I would if I were him. Nothing will fire up his base like blowing up a bunch of stuff in a country with brown people, even if his base’s kids will eventually be the ones who die cleaning up the mess over there. And (as we saw in 2003-4), lots of people in the middle are gullible enough to believe what the WH says in the lead-up to a war, and then reluctant to change leadership once the war is underway.
Agreed, it sounds like you are weighing the economy quite heavily. It’s a valid approach, but 18 months is a hugely long window for the economy to change (like anything else).
Another issue with overweighting the economic aspect… historically, if the stock market was doing well, average Americans would see that in their paycheck. That’s been less true for quite a while. In fact one could argue this is why average people didn’t choose Hillary as a third Obama term. The economy was doing quite well but average Americans weren’t seeing it, and many still aren’t.
As for the incumbent advantage… well, there is that, but Trump’s popularity in early presidency has (AFAIK) been lower than anyone else in early presidency. His popularity doesn’t seem to be tracking with the economy.
Mainly this part. But the timing will be everything. The election has to happen before a lot of the MAGAbots kids are killed over there.
And the Iran hostage crisis hurt Carter.
Harris could win vs Trump.
Biden is very popular among minorities, but yes, gaffes.
Kind of funny (and sadly, true) we think gaffes might be a problem for Biden, when Trump is basically a walking 24/7 gaffe machine.
It has never been a fair fight wrt the standards Dem candidates are held to vs Trump and I just don’t fucking understand it.
That’s certainly true.
Obama didn’t have very long coat tails in general, and Hilary wasn’t VP or anything that could be associated with the economy. Whether or not Obama’s coat tails help Biden now (providing he gets the nod) remains to be seen - it’s been four years. Maybe if Obama enthusiastically endorses Biden, early and often, maybe it will. Maybe.
That’s the elephant in the room - I no longer trust polls when it comes to Trump. All the polls said Hilary was going to win, and she didn’t. Back in 2016, there were polls saying that various Democrats would beat Trump head-to-head.
I do think there is a real factor of “you aren’t really going to vote for HIM, are you?” behind a lot of the polling and coverage of the 2016 election. And I don’t think that has gotten any less. There is some truth to the idea of living in a bubble.
But we shall see. I find it interesting that, even on a message board as progressive-leaning (to say the least) as the SDMB, as of this post those voting that they think Trump will be re-elected are outnumbered by those thinking he won’t - by one vote.
Regards,
Shodan
Most of the national polls said she would win the popular vote, and she did. The state polls were much more mixed, but such that Nate Silver rated that she had about a 70% chance or so of winning the electoral college (which meant Trump had a 30% chance). The polls really weren’t that far off – this narrative is vastly inflated.
Perhaps you’re right, and Trmp not allowing the election to proceed is too far. One of Trmp’s biggest assets, though, is people’s willingness to underestimate him. He comes across as a complete incompetent buffoon, but he is PRESIDENT OF THE UNITED STATES! As they say in sports, scoreboard, bitches.
Maybe what will actually happen is Tr*mp simply refusing to acknowledge any loss.
ETA: I should say, I think this is an extremely germane point, because you all are having this discussion like nothing in US American politics has changed. Tr*mp has changed everything, and if the Democrats don’t see that and work with it, they will lose again.
Apart from being wrong, they weren’t that far off.
My point is that I don’t think the polls are that far off now.
Old joke: Football player is about to lose his eligibility for flunking all his courses. The coach pleads with the professor to pass the player. Finally, the professor agrees that if the player can answer one question right, he will pass. The coach and player agree.
The professor asks the player: “How much is nine plus four?” The player, after much thought, answers “Eleven?” The professor says, “Sorry, you flunk.”
“Oh come on, prof!” says the coach. “He was only off by one!”
My concern is that the polls are only off by one.
Regards,
Shodan
The polls are almost always off by one or two or more. If someone says the polling aggregate shows X winning 51 49, and another guy says the polling shows that X will lose 60 40, then who is more accurate when the result is X losing 51 49? I think the first one.
If I say a probability of a certain event is 30%, and it happens, was I wrong? Am I only correct if I said 51% or higher?
Personally, if the weather forecast says 30% of rain, and it actually rains, I wouldn’t call the weather service and tell them they were wrong.
The big difference is that back then we were guessing, and now we’re factoring in observed events.
Yet again I point out… even if polls are giving unreliable figures, that unreliable figure has gone down, not up. Given that Trump supporters are more emboldened, it seems unlikely that more people are lying to the polls, and more likely that he’s just getting less popular.
And… Trump’s behavior and governance is now a revealed factor as well. Some people approve, some people don’t, but nobody can now say “Give him a chance to show us who he is.” That’s going to peel off some conservatives and swing voters.
Polls are not forecasts. They are a sampling exercise. If the observed behavior doesn’t match the sampling behavior, then they were wrong.
When you compare polling results to election outcomes, you are treating the poll as a forecast. (Unless it’s an exit poll.)
Also, many people & organizations made predictions (i.e. forecasts) based on polls. The 30% probability of Trump win was a forecast by FiveThirtyEight based on polling data.
Polls are not probabilistic exercises, period. You compared them to weather forecasts and that’s wrong.
This is a non-sequitur. Polling data is polling data. Forecasts are forecasts. One is probabilistic, one isn’t.
Trump barely squeaked by in 2016, there is zero chance he does better in 2020.
It depends on the state of the country and the quality of the opposition. I’d also keep open the possibility of a third party spoiler, who comes in and shaves off just enough votes to screw over the Dem nominee. Maybe Howard Schultz or Jill Stein runs, or maybe a jilted Democrat like Bernie or Yang gets disgruntled because he’s not taken as seriously as he believes he should be.
This is ignorance of what polls represent. A poll reported as a 4 point lead (as the average of the polls said before the 2016 election) is NOT stating that the result will be 4 and results with a lead of 6 or 2 are not “wrong”. Individual polls and polls in aggregate are stating probabilities. There are confidence intervals in polls which do not disappear when they are aggregated. And polls must be interpreted with awareness that polling errors often correlate together. If one is for some reason other than sampling off by 2 one way the other, others will likely also be too.
If you believe that the polls saying candidate A is leading the popular vote by 4% and she ends up leading the actual vote by 2% means that polls are worthless then you are simply wrong. And that is precisely what happened in 2016. 2 points off one way or the other is the rough average for how the national polls perform. And have less confidence in the precision of state level ones.
Forecasting is done by using the polls and by interpreting them. It was very reasonable to look at the polls and to state with some confidence that the odds were good that the result would fall somewhere roughly between Clinton +6 and Clinton +2. It was unreasonable to look at them and say that the polls meant the result was going to be Clinton +4 exactly. And smaller of a Clinton blow out of +8 or of a tie.
Problem was the exact way the votes played out to Clinton +2, within the expected result range, which resulted in a Trump EC win. Something that 538 in fact modeled as a distinct possible outcome in 10% of their simulation runs.
If I order a lab test and ignorantly misinterpret the result of the test then I am wrong, not the test.
The thing is, he doesn’t have to do better. All he has to do is the same as in 2016.