As you note, Trump's net-approval floor - more likely his net DIS-approval floor- is probably higher that it might be in a non-polarized environment. That said, I could see it slipping into the mid-30s by the midterms (all things being equal), because I think that's the percentage of the hardest core of his hardcore MAGA base.
Elliott -suppose Trump's approval rating drops to the 30 - 35% range. What does that do for the Dems Senate chances in 2026? It's a rough map, and sure Trump with that low level of support probably serves up NC and ME on a silver platter, but is it enough to get some combination of the remaining seats up for grabs to get a majority? IA, OH (especially if Ramaswamy is the gubernatorial candidate), TX, FL are up there, and there's precedent for Trump costing the GOP Senate seats in MT and AL, but is there any idea what kind of impact a lame duck President this unpopular has on midterms?
To me, 13% represents the absolute core of MAGA supporters -- based on the polling numbers that represent support for Trump's fantasy to annex Canada. The number also resembles support for George Wallace, back in the day. I will eventually do a cartoon, citing these numbers.
So he'll wag the dog and start a war. Hope he doesn't and the particulars of the war matter. A war w NATO probably won't help him. But some others might.
Hi Elliott, thanks as always for thoughtful and thought provoking analysis!!! I haven’t looked closely enough to answer this question but wouldn’t it be helpful to normalize for how much there is to lose? For example if someone (A) starts out at 100 days with 60% approval vs. someone (B) starts with 40% won’t that change the prediction? President B will obviously never get to 0. I believe the lowest presidential approval rating ever is about 20% and Nixon only got down to 24%. It seems that there is a lower limit that perhaps is different in polarized vs non polarized times, but it’s something, so Trump can’t lose 40 or even likely 25… could he lose 20? So perhaps combining the two to account for how much there is to lose to get to a historical minimum ( president A is 35 pts from 25 but B is only 15 points from 25). I bet this is what you are doing intuitively. Thank you! Best, Marsha
I was having the same thought, I would guess a regression that included starting approval rating would have Trump declining less just because he's starting from a much lower point.
I was thinking about this too, but didn't want to overfit the model. Intuitively, I think this is the right way to think about what's going on with Trump, and why I think we're unlikely to see a -40 net rating, for example.
Is there any way to model (by looking at other countries) how low approval has to get before there is some kind of mass effort to push the leader out? I agree it’s unlikely to get there here and now with polarization, but how can any leader in any system survive 70 percent disapproval?
My impression of the scholarship is that institutions in America are too strong to allow a mass uprising to remove Trump from the White House if he hit 70%, but that’s getting near Arab Spring territory. Truman’s disapproval approached 70% in early 1952, but at that point there wasn’t a “need” (so to speak) to remove him since he wasn’t running for president again. Nixon peaked at 65%, and then he resigned.
Great question and really interesting response! What do you know about the idea that it takes 3.5% of a population to stop authoritarianism? I haven’t seen any analysis of this and don’t know if it applies equally everywhere. Appreciate any insight you have on this!
Here’s one article. I believe Ruth Ben-Ghiat has done some work on this, but perhaps it’s someone else… in any case a nagging question I’d be interested in for sure!
I simultaneously want Donald Trump to wreck his particular brand of politics as much as possible but want the least damage done to the United States of America as can be. Tricky stuff.
As you note, Trump's net-approval floor - more likely his net DIS-approval floor- is probably higher that it might be in a non-polarized environment. That said, I could see it slipping into the mid-30s by the midterms (all things being equal), because I think that's the percentage of the hardest core of his hardcore MAGA base.
Someone may have noticed this already. Trump was inaugurated Jan 20, 2025 not Jan 1
——
Well, because that's the length of time between January 1, 2025, when Trump was inaugurated as president for a second time
Elliott -suppose Trump's approval rating drops to the 30 - 35% range. What does that do for the Dems Senate chances in 2026? It's a rough map, and sure Trump with that low level of support probably serves up NC and ME on a silver platter, but is it enough to get some combination of the remaining seats up for grabs to get a majority? IA, OH (especially if Ramaswamy is the gubernatorial candidate), TX, FL are up there, and there's precedent for Trump costing the GOP Senate seats in MT and AL, but is there any idea what kind of impact a lame duck President this unpopular has on midterms?
How low could Trump go? 13 Percent approval.
To me, 13% represents the absolute core of MAGA supporters -- based on the polling numbers that represent support for Trump's fantasy to annex Canada. The number also resembles support for George Wallace, back in the day. I will eventually do a cartoon, citing these numbers.
In the meantime -- great data.
So he'll wag the dog and start a war. Hope he doesn't and the particulars of the war matter. A war w NATO probably won't help him. But some others might.
Hi Elliott, thanks as always for thoughtful and thought provoking analysis!!! I haven’t looked closely enough to answer this question but wouldn’t it be helpful to normalize for how much there is to lose? For example if someone (A) starts out at 100 days with 60% approval vs. someone (B) starts with 40% won’t that change the prediction? President B will obviously never get to 0. I believe the lowest presidential approval rating ever is about 20% and Nixon only got down to 24%. It seems that there is a lower limit that perhaps is different in polarized vs non polarized times, but it’s something, so Trump can’t lose 40 or even likely 25… could he lose 20? So perhaps combining the two to account for how much there is to lose to get to a historical minimum ( president A is 35 pts from 25 but B is only 15 points from 25). I bet this is what you are doing intuitively. Thank you! Best, Marsha
I was having the same thought, I would guess a regression that included starting approval rating would have Trump declining less just because he's starting from a much lower point.
I was thinking about this too, but didn't want to overfit the model. Intuitively, I think this is the right way to think about what's going on with Trump, and why I think we're unlikely to see a -40 net rating, for example.
Is there any way to model (by looking at other countries) how low approval has to get before there is some kind of mass effort to push the leader out? I agree it’s unlikely to get there here and now with polarization, but how can any leader in any system survive 70 percent disapproval?
My impression of the scholarship is that institutions in America are too strong to allow a mass uprising to remove Trump from the White House if he hit 70%, but that’s getting near Arab Spring territory. Truman’s disapproval approached 70% in early 1952, but at that point there wasn’t a “need” (so to speak) to remove him since he wasn’t running for president again. Nixon peaked at 65%, and then he resigned.
Great question and really interesting response! What do you know about the idea that it takes 3.5% of a population to stop authoritarianism? I haven’t seen any analysis of this and don’t know if it applies equally everywhere. Appreciate any insight you have on this!
Have never heard of this. I'll read around
Here’s one article. I believe Ruth Ben-Ghiat has done some work on this, but perhaps it’s someone else… in any case a nagging question I’d be interested in for sure!
https://www.bbc.com/future/article/20190513-it-only-takes-35-of-people-to-change-the-world
In the U.S. context in 2025, 3.5% would be roughly 13m people. Doesn't immediately strike me as a lot
I simultaneously want Donald Trump to wreck his particular brand of politics as much as possible but want the least damage done to the United States of America as can be. Tricky stuff.