Predictions for 2028
93.6% chance of Republican dwindling.
I’m going to present two models of how 2028 could go, and then explain which one is more likely. The two big variables are foreign policy and economics.
Normalization (93.6% likely):
Nightmare (6.4% likely):
Methodology:
First I took the average shift between 2020 and 2024 for all 50 states and found that it was +2.775% for Trump. Only 76% of votes have been counted in Arizona as of this analysis, so this might be slightly inaccurate, but reporting is +94% complete in all others.
There are two ways to view this shift:
As a unique good Trump / bad Kamala effect;
As part of a long term demographic / ideological shift among voters which will continue into 2028.
If it’s the former, then the uniquely good Trump / uniquely bad Kamala effect goes away in 2028. Democrats could run Martin Heinrich, and Republicans are stuck with JD Vance because Trump still controls the primary process.
Vance’s numbers aren’t great. Compare them to Trump’s:
Vance is at net negative -6.7, whereas Trump is net negative at -8.6. That may look worse for Trump, but you have to put it in the context. Only 82.9% of voters had any opinion at all on Vance, whereas 95.8% of voters have an opinion on Trump. If Vance is the candidate, one of two things are going to happen:
12.9% of voters are going to form an opinion on him, good or bad.
He’s going to remain an ambiguous, uninspiring, unexciting figure.
If Vance can’t capture low propensity voters in the rust belt the way that Trump has, on the force of pure cult of personality, then the Trump effect goes away, and things “normalize” toward 2020.
New York as a Swing State:
In the “normalization” model, Republicans are going to struggle to hold onto Georgia, North Carolina, and Pennsylvania. If they win all three, they are guaranteed to hit 286. But if they miss just one of them, they will be sitting on a razor's edge between 268 and 270. It’s going to be a very close election.
At the same time, the blue wall is going to erode and the number of potential swing states is going to shift.
Arizona and Nevada will lean more heavily toward Republicans due to the long-term trends of whiteshift or de-racialization among Hispanics.
Wisconsin and Michigan will remain swing states.
However, a new crop of swing states will open up in the northeast: New Jersey, New Hampshire, and even New York will become competitive battlegrounds.
If you’ve never been to New York, you might imagine that the whole state is just one big city, and maybe some suburbs. In reality, western New York is culturally similar to Pennsylvania, Ohio, and New Hampshire. I’ve drawn a red line on the map above to show how western New York clusters more with Pennsylvania than with most of the rest of New England.
New Jersey is an exceptional state in that it is highly educated but also relatively conservative. For more data on that, check out my article on trans identification, and how New Jersey is a consistent outlier.
What if Trump does well?
What if Trump ends the war in Ukraine and Gaza? What if he brings down inflation, brings back jobs, and builds a wall? What if opioid deaths, suicides, and all deaths of despair start to decrease? What if RFK bans the use of pesticides and microplastics and makes America healthy again, reversing our downward spiral toward lower life expectancies? What if Trump literally Makes America Great Again?
Also, what if the Democrats nominate another uniquely unpopular woman of color, like AOC? What if the Kamala effect is replaced with the AOC effect, and Trump establishes Reagan-levels of popularity?
In that case, we should expect the Trump effect to continue into 2028. Vance will start off with an almost guaranteed 285 electoral votes. The only question is how big the blowout will be. Will Republicans sweep not just Michigan and New Hampshire, but also New Jersey and New York? Democrats will have to fight for their lives not to lose New Mexico, Wisconsin, Virginia, and even previous blue strongholds like Minnesota and, yes, even Illinois.
That would mean Republicans could grab as many as 404 electoral votes. That would be the biggest electoral victory since Bush in 1988. Since it would be Vance’s first term, we could reasonably expect him to win again in 2032, securing Republican dominance through 2036.
Elena Kagan is currently 64 years old. In 12 years, she will be 76. Sonia Sotomayor is 70, and will be 86. John Roberts is 69, and will be 85. It seems likely that at least one of these liberal or moderate justices will be up for replacement, ensuring a 7/9 majority for Republicans.
This is the nightmare scenario for Democrats, and the wet dream of Republicans. This nightmare scenario is unlikely for the following reasons.
Caveat on Ukraine:
I want to be fair and make it clear, before badmouthing Trump, that Biden has been dragging his feet on Ukraine and not applying maximum pressure to Russia. There is also evidence that Kamala would be even worse and would further abandon Ukraine in an optical “good luck, honey! We’re rooting for you!” sort of way.
Nonetheless, I morally supported the Democrats simply because they refused to tolerate the kind of Putin-washing coming from MTG and Vance, and big networked or connected donors like Sacks and Musk. At least rhetorically, Democrats were better, and Republicans were worse. I will consider in this analysis the slightly possibility that Trump will be better on Ukraine than Biden. But before considering that slight possibility, I will consider the more likely possibility that he will side with Elon, Vance, and David Sacks.
I would also like to add, before bashing Trump’s foreign policy, that a disaster in Ukraine may have been baked into the cake for either Kamala or Trump. Just as the president doesn’t really control the fed, or the supreme court, or gas prices, or many other things that voters magically believe the president controls, the president is limited in their ability to bend the long arm of the “deep state,” the state department, the foreign policy establishment, and the military bureaucracy. Just because Ukraine is going to be given to Putin doesn’t mean that it is entirely Trump’s fault, and Biden certainly shares blame here. But whether or not it is his fault in a cosmic moral sense, he will ultimately have to deal with the consequences. Onto the analysis:
Ukraine:
If Trump gives Ukraine to Putin, America will become discredited in the eyes of France, Germany, and the rest of pro-Ukrainian Europe. A majority of Europeans support Ukraine, with the exception of Slovakians, Greeks, and Bulgarians.
It is true that between 2022 and 2023, support fell from 91% to 65% for sanctions, and 91% to 49% for weapons. However, only 41% of Europeans last year said "the most important thing is to stop the war as soon as possible, even if it means Ukraine giving control of areas to Russia.” As of 2023, a majority of the European population wanted to see the war continue.
Even if polling since 2023 has slipped further, to the point where the European public is tired of the war in Ukraine, we have to distinguish between popular opinion and the political establishment. The European bureaucracy and elites have been staunchly pro-Ukraine, because they believed that this is what America wanted. They were acting as loyal vassals. A Trump betrayal would send shockwaves through the European establishment, and result in serious geopolitical realignment.
Imagine the visceral hatred that the European establishment will feel for America. Billions (trillions?) of dollars wasted. Deindustrialization. The rise of the far right. Mass immigration. COVID hysteria. All of this is the result of Europe trusting and following America’s lead. And what was it all for? Just for Trump to betray Ukraine? Was it all in vain?
Being completely burned and out of options, European policymakers might begin a swift and radical swing toward Russia and China. If America refuses to guarantee European security and economic growth, then Europe has no other choice than to become a vassal of Eurasia. It simply does not have the demographic or economic growth necessary to provide for its own security. The “European Army” is a pipe dream.
It is hard to imagine Trump “threading the needle” and somehow abandoning Ukraine but simultaneously assuaging the feeling of raw betrayal on the part of Europe. At the very least, a Ukrainian betrayal will lead to an immediate cooling of relations between the EU and America. Europe and America might even begin a trade war in response to the Trump tariffs, and Russia could resume repairs of Nordstream.
This would be a total foreign policy disaster. Isolationists would love it, but it would destroy global confidence in the American empire and threaten the dollar as the world’s reserve currency. If Trump is willing to betray its greatest ally (Europe), who is not willing to betray? How can anyone trust America after this?
The war is already going badly under Biden, and Trump tariffs are going to destroy the economic basis for NATO. The Europeans are going to be really mad when they find out they destroyed their economies for nothing, and have to go crawling back to Putin for oil. That sort of humiliation has real consequences. Europeans are good at holding grudges.
How would the CIA feel about all this? The mission of the CIA is to expand American geopolitical influence over the globe. Would it take all of this sitting down? Or would it seek to punish the Vance/Thiel wing of the Republican Party? Maybe the CIA would get “creative.” It’s also possible that the CIA would prefer to abandon Ukraine in favor of a war with Iran. Such a large and powerful organization is not monolithic, and has many internal competing factions. It’s hard to speculate on which one will win without any public evidence.
Iran:
In the Middle East, Trump’s full backing of Israel could result in a war with Iran, repeating the failures of Iraq and Afghanistan on an epic scale. Thousands of Americans would die, trillions of dollars would be spent. Ultimately, Iran would not become a liberal democracy, but a breeding ground for international terrorism. Russia and China could covertly fund Iran (just like Reagan covertly funded Iran) and drain the American Empire. Israel might emerge victorious, but at what cost?
