Oct 16, 2020

Rachel Bitecofer's final election prediction.

Rachel Bitecofer's final election prediction.

She explains how her model has changed — slightly.

I’m Isaac Saul, and this is Tangle: an independent, ad-free, subscriber-supported politics newsletter that covers the best arguments from both sides. If you found this online, you should sign up for the newsletter below:

Subscribe now


Today’s read: 20 minutes.

This is the longest Tangle’s ever, but it’s worth it. In today’s edition, I’m talking to Rachel Bitecofer again. She’s the pollster I interviewed last winter who — at the time — had already released her election predictions for the presidential, Senate and House races.

Bitecofer (right) on CNN in February discussing her election models.

When I first interviewed Rachel Bitecofer in February of 2020, the country was a vastly different place than it is right now.

Most notably, COVID-19 was hardly on our radar and we did not yet have any idea who the Democratic candidates for president and vice president would be. I published my first interview with Bitecofer on February 12th, and it wasn’t until February 27th that I wrote the Tangle edition “Coronavirus: Is it time to worry?”

Even then, though, Bitecofer was bullish on something nobody else was: she believed Donald Trump was going to have a very hard time getting reelected. Back then, most pollsters thought the Trump economy and record-low unemployment would carry him to victory in swing states. But Bitecofer and her model are different. The distinguishing principles of her forecasting are that she bases it on “negative partisanship” and demographic realignment. It’s a new framework for predicting election outcomes, one that focuses far less on persuadable voters and far more on party loyalty and negative feelings people have about the opposition.

In 2016, when Trump was elected, Bitecofer was still a professor and had not yet become the election modeling expert many consider her to be today. But her rise to notoriety came after her model predicted the size of the 2018 blue wave almost perfectly months before election day. That blue wave, Bitecofer told everyone, was going to come because of the negative partisanship data her model included.

Unlike other pollsters, she doesn’t rely heavily on horse-race polling in battleground states, or like the way polls spit out different results week over week. Instead, she uses polling to garner incumbent approval ratings, but that’s just about it. Back in February, Bitecofer’s model was showing extremely strong negative partisan feelings towards Trump and changes in voter demographics. This, she argued as early as July of 2019, was going to sink him. And take the Republican-controlled Senate down, too.

That’s why, even after the COVID-19 pandemic and the outcome of the Democratic primaries, and basically every other major news story in the last nine months, Bitecofer’s model has barely budged. Instead, her bullishness on Democrats winning the presidency, retaking the Senate and picking up the House has grown — but only modestly.

As she says on her website, the “difference between the pre and post-pandemic environment, both due to the severity of the pandemic from the mismanagement of the response and due to the economic damage from it, is that Trump is also suffering greater loss of support from ‘pure’ Independents than he might have otherwise — the persuasion band is larger than it might have been. As it turns out, even in the age of polarization, voters become more persuadable when your greed and gross incompetence is literally killing them.”

Which brings me to one other thing about Bitecofer that makes her unique in the polling world: she’s not shy about her politics. Bitecofer is a liberal, a passionate and outspoken liberal who loathes what the president stands for and what she views as an open attempt to subvert democracy as we know it. She’s currently working as an advisor to The Lincoln Project, a group of never-Trump Republicans who are raising money and running ads against the president. As a result, she’s an entertaining pollster to talk to — unlike any other I’ve interviewed. She’s a smoothie of data and personal opinion, cussing and laughs, all wrapped up into the fact that she’s a woman in a male-dominated field with a chip on her shoulder trying to prove the longstanding norms on election modeling are broken — and that her model is the future.

With just a couple of weeks to go until election day, I wanted to follow up with Bitecofer and see where her model stands now. Last week, she released her election update, which shows Joe Biden with an 81.1% chance of winning the election and Democrats with a 73.9% chance of taking over the Senate. So on Monday, I called her up and asked her why. We rehashed some things from 2016, and I pressed her on whether her own personal politics might impact her modeling. We also touch on the potential for Florida results to end the election on election night, rather than the election ending after days of counting mail-in votes across swing states.

This interview has been edited for length and clarity. It also contains some explicit language and annotations for additional context that are marked with asterisks.


Tangle: To sort of press rewind for a sec, the last time we spoke was early February. So much has happened since then. We now know the Democratic nominee, the vice presidential nominee, we’ve been through a few million “bombshell stories,” and COVID-19.

Back then, you told me that you expected Donald Trump to lose in November, and your main reason was that the Democratic base was activated. Turnout was increasing about 10 points in all these election cycles since Trump got elected — and you said all the stories to come aside, the fundamentals of that reality won’t change. The exact quote I have was, “Unless Donald Trump can find more white working-class voters to offset it or fracture the Democratic coalition, unless he can come up with a way to win without being a majority vote winner, it really is going to be very difficult for him to re-capture the presidency.” So… I guess the first question is, has your forecast changed at all?

Bitecofer: No. I mean, keep in mind we've gone through a once in a century global pandemic. Right? And you know he [Trump] has mismanaged it terribly. This is an objective fact, right? It becomes subjective once it gets into the partisan environment, but this is a global phenomenon. Other countries are going through it -- it's not American centric, and you can thus measure America's pandemic against its fellow democracies and countries and see how we are faring, and we are faring objectively terribly. Right? And yet, there's been no real democratic price for him. There's been a mild price, and we'll talk about that specifically. But not the total of what we would see if we were not a country whose body politic is deeply sick with hyperpartisanship and polarization.

What we should be seeing is a massive and total rejection of him. He should be polling like George W. Bush right after the financial crisis hit. When he [Bush] was finishing his term, he got down into the low 30s on approval and that was because finally, some Republicans started saying, “okay this guy sucks, alright?” [laughs]. As long as you have 90% approval locked in amongst all Republicans, you are always going to be held up because there are independent leaning Republicans who behave exactly like Republican partisans. And between those two groups, you're going to be at 40 percent-ish. So Trump is always going to be around 40 percent-ish because those groups refuse to acknowledge reality about his performance.

With the Bush administration, that was kind of true going through eight years of his presidency. And then in his second term, the Iraq War reality started to settle in. But once the economic collapse came in and really got into the last couple of months, suddenly his approval started to drop below 40.* And when you look at why that happened, it happened because his approval amongst Republicans at that point went down into the 70’s. Even into the high 60’s. With Trump, that has never happened. He's always been at 90% or above, and that's what's keeping him competitive. He's set to get shellacked, but he's still not Jimmy Carter in 1980 shellacked.** I mean, that's what we should be seeing in this election, if we had a healthy democracy, and a healthy public, because it's the citizens that are sick.


*In the wake of 9/11, according to Gallup, George W. Bush’s approval rating got as high as 90%. Before he left office, it had dipped as low as 25%. Trump’s approval rating is currently about 46%, according to Gallup, and the lowest it ever got was about 35% right before his inauguration. He remains above 90% amongst registered Republican voters.
**In 1980, Ronald Reagan defeated Jimmy Carter in one of the most lopsided victories in presidential election history, by a 489-49 electoral college margin. Carter won only Georgia (his home state), Maryland, Minnesota, Hawaii, West Virginia, Rhode Island, and the District of Columbia.

So that said, when I talked to you in February, we had one effect. We had this negative partisanship effect that is the basis of my theory and the original forecast that said — even then — when we get to the fall of 2020 we'll be seeing the Democratic nominee, whoever it is assuming it’s not Bernie Sanders,* in a very dominant position. And that the midwest would be obviously out of reach for the Republicans and that's exactly what we see right now. We see the midwest out of reach for the Republicans. I wrote this in the forecast and you can go back and look at it — it was a profound misunderstanding of how Trump won the midwest in the first place. It wasn't winning over this big revolution of white working-class voters. It was enough voters not showing up that could have voted, it was people voting third party ballots and Black turnout decreased 5% across the board in 2016 from 2012.


*Bitecofer said throughout the Democratic primary that if Bernie Sanders won, the general election would be far closer than expected — unless he picked a moderate running mate.

Of course, in my theory, I also talk about a certain segment of independents. Because most independents are leaners, so leaners are basically closet partisans, they are not persuadable. I can show you in the data that independent-leaners are almost no different in their vote propensity for Trump than people who are admitted Republicans. My theory has always argued that the pure independents were going to break against Trump this cycle.

In 2016, Trump had the advantage of being the outsider coming against an incumbent power that has been in power for eight years. He was the anti-status quo candidate. Pure independents are generally not highly informed, highly engaged voters. And they generally don't like the party in power. And so my theory at the time was that they would break in favor of Biden. Now, keep in mind, we're talking about 10% of voters in the election. Not the 33% you see on the surveys, because once you remove the leaners you get down to the 10 or 10-ish percent of the electorate falling into the pure independent category.

And then between these two parties, you have to ask is there a breaking in favor of one party or the other. Usually, we're talking 45-55. This year we have this second effect, I'm calling it the pandemic effect. Instead of it being 45-55, these pure independents are more like 60-40 in favor of Biden, and so when we look at why Biden's advantages have extended in the last couple of months, it's that extended growth amongst that group of voters and some of them are seniors.

So my sole update [the one released this week] talks about two effects, right? And I think that’s why we're seeing, number one, a bigger advantage in the midwest than I expected, though I certainly did expect an obvious advantage for Democrats in the midwest to close out. But also a deeper map. We're really seeing now potential for Texas to flip. And how we know that that's real potential is the Biden campaign spent $6 million there.* The Lincoln Project is spending millions there now too, right? And they wouldn't be doing that if they didn't really feel that there is some potentiality for that state to be competitive. The state has really helped out by the way, by the fact that it's had a ton of down-ballot activity in the state legislature. The state legislature is very likely going to flip control, they're nine seats away, and almost half the competitive House seats are in Texas. So, anyway, a deeper map. And it's because of these two effects working together.


*The Joe Biden campaign has been surprisingly aggressive in Texas, despite the fact many Democrats didn’t think they’d have a shot there this year. Typically, spending money like this is indicative that the campaign as private polling or non-public data showing them they have an opportunity to win.

Tangle: That kind of leads me into my next question. The model on the website you have, it looks like there are really four serious toss-ups that you have, which are Iowa, Georgia, North Carolina and Texas, and the others have a lean. So I took the liberty of just turning those leans red and blue, and it came out to about a 320 to 143 electoral college lead for Biden over Trump, with those four states remaining as toss-ups. So, I guess I'm just curious from your perspective, a) are we really looking at a landslide of that proportion? And b) if so, it seems like you guys feel pretty strongly that the Democratic Senate majority is in play right now, right?

Bitecofer: Oh yes. Basically my view for 2020 in Congress is not only are Democrats going to hold onto their 40-seat gain in the House, they are going to be on offense. And here is where that offense is going to be playing now. And in terms of the Senate, I was arguing that Colorado and Arizona were basically done deals.* When my first forecast first released I called Arizona, Colorado and Maine as likely Democratic pickups. But they need to net four though, not three, because they are probably going to lose Doug Jones' seat down in Alabama.**


*Colorado and Arizona are now widely considered out of reach for the Trump campaign. In Arizona, the Republican Senate incumbent Martha McSally has started distancing herself from the Trump campaign. A year ago, that was not something many pollsters were sure of. Bitecofer was.
**Republicans currently have a 53-45 seat advantage in the Senate, with two independents who typically vote with Democrats. That amounts to a 53-47 advantage. If Democrats lose Doug Jones’ seat, which they likely will, that puts it at 54-46 Republican advantage, meaning Democrats need to net four other seats to force a Senate tie, which — if Biden were president — would give them the tie breaking vote from Vice President Harris and thus, a Senate majority.

Trump is going to win Alabama, so you have to imagine a scenario in which Alabama people voting for Trump are going to crossover and vote for Doug Jones. I think that’s more likely to happen in South Carolina, because Trump is going to win South Carolina but I think it’s more likely that there’s going to be somewhat leaning Republican independents who will vote against Lindsey Graham. And most of Jaime Harrison's hopes and dreams come from the potential of a huge Black voter turnout surge.

If he didn’t have that, like Amy McGrath doesn't have that [McGrath is the Democratic challenger to Senate Majority Leader Mitch McConnell in Kentucky], that’s why Harrison is in much better shape than she is. His pathway doesn't have to rely purely on Trump-Harrison voters. And so Doug Jones needs Trump-Jones voters and a lot of them. And he's got more Black voters than McGrath, so he's in a better spot and him and Harrison are kind of in equal positions in that regard, but what Harrison has that Jones doesn't have is an enemy. Harrison is running against Lindsey Graham, and Lindsey Graham is maligned.*


*Democratic challenger Jaime Harrison was performing very well in polls against Sen. Graham until this week, when a new poll showed Graham with a 48-42 edge. Graham is one of the most disliked Republicans by the left, and a historic amount of money has poured into Harrison’s campaign from across the U.S.

On the other hand, Doug Jones is running against an ex-Alabama football coach [laughs]. So that fourth seat to me has always been North Carolina, and the sex scandal down there notwithstanding.* Go figure, when you go with a mediocre white guy, what do you get every time? It's a sex scandal! [laughs] But here's the thing, with the COVID-19 screw-up and Trump being such a circus freak — he has now caused a tightly married COVID-19 outbreak to the White House — it's a freak show! It has caused that persuasion band amongst those pure independents to be so deep, so much bigger than what it would have been previously, that the Senate map — you've got Iowa, you've got the two Georgia senate seats, you've got Montana and you've got Kansas. So the GOP has basically five or six pathways they have to defend to keep that 4th seat, you know what I mean? And that's a lot of holes in a dike that you have to fill to keep that majority. Do you see what I'm saying?


*Cal Cunningham, the Democratic challenger to Thom Tillis, had his campaign thrown into a frenzy after an apparent mistress leaked romantic text messages from Cunningham to the press. Cunningham is married with children. While his approval numbers plummeted in the wake of the scandal, the percentage of voters who said they’d cast a ballot for him remained steady.

Tangle: Yeah, totally. Not a good position to be in.

Bitecofer: If I'm McConnell, I've gotta defend six different pathways to that fourth seat. All of them, every day, look increasingly harder to defend, you know? The advantage that the Republican party has is that Democrats are still not Republicans in terms of electioneering and messaging. If these fundamentals and circumstances that Democrats have belonged to Republicans, which of course they wouldn’t because they’re so thoroughly stupid only the modern GOP could create them — but let’s say that they did, the Republicans would exploit them so much better than the Democrats do that it would just be a bloodletting. Because Democrats are not going to exploit the political moment, it’s hard to understand just how much natural oomph [they’ll] get out of it.*


* The term “electioneering” sounds sinister but it’s really not — it’s just a reference to campaigning. In many political circles, it’s pretty accepted that Republicans are better and more relentless campaigners than Democrats — and typically do a better job energizing their base. Bitecofer seems to be referring to this reality here, saying Republicans would do a better job exploiting a president like Trump to energize their base if the roles were reversed. The famous quote from the show Newsroom sort of illustrates this: “If liberals are so fucking smart, how come they lose so goddamn always?”

Tangle: So I’m curious, I mean one of the things about you and the model that you run is that in a lot of ways it’s supposed to be unique from the FiveThirtyEight, Cook Political, and so on. And I'll cover that in this interview when I write this up, what the differences are between you and them, but something that struck me was that I looked at your model and then I looked at FiveThirtyEight’s model and the outcomes are almost identical. It's an 88% chance of a Biden win in yours and FiveThirtyEight has it at 86 in 100. So I'm curious, if the models are so different, why are they producing such similar results? What's happening there?

Bitecofer: Let me tell you this, number one, I'm putting out this stuff months and months and months ahead of time. I did it in 2018, and I was like “this stuff is going to seem really radical,” because the conventional wisdom on election Twitter — and I wrote this into the original forecast — the conventional wisdom at that time was having the conversation, asking whether or not Democrats can pick up 23 seats and flip the House. And at that time David Wasserman had put out an analysis in a piece that he was really, really passionately defending, arguing that Democrats would have to win the popular national vote in the House by at least 10 points to win 23 seats because of gerrymandering.*


*Worth noting here that Wasserman — who is one of the best known pollsters in America — and Bitecofer have a little bit of history. Bitecofer has repeatedly criticized Wasserman, Nate Silver and other well-known pollsters on the record for dismissing her work as unserious. And she’s got a point. In a not so subtle tweet in August of 2019, Wasserman wrote, “You know which forecasters I don't take seriously? People who tell you their secret ‘model’ can predict the outcome of the 2020 election (even down to the EC count), before we know trivial details like, say, the identity of the Dem nominee or the state of the economy.”

And so when I came out with my forecast, I said it’s not even a question as to whether Democrats are going to flip the House, that is a done deal, you can take it to the bank. The only question is how many more seats and my guesstimate is they're going to pick up like 42 seats.* So I said in the forecast this might seem radical today but my goal, what should happen, is in September or October of 2018 this forecast should be the mainstream forecast. So what you're referring to now is exactly the same thing. And it’s exactly what I want to see.


*Bitecofer almost nailed it: Democrats picked up 41 seats.

When I put out something in 2019 before the nominee, before the pandemic, before all these things happen, and I’m saying “hey, this is how shit is going to look in the fall of 2020 when Nate Silver's gold standard probability model drops,” I want his shit to look like my shit, you know? [Laughs]

The thing is, and a lot of people who have covered my work don’t get this nuance, what is unique about my work is the lead time, right? Because everyone gets there in the end. We’re all there.* What is unique about my work is the forecast. It’s the ability of it to say this is how shit is going to be five months from now or a year from now even before there is a nominee and it doesn’t even matter who the nominee is, right? That’s what differentiates my work from the other models.


*Again, Bitecofer has a point. Her model has barely changed, but many of the pollsters who criticize her work have seen their models move closer and closer to hers now — and today, their forecasts are nearly identical.

Tangle: So the last time that we spoke, I asked you explicitly after you showed me your model: how does Trump pull this out? And you gave me two scenarios: You said either there is a legit third party challenger, someone like Tulsi Gabbard, who fractures the party. Or, Democrats just totally blow the VP pick by not uniting the party. So there's no legit third party challenger that’s going to siphon off too many votes, and I know from reading your stuff that you are really high on Kamala Harris and think she was a great pick for VP to unite the party. So those two options are off the board. So, I’ll ask you again now: What's the flip side of this? How does Trump pull this out? What's an outcome or a scenario where your model might be wrong, Nate Silver might be wrong, etc.?

Bitecofer: That’s a great question. I cannot underscore enough — and you'll notice I’m doing it now in my public appearances and Twitter feed as much as I possibly can… in 2016, the data wasn’t wrong, the data was sending a very clear signal: A big chunk of the electorate dislikes both Trump and Clinton, and they are unusually unsettled about who to choose.

Normally, when you take the Democratic nominee and Republican nominee and you total up the people who are voting for them in a presidential election, you get less than 10% of people who say they are undecided. And it’s usually around 6% and it’s definitely around 6% in the last couple of weeks of the election cycle. In 2016, that was not happening. It was not happening a month out and not happening a week out. They’d total these numbers up and it’d still be well above 10% and sometimes up to 15% of the vote was missing.

So, what should have been happening all that time was the narrative, the expert narrative, the frame, on all the TV sets and all the election coverage, should have been “we are missing a big chunk of the two-party vote in these polls and we're only seeing 85, 86 percent of the vote. And it looks like the electorate is having a hard time deciding.” There's that pure independent that basically couldn’t decide because partisans already know what they are going to do and that includes independents that lean Democrat or Republican. I wasn't an expert then, I was just a college professor, but experts should have been saying we have no idea what’s going to happen. And that’s really important to understand what’s happening now.

Because yes, college-educated voters were slightly over-weighted in some state polls. But the fundamental problem was not the data, it was the interpretation of that data. Think about how much that would have changed, how people would have behaved if they thought about the election and instead of everyone saying Clinton has got this in the bag, we had the uncertainty explained to people and also accounted for. The reason why that's so important to understand now is one, that the small errors in polling were unweighted non-college educated voters — all that stuff has been corrected now.* In fact, if there is a bias in polling this year, it’s a bias toward Trump or toward Republicans because many polls are weighing the electorate very heavily in favor of non-college-educated.


*One narrative that emerged after 2016, that I have written about in Tangle, is that polls did not give proper weight to non-college educated white voters, who overwhelmingly support Trump. Whether that has actually been fixed this time around, as Bitecofer claims, is a topic of much debate.

Still, these polls are producing pretty sizable advantages for the Democrats. And number two, there can only be so much disconnect between analytics and outcome, so unless these fundamentals change, if we were to hold the election today, Joe Biden should win. He should win anywhere where the swing state polling margins are above the margin of error, and I’m very confident in the quality of the data this cycle. Generally speaking, there’s only so much room for the data and reality to disconnect.

In other words, for Trump to win an election through voting we need to see different fundamentals in the data. The way Trump can quote unquote pull it off is to do what he and Attorney General Barr and Ronna McDaniel — who is head of the RNC — have been very forthcoming about. She, through her actions directing the RNC’s legal strategy, which they have to file on the public docket, are attempting as much as possible to make it very difficult to cast a ballot and have that ballot counted.*

And their legal strategy, the Trump team will straight up tell you this, “the fight begins November 4th.” What they mean by that is they are going to attempt to keep ballots that are cast from being counted. And so can he retain the presidency via manipulation of the counting of ballots? That's not something that this model can measure. This model is considering factors that are based on voter sentiment and fundamentals and preference. And they assume that every vote that gets cast gets counted. I can't model for some other type of election.



Tangle: That sort of covers my next question which was going to be about whether this model accounts for that variability. One last thing before I let you go, speaking of things that have changed, I noticed that you are now working with The Lincoln Project and I'm just curious how you ended up working with them? What has your role been over there? And I think something my readers are probably going to be wondering is whether your own political biases could or may have any impact on how you're seeing the election or modeling this thing out?

Bitecofer: Right. So I am a senior advisor at The Lincoln Project, I was approached to join the senior advisory board I think it was probably not long after we spoke, so it must have been March. But the role sounds more glorious than it is. It is attending meetings, getting briefed on things they’re up to and providing feedback. I wish it was a more hands-on position, it’s not a paid position certainly, and it doesn't get me a ticket into the brain trust of the group where I’m getting to build a strategy or do ads or anything like that. That is unfortunately not what I'm doing. But I do talk about them and promote the work they do.

The Lincoln Project, of course, is comprised of eight principals who are all former stars of Republican electioneering and campaigns. Major, major players, people like Stuart Stevens who has been on five presidential campaigns, two all the way to the White House. They’re Republicans who found themselves exiled or self-exiling from a political party that basically had a civil war, and they ended up losing. What they are trying to do is save America from an element of the party that has taken over and is spearheaded by people who don’t believe in democratic norms or principles.

I mean, obviously, you don’t if you are willing to try and disqualify millions and millions of ballots to retain power in an election. My work with the Lincoln Project is about democratic norms, there’s no policy issues or partisan elements to that, which is how I see it. Given that I’m working with Republicans, I don’t think that there is much partisan nature to it, frankly.

But in terms of how it might be impacting my rankings, I’ll tell you this: I’m the only woman doing what I’m doing. When I came out into the field in 2018, it was a little hostile, but after my forecast proved to be correct in 2018, it became a lot hostile. I, more than any person in the game, have to be correct. So I don’t make any decisions or ratings or rankings based on any other considerations than being correct. Obviously, that is my goal, and the nice thing about election analysis and forecasting is that there is going to be a judgment day, right? So you have no choice but to be brutally honest about it.


Tangle: Yeah that’s a good way to put it. [Laughs]. Rachel, thanks so much for the time, it’s always fascinating to hear from you, and I’m really interested to see how this all plays out. Maybe we can do a post-election circle up and figure out how things did play out. Election night is going to be wild.

Bitecofer: God, I’m just hoping like hell. Florida is kind of the stopgap, frankly. Luckily, Republicans are the ones that invented vote by mail and they invented it in Florida and they’ve been using it for a very long time to actually very successfully hold onto statewide elections, even as their numbers [became] disadvantaged. There’s fewer Republicans in Florida than there are Democrats, but they always win these statewide elections, and that’s why Democrats are always like “how are Republicans beating us?” And I’m like, “dude, because they fucking turn out at 70% and you don’t, and they do it through mail voting in Florida.”

You know, it was once Democrats were like, “oh we should start using absentee balloting the way Republicans do,” that’s why Trump made it his mission to come after it. And he was already wanting to come after it before the pandemic, because — this dumbass — he thought he was going to win... remember the blue wave in 2018, I didn’t think to prep all my people that were following me that it might take a week before all these Congressional seats, especially out in California, will come in...

California leans so much on small “l” liberal voting systems and they have mail voting and a really expansive voting rights system, so I didn’t think to warn people that I’m going to be right but it’s going to be a week delay. So one of the reasons I didn’t get the big bump — other than [the] fact that Silver and those guys decided to bury me — was the fact I was right after a week. Well, Trump thinks it’s because people faked those ballots. He thinks he initially survived 2018 and then somehow all these magical ballots showed up. So he was already coming after voting by mail.

Long story short, Florida is the stopgap because they already have the infrastructure to handle the counting and they have codified it dude, where Barr can't fuck with it because remember in the Florida legal system they have to do the count a certain way, and they have it all certified a certain way so that you can’t just be [Republican Gov. Ron] DeSantis and say, “I’m going to go into work and help my party out.” It’s all codified in Florida legal code. It's nice! And if Trump loses Florida, then it doesn’t matter if it takes a week to count shit in Wisconsin.


Tangle: That did stick out to me, that you have that 71% chance of Florida going to Biden, which to me that seems like the election is over on election night if that happens.

Bitecofer: Yeah, well, we'll find out. I am tracking vote by mail returns, which is a noisy signal because of Trump. When you have an elite signal, like the president says “hey maybe bleach can kill the coronavirus if somehow we can get it in the body.” You know, he didn’t say inject bleach, which is the Democratic way to say it, but he did muse out loud in a presser that bleach really affected coronavirus and it’d be nice if we could find a way to get bleach into the body. He had a conversation with himself out loud about it.

Well, people are so stupid, not just Americans, but humans are so stupid, and when the President of the United States who has that much authority and legitimization says something like that, there are people who straight up will go and buy bleach. So imagine then instead, if you have that president, a man with that kind of authority telling his partisans, “hey voting by mail is evil, it’s fraudulent, don’t do it,” so what we are seeing is a huge shift at a mass level — it’s just something you never see — we're seeing it where Republicans are eschewing voting by mail because Trump has told them not to.

Now that is terrifying the hell out of the GOP because that’s how they've been winning all these elections and he's basically undercut their system. Now, they have to count on the fact that these guys are going to show up instead on election day. And the nice thing about doing vote by mail was that when the people didn’t turn in a ballot, they had a system to track them. So every couple of days they could knock on the door, call them up. If you can't get the people to vote through that system and you have to rely on election day turnout, which is what Democrats were doing for the longest time and why they were getting their asses kicked, you get one shot, and if they don't show up, you're fucked. And so that’s why the Republican party — the local chapters down in Florida — they’re really, really nervous. And what we’re seeing in the data is really not good for them.*

*On this, Bitecofer is absolutely right. In fact, Trump apparently realized his mistake, and in early August began saying publicly that the Florida vote by mail system was safe and secure — a huge reversal after a summer of discrediting the practice. His change of tone was widely attributed to the fact Florida officials were infuriated that he was undercutting one of their major advantages in a crucial swing state.

Tangle: Yeah, wow. Interesting to think about. Well, Rachel, I'm looking forward to publishing this. Just like last time it'll be a transcription of our convo. I’m really looking forward to getting this out.

Bitecofer: Cool, cool, cool. It was nice talking to you. Good luck to you. And good luck to the country, cause… you know. I really mean it, good luck dude!

Tangle: [Laughs]. I think we’ll be okay. Thank you, I appreciate it.


Want more?

If you enjoyed this interview, consider subscribing to Tangle. The newsletter comes out once a day, Monday-Friday, around noon Eastern. Right to your inbox.

Subscribe now

Subscribe to Tangle

Join 120,000+ people getting Tangle directly to their inbox!