let's just assume a Trump vs Biden matchup. I mean, at 40-50% ahead in both camps, we're talking Z scores in the 10+ range, which means virtually no probability for error here. Not saying it can't happen, but it would take either a nation wide shift in polling, or some sort of issue just flat out stopping those two from running. Given polls are snapshots in time, well, these things CAN change, which is why I've been reluctant to do a forecast like this, but assuming that they don't yeah. 

The general election

Right now, Trump has a 1.5% lead in national polling. I crap you not. Trump is actually ahead. And that ain't good because, well, remember discussions from last time about how the dems need a 3%+ lead just to maintain a roughly 50% shot at the presidency? Yeah...and they're 4.5 points underwater...

I mean, remember how razor thin the margins were with Biden having a 7.2% lead in the polls last time (which translated to a 4.5% lead in practice?). What about in 2016 about how Hillary lost with a 3.2% lead in polling and a 2.1% lead in the popular vote? So yeah, the dems are significantly underwater right now. 

If we were to apply my 2020 electoral college to popular vote model to current 2024 polling, we would get roughly a 85-86% chance of a Trump win, or a 14-15% chance of a Biden win. 

Or how about this? Biden won by 4.5% in the popular vote last time, and he's now down 1.5% to Trump. So say we took 2020's election results, and shifted the electoral college by 6 points. This is basically what the prediction would look like:

State	Election result 2020	2024 estimate	Z score	% Biden Win	% Trump Win	EV if Biden wins	EV if Trump wins
Iowa	-8.2%	-14.2%	-3.55	<0.1%	99.9%<	413	131
Ohio	-8.0%	-14.0%	-3.5	<0.1%	99.9%<	407	148
ME-2	-7.4%	-13.4%	-3.35	<0.1%	99.9%<	390	149
Texas	-5.6%	-11.6%	-2.9	<0.1%	99.9%<	389	189
Florida	-3.4%	-9.4%	-2.35	0.1%	99.9%	349	219
North Carolina	-1.4%	-7.4%	-1.85	3.2%	96.8%	319	235
Georgia	0.2%	-5.8%	-1.45	7.4%	92.6%	303	251
Arizona	0.3%	-5.7%	-1.43	7.7%	92.3%	287	262
Wisconsin	0.6%	-5.4%	-1.35	8.9%	91.1%	276	272
Pennsylvania	1.2%	-4.8%	-1.2	11.5%	88.5%	266	291
Nevada	2.4%	-3.6%	-0.9	18.4%	81.6%	247	297
Michigan	2.8%	-3.2%	-0.8	21.2%	78.8%	241	312
NE-2	6.5%	0.5%	0.13	55.2%	44.8%	226	313
Minnesota	7.1%	1.1%	0.28	61.0%	39.0%	225	323
New Hampshire	7.4%	1.4%	0.35	63.7%	36.3%	215	327
Maine	9.1%	3.1%	0.78	78.2%	21.8%	211	328
Virginia	10.1%	4.1%	1.03	84.8%	15.2%	209	341
New Mexico	10.8%	4.8%	1.2	88.5%	11.5%	196	346
Colorado	13.5%	7.5%	1.88	97.0%	3.0%	191	356
New Jersey	15.9%	8.9%	2.23	98.7%	1.3%	181	370

Now, there's a lot to unpack here. For reference, I shifted the entire 2020 election over to Trump by 6 points, accounting for shifts in polling. This caused 6 states to defect to Trump: Georgia, Arizona, Wisconsin, Pennsylvania, Nevada, and Michigan. All the usual suspects that seem to define who wins a modern presidential election, with 5 of those (minus nevada) being the ones that swung from Trump in 2016 to Biden in 2020. A 6 point shift toward Trump would represent a shift back to 2016. Thankfully, after Michigan, Biden kind of has a nice plateau to act as a buffer. Still, this would quickly being NE2, Minnesota, and New Hampshire into swing state distance, with Maine, Virginia, New Mexico, and Colorado, all comfortable non swing states becoming swing states. On the flip side, Iowa, Ohio, ME2, Texas and Florida all become so out of reach for democrats that they stand virtually no chance of winning there. And you can see how that translates into probabilities. This is why I rarely look at states outside of the margin of error, you very quickly end up with scenarios where one candidate has a 99.9%+ chance of winning and yeah it's not even worth measuring.

But all in all? Biden, based on this exercise, currently has a 8.9% chance of winning, with Trump having a 91.1% chance. This is literally worse than Trump's 2020 chances vs Biden. The electoral margin is expected to be 312 to 226 in Trump's favor, shifting from the 303-235 the post 2024 electoral map would yield for the same results as 2020, and resemble Trump's 2016 electoral victory. 

So yeah, if things continue like this, democrats are SCREWED.

Of course, I expected the dems to be screwed in the 2022 mid terms this time in 2021, but things changed. And things CAN change. Going back to 2020 actual numbers vs 2024's polling, there are a lot more undecided voters. Biden won 51-47 to Trump in 2020. Right now, however, we're talking 46-44. And I would say Biden stands a better chance than is portrayed by this logic: the people who are going to vote trump have already made up their mind, people are going ride or die on him. They dont care about his criminal charges, they dont care about January 6th, they are die hard BELIEVERS. Meanwhile what happens on the dem side? Well, a lot of people dont actually LIKE biden. Myself included. I admit I have warmed up to him as I see him doing some progressive stuff, but I'm not passionate for the guy. I mean, I have entertained going for Marianne Williamson and Cornel west over him.

But...at the end of the day, the dems know most will vote for him anyway whether they like it or not, and you're right, we will. Even I will probably be voting for Biden this time whereas I didn't in 2016 and 2020. And as long as the dems dont really flaunt this fact and shove it in peoples' faces without offering anything, there doesnt stand to be a reason why most swing voters won't just...do it again.

I mean, do they really want trump? When you're not in the cult, Biden might actually...appeal to you. He's a relatively inoffensive centrist to normie swing voters and while his record isn't amazing and people have obvious questions over stuff like the economy, inflation, etc, I mean, it's not really his fault.

So idk, it really depends what's worse, peoples' grievances against Biden, or people being so horrified the orange nutcase, who should be behind bars btw in my opinion, is running again and actually has a chance. I honestly think the fear of another trump presidency will motivate swing voters who currently can't commit to Biden currently more than whatever Biden's flaws are. I dont think this is 2016 again. Trump is more dangerous this time, we know what we're getting, and Biden has made some of the compromises needed to win over voters he didn't necessarily win in 2016.

As such, I could see the 2024 election follow 2020 pretty closely, just as the 2022 mid terms did. 

Still, based on the existing polling data and where we are now, if the election were held today, Trump wins overwhelmingly. It's actually scary how strong his numbers are. Not even Clinton polled this bad last time around. If anything throughout 2015 she was the overwhelming favorite nationally and would crush republicans. It only evened up as we got into 2016. It's possible that will happen here, where the poll numbers will spike for Biden and lag for Trump. Perhaps trump is the clinton of 2024. Kind of funny to see.

Anyway, that's why I don't normally do forecasts like this that early. It's a lot of effort (I literally just spent 2 hours making this) and honestly, the whole thing could be quickly invalidated. But I did want to explain exactly where the current polling stands. Biden seems as screwed as Trump did in 2016 when I started doing forecasts, although given how that turned out, again, that could easily happen here. We'll just have to wait and see.

Anyway, expect me to have more accurate election predictions come mid 2024.