It is intellectually interesting to be going into an election with such strongly competing information about who is likely to prevail. All but a few insurgent poll outfits show an easy Biden victory. If these models are in any way accurate, Trump has no chance.
The few insurgents, meanwhile, predict a massive Trump victory.
My sense is that the polls are models, and our ability to predict complex phenomena well enough to model it is quite weak. We think we can because we can model things like physics well enough to put a probe on Mars, or hit an asteroid moving a million miles an hour and millions of miles away. Yet those are relatively simple things to do, because the forces acting on the planet Mars or the asteroid are few in number. You can model the increasing effect of gravity as the asteroid approaches the sun using an inverse square law we've known since Newton.
Try modeling which way a small, light balloon will go in a windstorm here on Earth, though, and you can't model it accurately for a hundred yards. The fluid dynamics of the atmosphere are too complex. We deal with turbulence by the application of main force, not by understanding exactly how it will affect an aircraft.
Our models of complex phenomena are reliably terrible. Viral models were wildly inaccurate; economic models are never any good; climate change theory is based around models we have no reason to trust, and good reasons to distrust if we are honest with ourselves about their predictive history since the 1970s or so.
So I'm going to found my guess on what I think I know about American politics, rather than the modeled outcomes. What I think I know is this:
* Americans hate political violence, and political violence by the Left has been ongoing since the summer. Democrats have actively abetted it in Democratic cities, as for example by not prosecuting rioters. Democrat-controlled cities are boarded up today, and everyone knows which side is the danger causing it. Americans will not like this.
* Though the police are having a bad year in terms of public confidence, they're still more trusted than any branch of elected government. Every major police organization has endorsed the President's re-election as a means of quelling the violence.
* Restrictions on freedom anger Americans, and the Biden campaign is promising new lockdowns and mandates if elected.
* Americans hate being told that they are racists, or that America is racist, and love being made to feel patriotic and hopeful. The Left has built its campaign around the 1619 Project and racial resentment, while the Trump campaign has been relentlessly patriotic and has engaged in positive outreach to minorities.
* Americans love economic booms and hate recessions or depressions. Trump presided over a wild economic boom until COVID, and even now we are in a V-shaped recovery that will flesh out into a boom if he is re-elected. Biden is promising economy-killing tax increases and lockdowns (and nobody loves paying more taxes anyway).
* Trump has started no new wars, and has slimmed our footprint in existing wars as far as he can manage. He has avoided creating pools of instability out of which things like ISIS can spring. He has also overseen the development of unheard-of normalization between Israel and Arab states, and economic normalization in the Kosovo region. Peace and Prosperity are hard to beat.
* The intense hatred facing Trump is largely a class hatred, and although this class controls nearly all of our media and much of our government, it is not large. Elizabeth Warren was the candidate of the college educated white women who make up its core, and she didn't even win her own state in the Democratic primary. They will doubtless persuade many voters through their relentless control of information, but many more regard them with scorn.
* Biden's record includes the 1994 Crime Bill, and repeated threats to cut Social Security. Even his own machine voters have to approach him out of a sense of party loyalty rather than personal affection. Trump, meanwhile, draws rock-star crowds everywhere he goes.
For all those reasons, then, I think Donald Trump will easily win tomorrow. Maybe I'm wrong; maybe the experts will prove right, and the models will work for once.
If so, I'll be surprised. I've been surprised before; I was surprised that Trump won the Republican primary, and again when he got elected. I was surprised when he achieved a major development in Middle East peace, when he didn't start new wars, and when his Presidency was marked by effective outreach to non-white voters rather than a Brazilianization of our politics into racialized factions.
He's surprised me a lot, in other words, but generally in ways that worked to his advantage. My guess is that the surprise that is coming will be for his opponents, as it has often been.
