538 rating drops Biden to 69% chance

[jorgevlorgan]

Even though he's leading nationally by 8.3%..

[BannonsLiver]

1. Clickety clicks for the guy who never actually says anything

It’s a business, after all.

[SCliberal91294]

2. That is only because of the amount of days remaining

If Biden still has an 8-10% lead with 20 days to go, the percentage will be much higher.

[peggysue2]

3. And the Economist gives Biden a 90% chance

https://projects.economist.com/us-2020-forecast/president?gclsrc=aw.ds&gclid=EAIaIQobChMIqK-22sG56wIVEwCGCh15sw3-EAAYASAAEgLPdfD_BwE&gclsrc=aw.ds

Let's not dwell on the polls. Unless, of course, they make everyone more resolute to GOTV. Because in the end, that's the only thing that matters.

[ananda]

16. 90% is quite likely.

It will be a landslide.

[peggysue2]

32. I agree!

A this moment in time, Biden is on his way to a major win, even landslide territory. Charlie Cook agrees and he's hardly someone known for outlandish predictions. We're entering the final stretch and lots of things can happen. But one thing remains constant:

Trump is unfit for the job and the signs are all around us. The economy is teetering despite the spikes in the stock market. We have joblessness, eviction threats looming and record bankruptcies on the horizon. Not to mention the ongoing Covid-19 disaster which is something Trump and his enablers believe they can hide. Then there's the deliberate upheaval in postal services, something many, many Americans rely on.

No time to get complacent but also no time to turn into nervous nellies.

We're going to win this thing. Even Don Jr. thinks so.

[a kennedy]

40. Yes, yes, I WANT A CRUSHING, A CRUSHING, no doubt about it. No possibility

of doubt, a CRUSHING.

[brooklynite]

4. Because he's not...

He's leading in ONE poll by 8.3%. The model takes into account multiple polls and changes whenever new data becomes available.

Also, remember that a national poll says nothing about State by State results.

[jorgevlorgan]

20. 538 average is actually 8.3

Some have him at 10 others at 4, others at 8.

[Javaman]

5. By the end of next week, he'll be back up

So much time for the orange asshole to continue to fuck things up.

I expect tRumps response to Laura will be worse than morons*

[still_one]

6. Forget the polls, just get out the vote, whether through call bankin or knocking on doors

[Initech]

7. After 2016, I will never in a million years trust anything the polls say.

[budkin]

10. National polls were pretty much on the money

There were a few state polls that were wrong, and we know why...

[LincolnRossiter]

8. I think Biden as a 2:1 favorite under current conditions is about right.

Presumably you have some data that you want to share that suggests to you that this isn’t right....

I’ll give you a quick analytical head start, though. Silver’s overall projection is very heavily (almost entirely) weighted toward the final results in the electoral college vote—not the popular vote spread. Which sort of makes sense when you consider that that’s how we elect the president.

[qazplm135]

9. I'd buy there's a 3 in 10 chance Trump wins

This was always going to tighten, and Biden isn't winning by high single digits.

I think he wins of course, but I don't think it's the blowout people think or want.

A lot of those white women WANT to vote for him, he just keeps making it hard for them.

[Awsi Dooger]

12. Your first sentence is perfectly accurate

"This was always going to tighten, and Biden isn't winning by high single digits."

I don't believe in high margins during such a polarized era. I realize that is not popular here, where logical parameters are ignored and everyone prefers cheerleader mode. But cheerleader mode is rubbish, since it lends to the nonsensical 60-62% Biden share of the vote.

If someone in February had offered 69% Biden edge at this point I would have been thrilled/shocked. Every situational advantage is with the incumbent. Basically the challenger cannot win. The incumbent has to lose it. Tragically it required coronavirus for the incumbent to be losing it. But that's where we are and now we need to hold on.

To win by big margin we needed a young charismatic "vote for" candidate. But the field was hardly ideal. I hoped Beto would emerge since he seemingly had greater potential toward that vital blueprint than anyone else. Once he fizzled out we were always going to be reliant on "vote against," which is hardly ideal and carries lower upside.

[SCliberal91294]

13. "Tightening"

I'm not sure about this, the higher quality polls do not reflect this and there are very few undecided voters. I really don't see how anybody would change their mind from Biden to Loser Donald.

[qazplm135]

15. I think they will, we will see

If unemployment drops to say 6-7 points by election day, not guaranteed but not impossible either, then some of his potential voters will see this as a recovery and credit him.

I trust polling overall, but I don't believe that just because someone tells someone in July that they are "locked" in to voting for Biden that means they can't possibly change their mind by election day.

People are fickle, they don't pay attention, and they can be swayed.

Again, don't think Trump wins. I don't think he can attract many NEW voters, and I think Biden will definitely attract new voters, people who sat out, people who thought Trump wasn't going to be that bad, people who thought the Dem had it in the bag, etc.

But I do think we are looking at a sub 5 percent win. Now, that doesn't mean it isn't an EC blowout. It wouldn't have taken much for Hillary to win in the mid 350s.

[SCliberal91294]

21. Unemployment Rate

Is projected to go back up, or stay around 9-10% in most projections. But yeah, people are fickle. I just want to win by as many as possible for the senate etc. 5% won't cut it, we need 7-8%. The challenger does better with late deciders, so this is possible.

[qazplm135]

24. the one thing Trump seems to have is insane levels of good luck

I'd actually just expect the economy to do better than expected the rest of the way.

[SCliberal91294]

27. Here are the projections, please click on forecast

https://tradingeconomics.com/united-states/unemployment-rate

[qazplm135]

31. projections are projections

they aren't reality til they are.

And look at that wide band in that projection, it's anywhere from 5 to 15 percent.

[Demsrule86]

42. A million people losing jobs every week...so the economy sucks and will only worsen.

[jorgevlorgan]

22. That's the same as 2016

Not comforting.

[qazplm135]

26. it shouldn't be

if you want comfort, Trump's America is not where to look for it.

This is going to be a decades long battle. Trump was just a part of it. It's not over when he's gone.

Half the voting population is crazy, ignorant or easily conned.

[jorgevlorgan]

28. Yep

...and that sucks

[qazplm135]

30. it probably sucks for the either side too

that they cannot maintain firm control for very long.

So which side fights harder will ultimately determine things.

[budkin]

11. Looks like it's the battleground polls that have tightened which caused this result

Trump now leads in NC and OH which Biden had been leading most of the summer.

[Awsi Dooger]

23. North Carolina had 43% conservatives in 2016 exit poll

I have no idea how we can ignore a number like that. As I've emphasized, the Democratic nominee has not won a state above 37% conservatives since Bill Clinton in 1996.

Granted, these things can shift a few percent, but these are the numbers from 2016. I'll include only the states we are chasing or that are somewhere close to that vital 37% tipping point:

* Arizona 41% conservatives 27% liberals
* Colorado 35% conservatives 28% liberals
* Florida 36% conservatives 25% liberals
* Georgia 42% conservatives 22% liberals
* Indiana 38% conservatives 22% liberals
* Iowa 40% conservatives 23% liberals
* Michigan 36% conservatives 27% liberals
* Minnesota 32% conservatives 28% liberals
* Nevada 36% conservatives 25% liberals
* New Hampshire 31% conservatives 27% liberals
* New Mexico 34% conservatives 29% liberals
* North Carolina 43% conservatives 22% liberals
* Ohio 39% conservatives 20% liberals
* Pennsylvania 33% conservatives 27% liberals
* Texas 44% conservatives 20% liberals
* Virginia 33% conservatives 26% liberals
* Wisconsin 34% conservatives 25% liberals

As you can see, we are taking states for granted that are not far below 37, while overly optimistic in states that are 40 and above. The 538 model is going to understand the historical nature of states like Texas, Georgia, North Carolina, Arizona, etc. Other models are foolishly taking 2020 polling at face value while 538 does not. We have far less opportunity in those states than polling or current conventional wisdom suggests. That's why 538 allows a lower number toward Biden.

Florida is the gold mine. Florida changes the entire national dynamic if we pick it off. Nothing special needs to happen in Florida, other than competent messaging and emphasis. Florida is right there to be had at 36. The others are long shots disguised as 50/50.

I am unsure about Arizona. Typically it votes right of Florida and I would expect that again due to the 41% conservatives. But if any other state can potentially join our camp it is Arizona and not the silly flails in those 42 and above states. If Biden wins the 42+ states then the national margin is indeed 8+. But there is no logical reason to expect that margin to hold.

[qazplm135]

29. Yet Obama won it

and Biden is tied there in every single poll.

Add 5 points to liberals, subtract 5 points from conservatives, and now you have 38-29, which aint that far from Colorado which is solid blue. Plus, where are the 35 percent independents going? How many more liberals are coming out? How many conservatives are sitting home, voting I or voting Biden?

This is just too simple of an analysis you are doing.

[jorgevlorgan]

37. Actually Biden is leading there in the average of polls

But its important still to recognize it is still a more conservative state.


We really need to get tied of the Electoral College.

[qazplm135]

38. leading within the margin of error

is tied to me.

But it's not really a "more conservative state."

The Gov is a Dem.
A Dem is leading in the Senate race rather significantly.
A Dem is tied/leading within the MOE in the presidential race.
A Dem won there in 2008 presidential election and only lost by a few percentage points in 2012.
Heck, Hillary lost by less than 4 points.


That sounds awfully purple to me.

[Celerity]

41. +1000

I agree, it is reddish purple trending to pure purple, but also SO corrupt (the Rethugs there)

[Celerity]

14. Do these models take into account 2020-specific exogenous shocks?

1. COVID-19's driving of a massive number of voters to vote by mail

2. Trump and the Rethugs trying to destroy the mail-in ballot delivery system

3. Trump and the Rethugs trying to discredit mail-in voting

4.The largest ever, most sophisticated ever, systemic, multivariate voter suppression scheme (ie roll purging, caging, reduction of polling places, voter physical intimidation via a nationwide army of white power-fuelled goon squads, outright destruction of mail-in ballots by a multiplicity of tactics, etc etc etc.) And ALL that erected upon a decades old superstructure of illegal gerrymandering, which affects not only the US House, but also has given power to ratfucker Rethugs at state legislative levels to turbocharge the other forms of suppression.

[Doodley]

17. Too many are happy to accept 1000 dying a day as the new normal. They think Trump is strong on the

economy and they don't give Biden credit for anything.

[PTWB]

18. A big lead nationally means nothing

2016 came down to mere tenths of a percent in a few swing state.

Swing states are tightening, again, unfortunately. We have to GOTV.

[uponit7771]

25. +1

[Drunken Irishman]

34. A big lead absolutely means something.

Had Hillary won the popular vote by what Obama did in 2012, she's president.

[PTWB]

39. Biden could win the popular vote by 20m and still lose the election

If he wins that 20m in the wrong places. That’s the nightmare of the electoral college.

A big lead nationally doesn’t mean anything when the election comes down to a half dozen swing states.

Again, getting out the vote in those states is what will be effective. That’s how we win.

[BGBD]

19. It's just an artifact of the model

Today is long enough after the convention that the model is pricing in a polling bump when giving odds.

This year is weird and has few undecideds, meaning there isn't going to be a real convention bump for either candidate in all likelihood.

So, the model is seeing Biden +8.3 average and reading it in as something more like Biden + 7.0 or so. A week from now the model is going to be expecting a bump for Trump for the same reason. If the polls remain stable you will see Bidens odds run back up.

[octoberlib]

33. We need to start seeing more of Joe and Kamala . Constant TV appearances and such.

And they need to figure out how to connect with people in the swing states if they can't do rallies.

[Squinch]

35. Good. We need a reminder that this thing is far from over. GOTV and call your

reps and demand a detailed account about how they are safeguarding the vote.

[RhodeIslandOne]

36. Preparing you for the fix.