The Economist: Forecasting the US elections

[new_patomic]

Feels rather nitpicky to me.

I dont think anyone wants to repeat the stupidity of HuffPost or others saying Clinton was a 90+ percent shoein but even with uncertainty the sort of polling error and change of fortune it would take for Trump to win the PV would be simply extraordinary. It certainly isnt 1 in 20.

[The Mikado]

Feels rather nitpicky to me.

I dont think anyone wants to repeat the stupidity of HuffPost or others saying Clinton was a 90+ percent shoein but even with uncertainty the sort of polling error and change of fortune it would take for Trump to win the PV would be simply extraordinary.

^^^

There's no way you can feed in "9 point average lead for challenger three months out in the middle of a recession with the highest unemployment since The Great Depression" into an election model and not have the model spit out "LOL incumbent's f**ked." There's just no way around it, that's how models are.

Like, the only way you can talk about "Maybe Trump can come back" is gut/feeling stuff at this point and that's not what models do. If Trump does come back, it means everything people thought they knew about electoral politics is wrong, not that the models are wrong.

[new_patomic]

Feels rather nitpicky to me.

I dont think anyone wants to repeat the stupidity of HuffPost or others saying Clinton was a 90+ percent shoein but even with uncertainty the sort of polling error and change of fortune it would take for Trump to win the PV would be simply extraordinary.

^^^

There's no way you can feed in "9 point average lead for challenger three months out in the middle of a recession with the highest unemployment since The Great Depression" into an election model and not have the model spit out "LOL incumbent's f**ked." There's just no way around it, that's how models are.

Like, the only way you can talk about "Maybe Trump can come back" is gut/feeling stuff at this point and that's not what models do. If Trump does come back, it means everything people thought they knew about electoral politics is wrong, not that the models are wrong.

I suppose you could make a point about exogenous events. We don't know what will happen in August let alone October. We could have a cure, Biden could have a stroke, aliens could invade and the election gets cancelled, who knows.

But the point with those is that they are by their very nature unpredictable. Some amount of uncertainty should be baked in but it seems like they just want to find a statistical justification to say it should be higher because anything having 1 in 100 odds in July feels wrong.

[True Federalist (진정한 연방 주의자)]

The only way Trump could conceivably win the PV would be if rampant coronavirus virus keeps down the total vote in New York, California, and other Democratic strongholds while not affecting Republican strongholds.  But even if that were to happen, I think the net result would be that voters angry at Trump would be more likely to go to the polls than ones who are content with him.

[GeorgiaModerate]

Elliott Morris has a Q&A Twitter thread going on right now, taking questions about the model:

https://twitter.com/gelliottmorris/status/1288252792268783617

[wbrocks67]

It appears G. elliott got flack from Dave Wasserman and others and ended up reworking his model. Now he's claiming that Trump had a better week last week (not true, according to the polls), and Biden dropped a whole 2% in his model.

These people need to stop letting other people affect how they do their work

[GeorgiaModerate]

It appears G. elliott got flack from Dave Wasserman and others and ended up reworking his model. Now he's claiming that Trump had a better week last week (not true, according to the polls), and Biden dropped a whole 2% in his model.

These people need to stop letting other people affect how they do their work

I don't agree with the last part.  This isn't an exact science.  It's perfectly reasonable for other experts in the subject to offer constructive criticism, and I think that's the spirit in which it's been offered and received.  I think it's a good thing if it results in improvements in the model.  The Economist's model is out there before the 538 or NYT models, so it's going to be the first target of such criticism (and the Economist team deserves a ton of credit for being the first ones to publish!)

[wbrocks67]

It appears G. elliott got flack from Dave Wasserman and others and ended up reworking his model. Now he's claiming that Trump had a better week last week (not true, according to the polls), and Biden dropped a whole 2% in his model.

These people need to stop letting other people affect how they do their work

I don't agree with the last part.  This isn't an exact science.  It's perfectly reasonable for other experts in the subject to offer constructive criticism, and I think that's the spirit in which it's been offered and received.  I think it's a good thing if it results in improvements in the model.  The Economist's model is out there before the 538 or NYT models, so it's going to be the first target of such criticism (and the Economist team deserves a ton of credit for being the first ones to publish!)

Yeah, but if you're changing your model just because of criticism, that doesn't give me too much confidence in it. He should want to make it the best it can be, but he withered under literally one person making criticisms.

[GeorgiaModerate]

The map on the model page has now been improved.  You can now hover over any state and see the individual probability for that state.

https://projects.economist.com/us-2020-forecast/president

[Beefalow and the Consumer]

The map on the model page has now been improved.  You can now hover over any state and see the individual probability for that state.

https://projects.economist.com/us-2020-forecast/president

Nice. I was hoping they would add that.

AZ and NC are both on the verge of moving to likely Biden. (NC has waffled around that threshold for weeks). If they both make the jump, that would put 334 EVs in the "likely Biden" column, and make him the clear favorite in three large states past the tipping point. If that's the state of things on Nov 2, Biden will be a lock.

[Sir Mohamed]

I noticed their model doesn't take any 3rd parties into account. All the state by state vote share projections and the NPV exactly add up to exactly 100%. Otherwise, a fine page.

[pppolitics]

The map on the model page has now been improved.  You can now hover over any state and see the individual probability for that state.

https://projects.economist.com/us-2020-forecast/president

Nice!

That's one that I noticed was missing.

[Skye]

So, the past few days, several pundits (Silver, Cohn, Trende, Wasserman) have been giving their "thoughts" on the model (i.e. mostly critiques) and Morris has been defending it, but this by far the most confrontational of the interchanges:

https://twitter.com/NateSilver538/status/1288905871347646467

[new_patomic]

Silver just sounds more and more pissy.

[GeorgiaModerate]

Silver just sounds more and more pissy.

As others have said, Nate's position in this argument would be stronger if he'd just publish his own model.

[wbrocks67]

A) Nate hasn't even posted his model yet so he has no right to be offering any critiques at this moment in time
B) How is Eliott's model that bad? Biden literally has over a 90% chance of winning right now. I don't see the problem?
C) Eliott, however, is giving in again- saying that the popular vote, which shows, what Biden, +8/9, is still too high? When that's *literally* what the poll averages are showing right now?

It's almost as if 2016 screwed everyone so much in the heads that they just will refuse to believe that Biden could possibly be ahead this much, even though all the data and evidence is pointing to that.

[slothdem]

A) Nate hasn't even posted his model yet so he has no right to be offering any critiques at this moment in time

Maybe this is conspiratorial of me but I think one of the reasons why 538 hasn't posted their model is because it shows the same thing as Morris's. Like 1996 and 1984, this is a "cards face up" election.

[Skye]

C) Eliott, however, is giving in again- saying that the popular vote, which shows, what Biden, +8/9, is still too high? When that's *literally* what the poll averages are showing right now?

I think he's referring to the part of the model that says Biden has a 99% chance of winning the popular vote. That's basically uber safe territory and I definitely can see criticism towards that part.

[Beefalow and the Consumer]

A) Nate hasn't even posted his model yet so he has no right to be offering any critiques at this moment in time

Maybe this is conspiratorial of me but I think one of the reasons why 538 hasn't posted their model is because it shows the same thing as Morris's. Like 1996 and 1984, this is a "cards face up" election.

I remember one of my friends in 1996 telling me, "the media already decided Clinton's gonna win." I had no appropriate response for that.

[Del Tachi]

Nate Silver doesn't need his own model to recognize The Economist's is trash.  Their model is too reliant on "fundamentals", yet because so much of the economic date is so out-of-sample (at least considering post-WW2 elections) they arbitrarily adjust it to produce what looks (to them) to be a "realistic" scenario.  That's p-hacking.

[Ferguson97]

Nate Silver doesn't need his own model to recognize The Economist's is trash.  Their model is too reliant on "fundamentals", yet because so much of the economic date is so out-of-sample (at least considering post-WW2 elections) they arbitrarily adjust it to produce what looks (to them) to be a "realistic" scenario.  That's p-hacking.

What are "fundamentals" even supposed to mean?

[GeorgiaModerate]

https://twitter.com/gelliottmorris/status/1289198771780239364

The linked article is worth reading.

[GeorgiaModerate]

https://twitter.com/gelliottmorris/status/1289941988880445440

[GeorgiaModerate]

Interesting:

https://twitter.com/gelliottmorris/status/1290339003070324738

[wbrocks67]

Interesting:

https://twitter.com/gelliottmorris/status/1290339003070324738

Rasmussen has Trump's approval at like 40% among black voters. That in and of itself is enough to make their stuff completely irrelevant.