POSTSCRIPT:
It occurs to me that the one factor I'm ignoring is the psychological effect of COVID itself. I don't watch TV, but when I speak to those who do it's clear that they're being subjected to an intense psychological operation designed to make this disease seem terrifying and momentous. In fact -- per Worldometer -- there are only ~85,000 serious or critical cases of COVID currently in the entire world. 99% of cases are mild, and even then it's only 11 million people suffering the mild symptoms in the whole world. Almost everyone recovers.
It's not worth our attention, in my view; just one more risk among the many we run every day, and far from the worst one. I'm excluding it from my calculations because I think it's worth excluding, and I'm outside the psychological bubble that constantly reinforces it as a concern.
If Trump loses tomorrow, it'll be because of that psychological operation combined with the other ones that were designed to hide his accomplishments and play up his mistakes and bad behaviors. That will by itself be remarkable; it will underline that Americans are much more manipulable than is good for a free people.
Nevertheless, I don't expect it. I think the reason we keep hearing about the tyrannical governors and governments coming up with new restrictions is that most people -- not just most Americans -- have decided to proceed with their lives. They're not going to vote to cast themselves into a new darkness of suffering and poverty to address this annoyance. I could be wrong about that, too, of course; my tolerance for risk is very high.
8 comments:
My understanding is that polls have used landlines to contact voters. Currently only about 7% of households have landlines. That's going to mess with accurate predictions.
You're back!
The polls are garbage, and have been for awhile. Doesn't stop those wanna-be prophets from predicting everything, though.
I have seen it surmised that the polls are being misled with false responses too.
Correction: about 40% of houses have landlines. Only about 7% of households have landlines only.
You're back!
Sorta. I like having a place to jot down my thoughts if I have any. This set has been with me for a few days, and it's good to write them down so they can be out of my head. Now I can think about something else, you know?
I hope for a Trump landslide, one that far exceeds the margin of fraud, and that the Republicans keep the Senate. I pray for a miracle of peace, that the cities don't erupt, no matter who wins or what things look like as the absentee and mail-in ballots come in.
LittleRed1
https://twitter.com/Ymarsakar/status/1294394998197309442
That was my actually July/August predictions. Technically they aren't predictions. They are things I do to make happen, but that's another opsec problem issue.
My current "predictions" or prophecies are that the election remains contested until Nov 12-20. Main sewer claims Kama as victor. And they will realize what Barret's confirmation means by January 20th, and whomever survives past that, is the real Presidential victor.
I was surprised that Trump won the Republican primary, and again when he got elected. I was surprised when he achieved a major development in Middle East peace, when he didn't start new wars, and when his Presidency was marked by effective outreach to non-white voters rather than a Brazilianization of our politics into racialized factions.
Surprised... Trump is a type of Ymar. Is it so surprising that Ymar would win agaisnt Republicans and do such things?
Well, I suppose it is, but given you have been in communication with a Ymar for so many years, Grim, did you just not make the connection at the time?
I was not surprised at any of this, because I began realizing Trump was a type of Ymar, which is partially why I wasn't gonna jump on no "Trump band wagon".
Gringo, from what I've read the more fundamental problem is that pollsters have gone from about a 50% response rate before 2000 to about a 2% response rate now, regardless of the form of contact. The switch from landlines is a part of that but not the whole thing. The selection bias is tremendous, and most poll respondents are white female Millenials (anybody surprised?). Sampling until you get enough responses to build a decent model of the population is extremely costly, and so pollsters are forced to do things like use responses from a 23 year old with a degree in Peruvian Feminist Literature working as a barrista as a 'working white male'. Do you think the barrista has the same world view as a 23 year old working construction or in an Amazon fulfilment facility? Then add in the bias inherent in providing results to obviously biased organizations like the FNM, and it's now wonder that polls look to be pretty much exercises in GIGO.
Predictions ...
Trump wins but I think relatively narrowly again. 275 to 263 in the EV, popular vote 66.9M to 66M. He gives up MI, PA, and AZ, holds WI and FL, flips MN and NV (my surprises for the night). Despite all the current angst over it, PA is irrelevant.
Post a Comment