Netanyahu has been crying wolf for 10 years that the Iranians are just “weeks away from a nuclear weapon.” What if he turns out to be right, and the Iranians finally nuke Israel? Or what if they have another biochemical weapon in store (COVID 9000), and their drone swarms are able to penetrate the Iron Dome to deliver the payload? At the very least, the Iranians should be able to mine the straits of Hormuz, which will skyrocket oil prices. That’s great for Russia, but it wouldn’t reflect well on Trump, and might trigger a global recession.
Economics:
Tech has had two bubbles in the past three decades: the Dot-Com bubble of April 2000 and the Zoom bubble of October 2022. Are we headed for an AI bubble in the next four years?
Even more terrifying (or exciting), imagine that AI isn’t in a bubble, but fulfills all its promises. Here’s who loses their jobs:
That’s potentially 15% of the workforce. Let’s put this in context.
Even if Trump supercharges the economy and reduces unemployment to a historically low 2.5%, AI unemployment would result in a total unemployment rate of 18%, which would be higher than the worst of COVID and anything since the Great Depression. Even if truck drivers are safe, that would only prevent 2.19% of the job losses. In a very moderate scenario, we should expect AI to permanently disrupt 7% of the job market, leading to 9% unemployment, which would be just as bad as the height of the Great Recession in 2010.
Permanent loss:
It’s important to note here that there is a huge difference between “temporary job loss” and “permanent job loss.” An example of temporary job loss would be a recession, like COVID, where restaurants shrink and have to close down. As a result, waiters and baristas lose their jobs. However, after the recession is over, the jobs come back, and things are good again. Or, if one restaurant closes down, those same workers can move to a different town or state and start a new life in the same industry with the same skills.
Permanent job loss is a different beast. This is when an entire industry is fundamentally transformed, and once the jobs are gone, they are gone forever. They’re never coming back. This is what happened to the Rust Belt due to automation — it wasn’t just globalization, it was fundamental technological changes in the process. Even the Chinese are using robots. Free marketers will tell a coal miner to “just learn to code,” but what do you tell the low-level programmer who has his job stolen by AI? Learn to be smarter?
Job shift cost:
Yes, AI will create new jobs, as just automation created new jobs. But people aren’t fluid entities with a singular fungible “employability score” that can be flawlessly transferred from one industry to another. A person who has spent 20 years digging for coal, or programming excel spreadsheets, or truck driving isn’t just going to seamlessly find a new job the next week.
Job loss causes psychological harm, depression, and hopelessness. A percentage of people give up and turn to drugs. You can be a libertarian and not care about these unfortunate souls. Maybe you think they deserve to die if they can’t adjust. But I’m not speaking in moral terms, but factual ones. Job losses due to AI will hurt Trump, even if they simultaneously create new jobs in new industries.
Obama was able to be re-elected in 2012 because he could blame everything on Bush (and he was also able to make progress during the first term). But can Trump blame the AI revolution on Biden? Or will Democrats successfully claim that his deregulation led to the AI job bust? Even if that would be a lie, would it be convincing to voters?
For the record, I fully support 15% of people losing their jobs to AI. Anyone who does a rote or repetitive job should be unemployed and forced to reorganize their lives around a new set of skills. But for the purposes of elections, this would almost certainly hurt Trump.
Trump recession map (30% chance):
This is the map if in 2028:
10+ million Americans lose their jobs to AI;
And/or the AI bubble pops, causing a recession;
Trump abandons Ukraine and fights a trade war with Europe;
At least one country begins a process to leave the EU or NATO;
A failed war with Iran results in:
100,000 dead Israelis;
Biochemical terrorism;
Skyrocketing oil prices;
Global victory for Russia and China.
If Democrats are able to run a white-male-Bernie style candidate against Trump’s chaos, it will be a sweeping victory.
Summary.
2028 is going to be tough for Republicans. The unique charisma of Trump will be gone, and replaced with JD “hand on my chest” Vance.
Democrats might learn from the Kamala disaster and run a white male. New York will, surprisingly, become a swing state, partially because of its large Hispanic population, and partially because of its large population of non-college educated whites.
The economy under Trump might get better, but it might also get much worse for reasons out of Trump’s control. AI could increase GDP, but it could also be massively disruptive for lower and middle class white collar workers. It’s easy to say “new jobs will be created,” but tell it to the rust belt. Elections aren’t fair, but they do have an underlying logic.
Trump's foreign policy, as laid out by Vance, is highly risky. Blowing up Nordstream in 2022 and then abandoning Ukraine in 2025 is a really bad look. Germany and France are going to be furious that they wasted billions (trillions?) of dollars on a war where they were promised support.
The result could be a trade war between the EU and the United States. This conflict between America and its closest ally (Europe) will decrease confidence in the USD as the global reserve currency. American debt is safe to buy right now, but if America is seen as increasingly isolationist, countries like China will stop buying our debt. The deficit will suddenly gain relevance for the domestic economy. As countries dump the dollar, the USD will lose value. If you thought inflation under Biden was bad, wait until you see de-dollarization under Trump.
I am confident that Israel can win a war with Iran, but in the short term, the Iranian use of biochemical or nuclear terrorism would be bad for Trump. Oil prices would almost certainly spike, and Russia would be the biggest victor of a war with Iran. Russia would emerge as the world's biggest energy and agricultural power, fueling China's military machine. That might encourage China to move on Taiwan, and discredit the Trump/Vance foreign policy even further.
The best case scenario:
The best case scenario is that Trump will sideline Vance and do some surprising and unexpected good things:
Maintain the status quo in Ukraine and keep Europe happy;
Defeat Hamas, but simultaneously thread the needle and quietly restraint Israel from expanding a limited war in Gaza into a war with Iran;
Or, somehow defeat Iran quick enough that oil prices don’t skyrocket and Israel doesn’t experience mass casualties;
Keep the AI boom going, but compensate for inevitable job losses with 10 million new AI-proof “mystery jobs” that have good wages at entry level.
Here’s my estimated likelihood of these outcomes:
40% chance that Trump creates peace in the Middle East and neutralizes Iran without economic disturbances. He has surprised the world before with the Abraham Accords and could do it again.
40% chance that Trump ignores Vance, resists Putin in a way that maintains American credibility, and avoids a trade war with Europe.
40% chance that Trump avoids high unemployment (over 6%) AND has no recession for four years. 80% chance that one or another happens, or both. AI is ultimately good in the long term for the economy, but the next four years will upset voters. That’s not Trump’s fault, but he will have to deal with it, just like he dealt with COVID.
Individually, it seems like Trump has a chance of winning on 1/3 or 2/3 of these issues. But taken together, things are looking pretty bad. Even if you think it’s 50/50 rather than 40/60 odds, this is still flipping heads three times in a row. As a result, even a 50/50 shot on each issue only gives Trump a 12.5% chance of avoiding all three of these crises simultaneously over the next four years. It’s going to be a rough four years.
Given the challenges that Trump faces, I don’t expect Vance (or any other Republican) to win in 2028. Conservative dreams will be overturned. Abortion will be legalized at the federal level; trans gender surgeries will increase; immigration will continue; birth rates will drop. Democrats will win with a white male moderate in charge.
There is a small chance that Trump can overcome all these monumental hurdles and achieve Obama-Reagan levels of popularity. In that case, Vance could rewrite the map, winning huge upsets in New Jersey, New York, and even Illinois.
Unfortunately for Republicans, I don’t think Trump is competent enough to run the country at this point. He is 78, and when he finally ends his term, he will be older than Joe Biden is currently. Instead, Vance and his allies will run the government. It could look like a rerun of Bush-Cheney, where Russia is given a free pass to invade its neighbors, trillions are wasted in the Middle East, and the economy implodes. America becomes, once again, hated and distrusted by our closest ally (Europe). Democrats can try to turn things around in 2028, but some of the damage will be permanent.
Perhaps I am too pessimistic, and I would be happy to be wrong. Maybe the isolationists are right, and a Chinese takeover of Europe will be great for America. We will see.
Im actually quite optimistic about Ukraine. Z will f-up a peace deal big, we've had credible queues that the Kremlin takes liberal fearmongering on Trump as canon. They will be maximalists, refuse the first draft offered by Trump. He will feel personally humiliated and do in a year what Biden and rest of them hadnt done in 4. The war wont end in 2025. Maybe not even 2026. After that is anyone's guess.
The AI bit is easy. If AI implodes, there will be a few billions in bailouts and within a year no-one outside of the Bay will care. If AI works, just give people money. As long as swing-voters have cash in their hands or some make-work job, it doesnt matter.
The rest you might be correct, but I have no way to access rn, so will just take your word for it!
Great predictions, but I think you seriously overlooked the possibility of Trump's administration sliding into Authoritarianism (Rigged elections, Martial Law), invalidating the 2028 election prospects.