Part I: Democrats' risk-aversion helped to re-elect Trump

Part 1 of a 3-part autopsy on the Harris campaign. Plus, a GiveDirectly fundraiser!

Silver Bulletin is part of a group of Substacks proudly participating in a fundraiser for GiveDirectly, an organization that puts money directly into the hands of the world’s poorest people. I’m not much of a joiner or a “group project” guy, but I’m convinced that this is one of the most effective means of charitable giving. You can read much more about GiveDirectly in this post from friend-of-the-newsletter Matt Yglesias at Slow Boring. The goal is to raise $350,000 over the next 28 days, and GiveDirectly will match contributions up to $400,000.

In addition, I’m putting my money where my mouth is. I’ll donate half of gross annualized revenues generated by this post to GiveDirectly, as well as half of the revenues generated by the second post in this series¹, which is likely to run on Thursday. (I’ll tally up the numbers on Monday morning, 12/9, and indicate the amount in a Substack Notes post.) So if you’ve been on the fence about becoming a paid subscriber to Silver Bulletin, now is a good time: you’ll both help people in need and lock in access to the second and third parts of this series.

^^^ This part of the program is now over as of noon Eastern on Tuesday, but holy crap, it raised an additional $27,196.50 in addition to those of you who gave directly! Wasn’t expecting that and thank you so much!

---

A 3-part autopsy on Kamala Harris’s campaign

Well, I guess I just can’t quit 2024.

I thought I was almost done with the recaps and reviews. But then I saw, last week, the Pod Save America interview with several ex-Harris senior staffers. The staffers revealed a lot about their thought process. A thought process that, in many ways, was flawed, and reflects deeper problems in the Democratic ecosystem — and contributed to a series of decisions that helped re-elect Donald Trump.

I repeatedly went over the Pod Save America transcript over the long Thanksgiving weekend. I think of as a perfectly preserved, in situ fossilized record of how Democratic elites were behaving when the meteor (Trump 2.0) hit — although unlike the dinosaurs with the meteor (or the Hillary Clinton campaign in 2016) the staffers weren’t surprised by Trump’s win and were even expecting it.

So I’m just literally going to go through the whole transcript and annotate it, in three parts:

• Part I: Democrats’ risk-aversion helped to re-elect Trump

• Part II: The failed rebrand of Kamala Harris

• Part III: Nice guys finish last

The first part, today, will run free. Parts II and III will be mostly paywalled and I’ll add links here once they’re ready.

If you haven’t read it yet, I’d also recommend Monday’s post (think of it as Part 0 of this series²) on the shortcomings of what I call the Village — the term from my book for essentially the East Coast progressive establishment: the mainstream media, the academy, the mainline of the Democratic Party and its affiliated “groups” and advocacy organizations. I argued on Monday that these shortcomings are both institutional (the “expert class” hasn’t performed well in recent crises) and political (Democrats just lost an election to Trump). The Harris campaign, and Joe Biden’s campaign before it, are about as Village-y as it gets. So while Monday’s post was pretty abstract, this series can serve as something of a case study on how the Village performed in practice.

The bad habits of ‘the Village’ came to roost in the Harris campaign

Could Harris have won if she’d run a better campaign?

This is not the same question I asked in an earlier newsletter, which is whether Democrats could have won with a different candidate who had more distance from Biden. My answer to that was a tentative “yes”, mainly given Democratic outperformance in Senate races relative to the presidency. In the end, they did not lose to Donald Trump by all that much: Harris lost Pennsylvania by 1.7 percentage points, Michigan by 1.4 and Wisconsin by 0.9; those states would have sufficed to give her 270 electoral votes. There’s probably been an overcorrection toward deterministic explanations for electoral outcomes. Candidate quality may not matter all that much, but I’m comfortable saying the difference between a mediocre candidate and a great one exceeds the margin in these states.

Locking Harris in as the nominee, could she have won with a better campaign? Meaning, one that employed different strategies and tactics, emphasized different themes, or deployed resources differently? My initial instinct was to say “no”, but that’s before I read the Pod Save America interview.

Because I think this was a pretty bad campaign. Not terrible: the Village, despite its other flaws, usually has a high floor for basic technocratic competence. The Harris campaign’s data and analytics work seems to have been good; they got off to a quick start under tough circumstances, and Harris was well-prepared and effective in her debate. But if Harris was a C-plus candidate, the campaign probably deserves a C-minus grade — and lower than that if you account for the fact that many of the people running Harris’s campaign were also running Biden’s.

Indeed, it’s hard to untangle the following threads:

1. The effectiveness of Harris’s campaign once Joe Biden dropped out on July 21 and she quickly became the consensus replacement;

2. Biden’s insistence on running again even as his problems became increasingly clear;

3. The manner in which he dropped out — endorsing Harris in a way that hadn’t been telegraphed in advance, the staffers claim in the interview;

4. And Biden choosing Harris in 2020, a decision that was also influenced by some of the staffers; Jen O’Malley Dillon, Harris’s campaign chair, also played that role for Biden in both 2020 and 2024 and is featured prominently in the New York Times article from August 2020 explaining the rationale for the Harris pick.

Because of this entanglement, I’m inclined to read their actions less sympathetically. Throughout the Pod Save America transcript, for instance, the staffers complain over and over again about how little time they had to work with. Well, guess what: as Ben Yelin points out, that’s largely on them. A full 24 days elapsed between Biden’s disastrous debate and his withdraw from the race. And the problems with Biden were evident far sooner than that: for instance, in skipping what is typically a softball Super Bowl interview. If you’ve been living under a rock and have never watched the We’re All Trying To Find The Guy Who Did This sketch, now is a good time.

In what ways did the campaign embody the bad habits of the Village? There are basically five themes that you’ll see reflected throughout the series:

• First, it was too risk-averse at many key moments, despite correctly understanding that Harris was at best in a toss-up race if not an underdog.

• Second, it suffered from the problem of too much agreeableness, a characteristic Village trait since the Village tends to emphasize communal consensus over individual achievement. There was, most obviously, far too much deference to Biden. But there was also an unwillingness to ruffle feathers with other progressive constituencies who might have gotten upset by a bolder approach.

• Third, there was a real lack of agency. The staffers portray themselves as subject to tidal forces outside of their control and tend to blame “the system” — even when some of those forces were literally of their own making. They are unable to conceive of alternative strategies even given the benefit of hindsight. They are not exactly describing themselves as victims — highly agreeable people usually avoid doing that quite so explicitly. Rather, they almost seem to view themselves as Non-Player Characters, executing a prescribed set of instructions that unfortunately happened to produce a losing hand.

• Fourth, there was a tendency toward decision-making by consensus or committee. They lacked “vision”, and their candidate did too. The metaphor I can't get out of my head for Harris’s brand is some hopelessly confused concept for a restaurant: say, one billing itself as a “Cajun-influenced, vegan-friendly, Tex-Mex cantina and speakeasy”. If I accused such a restaurant of lacking vision, the proprietor might say “What do you mean? We're a Cajun-influenced, vegan-friendly, Tex-Mex cantina and speakeasy! The vision couldn't be any clearer!”. And maybe in the hands of a genius chef, this concept could work. But there’s a 98 percent chance it's just going to be a mess. We’ll cover this extensively in Part II.

• Fifth, with the with the partial exception of deputy campaign manager Quentin Fulks — who shows more self-awareness than the other staffers — they pretty clearly live within a bubble. They don't seem to understand how Harris’s brand came across to the general public or what arguments were plausibly likely to persuade swing voters. Groups like young men were treated as alien lifeforms who could only be studied from afar through focus groups and internal polls.

I'm bringing receipts. For the psychographic profile of the Village, including specifically how Democratic campaigns are staffed, see Chapter 5 of the book. But most of what I'm criticizing here is consistent with what we wrote about here at Silver Bulletin all campaign long, so I'll be liberal about linking to earlier newsletters. And I’m going to go through literally the entire 90-minute transcript, save for the round of goodbyes at the end. This is already getting to be a long post, so let's dig in.

Harris’s staff thought she was probably going to lose

Dan Pfeiffer: Welcome to Pod Save America. I’m Dan Pfeiffer. We have a special show for you today. I am here in Washington, D. C., where I’m about to sit down with the leadership of the Harris-Walz campaign for their first interview about what happened in the election. Last week, Michael Tyler, who’s the communications director for the campaign, called me and said that they were ready to speak and that they wanted to have that conversation on Pod Save America. This is the first time that any of them have done an interview since the election. They don’t pretend to have all the answers here. There’s way more to cover than we could possibly cover in one podcast. This is the beginning of a conversation about understanding what happened in 2024 and learning the lessons that Democrats are going to need going forward. Here in Washington with me are Harris-Walz campaign manager Jen O’Malley Dillon, Quentin Fulks, who’s the deputy campaign manager, and oversaw paid advertising, Stephanie Cutter, who oversaw messaging and communications, and joining by Zoom is David Plouffe, who consulted on all of it.

Let's do some table setting and introduce the cast of characters. I don’t know any of these people — although I suppose I’ve had mostly friendly online interactions with Pfeiffer over the years — so what I’m providing you with here is pure “outside view.” Pfeiffer is the co-host of Pod Save America — not one of the Harris staffers. However, he filled multiple roles in Barack Obama’s campaign and administration.

Plouffe, another Obama-era guy, was brought in as a consultant after the candidate swap, as Cutter basically was, though she had previously been working with Harris. O’Malley Dillon was “inherited” from Biden, however: she was the campaign chair for Biden 2024, in fact, and then also played that role for Harris. She bears the most responsibility for the campaign's overall direction. Fulks also transitioned over from essentially the same role on Biden’s team.

By the way, I don’t think there’s anything to criticize from Pfeiffer here. The interview style is friendly, but podcast interviews usually are. He’s asking most of the right follow-up questions and you can often get a lot out of your guests just by sitting back and letting them talk in circles, especially if they’re feeling comfortable and have their guards down. But he’s not too friendly: Pfeiffer hasn’t been afraid to criticize decisions made by the Biden and Harris campaigns. He’s getting a lot of newsworthy insight about the campaign's thinking into the public record.

Pfeiffer: Jen, Quentin, Stephanie, David, thanks for joining us. Thank you for doing this. Very much appreciate you having this conversation with us here on Pod Save America. Just to level set, Jen, and I’ll start with you: How did you feel going into Election Day? And at what point did you have a sense that things were beginning to break Trump’s way? Was there a county result? Something about the turnout? Like, was there a moment when you sort of understood that– how it was going to end?

Jen O’Malley Dillon: Well, I, the truth is that we really thought this was a very close race. We talked about the entire time we saw it as a margin of error race, almost the entire time the vice president was in the race. And we knew we had to have strong turnout on Election Day. We saw early vote really, uh, ending strong for us and saw, you know, the types of voters we wanted to see turnout. But, you know, we expected this to be close. We also expected that Florida was going to come in a bit redder. Virginia we knew was tracking to being, you know, that we were going to be ahead, but that we would be ahead by less than we were in 2020. So we were expecting to see that when we saw that.

Let me start with a note of praise: the Harris campaign does seem to have had a pretty realistic conception of where they stood. This is not a case where the internal polling was unrealistically rosy. They stuck to campaigning almost exclusively in the seven key swing states that mattered— as Trump mostly did. They didn’t panic about Virginia or New Hampshire or reach into Florida or Iowa, despite public polling that occasionally might have tempted them to do that.

Of course, this cuts both ways. Given that they saw themselves in a toss-up race — or really, as we’ll see later, probably slightly behind — they should have adopted some more risk-on, high variance strategies and shown more of a propensity to “shake things up”: that’s one of my big critiques. But this isn’t a Hillary Clinton thing where they weren’t even campaigning in the right states. The data/analytics work was solid.

O’Malley Dillon: We also did anticipate that the night would go relatively long because some of these states would take longer to come in. But I think it was, you know, really after polls closed, there was nothing that we saw throughout the day. There was nothing that we saw that told us there was overwhelming turnout or anything out of complete expectations on Trump’s side. But it really took us into the hours of, you know, polls closing for us to know for sure that things were not tightening. They were tight, but they weren’t tightening in the direction we needed them to be.

This part is also honest, I suspect. Nobody really knows anything on Election Day until the results start coming in.

Dan Pfeiffer: And is that just because Trump’s turnout was so high?

Jen O’Malley Dillon: Well, honestly, I think, um, I think it’s a little bit mixed. I mean, we saw certainly Trump turnout high in early vote. We really believe that to be mode shifting and that’s what I think it was. I think we also saw turnout was as expected in rural areas, like we didn’t see anything that said, you know, like maybe we saw in ’16 or even in ’20 that he had more turnout than we had anticipated, and our analytics really was quite close, much closer than it had been in ’20 and in ’16. But I do think we saw some lighter turnout in some of the areas we had hoped, but difference of a point here or there, which obviously in a close race makes a huge difference. And then I think we saw a little bit of a drop in support in a few areas for us. So that ultimately, I think, is why we weren’t able to, to close the gap. It wasn’t so much that what we were seeing in the battlegrounds was out of expectation or that he had some hidden turnout we hadn’t picked up on.

So far, so good. O’Malley Dillon is right there’s not any one neat turnout story to explain why Harris lost when Biden won. Some of Biden’s voters did stay home: the recalled 2020 vote among the 2024 electorate was Biden +1, according to the exit poll, as compared to Biden’s +4.5 actual margin of victory. There were slightly more vote-switchers among Biden’s voters too, though: 7 percent of Biden 2020 voters flipped to Trump or a third party, versus 5 percent of 2020 Trump voters. And Trump won the plurality of voters who voted this year but didn’t vote in 2020 or who voted for a third party. Overall turnout is down, but only slightly, from 158.6 million votes in 2020 to 155 million in 2024 with some votes still left to count.

Dan Pfeiffer: David, when you say it was a margin of error race you needed high turnout, what did your polls tell you where the race was heading into election day?

David Plouffe: Well, Dan, you and I talked prior to the election, and just to rewind, I think when Kamala Harris became the nominee she was behind, we kind of, you know, climbed back. And even post debate, you know, we still had ourselves down, you know, in the battleground states, but very close. And so I think by the end it was a jump ball race, and I think we needed some things to break our way. Maybe Trump’s Election Day turnout would underperform. Our election day turnout would, you know either be at level or over perform and, you know, we’d win more of the people who decided in the last three or four days.

Plouffe, consistent with O'Malley Dillon, claims that the campaign had itself slightly down in its internal polls even after Harris’s strong debate. And I believe him, I guess. There were bearish signals all along from the campaign and affiliated groups about its internal polling. I was curious about this in real time. Did the campaign think it would raise more money if it portrayed itself as the underdog? Possibly some of that. Were the outside groups leaking negative news because they wanted a strategic shake-up? Maybe some of that too. Mostly, though — again, total “outside view” — I suspect these leaks were honest because the campaign wanted people to understand that they had a difficult assignment and they didn’t want to be blamed if Harris wound up losing.

There’s no evidence that the campaign closed strongly

Plouffe: I think our data and the New York Times data and other public data suggested we did have some progress with undecideds at late October.

This bit I’m more skeptical about, however. True, it tracked with what the campaign was saying in real time: they started making some more optimistic noises about their internal polls toward the very end of the race. But that might have been an exercise in morale-boosting. The AP VoteCast exit poll found the vote was split about evenly among people who decided in the last few days. The network exit poll, meanwhile, found that Trump won voters who decided in the last week 51-42, tipping a pure toss-up race into one where Trump narrowly won all the swing states.

And given that Harris underperformed her polls, one partial explanation for that could be she lost the bulk of late-deciding voters who made their choice too late to be captured in surveys. Harris lost so-called “double haters”, people who had an unfavorable view of both Harris and Trump, 52-32. There's really no evidence that the campaign closed strongly — maybe the opposite, really.

Plouffe: So it was a dead heat race. But, you know, at the end of the day, you know, the political atmosphere was pretty brutal and that’s not an excuse.

It is an excuse. Maybe it’s a good excuse; we’ll discuss that. But this was not a campaign that seemed to think it had a lot of agency over the outcome.

Plouffe: You had right track, wrong track, I think 28-72, about 70 percent of the country saying they were angry and dissatisfied. You had Trump’s approval rating on his first term frustratingly high, 48 to 51, depending on the state. Obviously, the incumbent president’s approval rating around 38 to 41, depending on the state. And you know, I think the economy and inflation is still driving a lot of votes. So I think given that we had a challenging political environment, the fact that we got the race to dead heat was positive, but boy, it was slow moving.

I’ll give the campaign this. As I covered on Monday, Trump’s approval rating from his first term (52 percent) was considerably higher than Biden’s (42 percent). That’s a difficult problem to overcome.

The question is to what extent this was exogenous from decisions made by the campaign or by the Village and the Democratic Party before then. What I’m trying to avoid is allowing the campaign to get away with too much circular logic. If their argument is that “well, Trump won because he was more popular”, that’s like saying the other team won the football game because they scored more points. How did Trump become more popular when he exited his first term with a very low approval rating?

On Nov. 5, 2023 — a year before Election Day — Trump’s favorability rating was −14. By Election Day exit poll, it had risen to a comparatively balmy −4. Some of that was the assassination attempt against Trump. Some of it may have been a sort of nostalgia: George W. Bush’s favorability ratings have improved considerably from the time he left office, for instance. And some of it may have been the various trials against Trump, which made him more sympathetic among his base.

Still, even with Democrats given eight months to prosecute the case — because Trump locked up the GOP nomination in early March, and Biden was trying to make it before Harris was — Trump only became more popular. Biden leaned heavily into the theme that Trump was a threat to democracy. One potential problem with that, as I’ve seen others point out, is that it made Democrats seem like defenders of the institutions of an unpopular status quo, with Trump getting to play the role of “outsider” despite being a former president himself. And to the extent those institutions were performing poorly, that only made Trump look better by comparison.

The argument that Harris outperformed in battleground states is dubious

Plouffe: And I think we were focused on seven states. You know, that’s our windshield into the world, the battleground states. But, you know, what we saw on election day was, you know, New Jersey and California and Connecticut and New York, massive shifts. So I think where Kamala Harris campaigned, we were able to keep the tide down a little bit, but it ended up being a pretty strong, you know, tailwind for Donald Trump.

The staffers repeatedly suggest that they overperformed in the battleground states relative to the rest of the country. I don’t exactly think this is something they should be patting themselves on the back for, however. They lost those battleground states, after all. All of them.

And in the non-competitive states, Democratic turnout dropped off more, perhaps because voters weren’t excited about Harris’s message and really the only thing that worked was the message that “Trump bad, so you have no choice” — a message more compelling to a voter in a swing state where your vote is more likely to make a difference.

Basically, in noncompetitive states, Harris had both a turnout problem (voters not showing up) and a persuasion problem (voters switching to Trump). In swing states, it was more of just a persuasion problem. Still, the differences should not be exaggerated. Here’s a table showing how Harris performed relative to Biden 2020 in all 50 states:

Yes, the biggest declines were in the blue states Plouffe mentioned, plus Florida. Because these are high-population states, they had a big impact on the popular vote and reduced the Electoral College – popular vote gap. But Harris’s margin declined by an above-average amount in two swing states — Nevada and Arizona — and by about an average amount in Michigan. