As long as I live, I'll remember waking up early AM, 11-9-2016. Trump was president elect & my wife was in the bathroom, crying. Part of what I felt was betrayal by media & pollsters I trusted. Incredibly, there’s a chance that it’s happening again. Please hear me out.
Much has been made of Biden polling at roughly the same national lead (+10.5 today on 538) as Hillary at this point in the election cycle. And while we all know what happened with Hillary’s lead, we’re supposed to feel safer this time for two reasons.
One, unlike Hillary, Biden’s lead has been consistently large, whereas Hillary’s evaporated at multiple points in the 2-3 months approaching election day. You can see that here: https://projects.fivethirtyeight.com/2016-election-forecast/national-polls/
And two, unlike Hillary, who never topped 46% voter support in 538’s rolling averages, Biden has crossed the 50% support threshold. He’s basically stayed there since mid-June. There are far fewer undecided voters.
Setting aside (for a moment) that the president is selected by the electoral college, and not a national vote, my biggest concern in this context is the accuracy of the polls. In 2016, pollsters were horribly, shockingly wrong in swing states, particularly PA, WI and MI.
(To be honest, any media outlet writing a story about 2020 polls where Pennsylvania, Michigan and Wisconsin aren’t the headline and the first ten paragraphs of the story is doing voters a grave disservice.)
I took time getting to this point because I really want you to remember how WRONG almost ALL the pollsters were. What evidence do we have that they have fixed anything? As the great philosopher Dubya once said, “Fool me once, shame on you. Fool me — you can’t get fooled again.”
And we really can’t, so here’s the buried lede: Trafalgar has Trump up: .6 in MI; 2.3 in FL; +4 in AZ; 3.7 in OH. They don’t have a recent (i.e. post-Trump Covid & subsequent hysteria) WI poll. And they have Biden up 2.3 in PA. What does this mean for the electoral map?
It’s scary as hell. If Biden loses MI, FL & AZ as Trafalgar, the top swing-state pollster in 2016 expects, his path to an electoral college victory is extremely dicey.
He’d need to win a state like NC, where he is currently +2.7 on the RCP average. Great, except—2.7 was the EXACT amount the RCP average underestimated Trump in 2016.
Infuriating side note: much like in 2016, the difference between a Biden victory or loss in MI may be covered by the number of people voting for @HowieHawkins
of the @GreenPartyUS & @Jorgensen4POTUS; Trafalgar has them at 1.5 & 2.5—well more than Trump’s lead.
So: Biden may yet pull this out. But if you’re seeing endless headlines about Biden’s super fundraising & huge polling lead, & understandably thinking that this race is in the bag, well, I just wanted to draw your attention to some facts.
This election may basically be tied, or maybe leaning Trump. We may literally, unfathomably, be heading for an exact repeat as 2016 in terms of polling dynamics, and the expectations they create. You truly mustn’t be fooled again.
PS: None of this factors in the pernicious voter suppression, misinformation campaigns, mail fraud & other forms of outright cheating that all evidence seems to show the Trump campaign currently utilizing.
PPS: Please feel free to: encourage @Jorgensen4POTUS & @HowieHawkins to drop out; send personal emails or call anyone & everyone you know, specially in a swing state, and beg them to vote for Biden; contact the Biden campaign and see if they need help. https://joebiden.com/take-action/ 
You can follow @BillWerde.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: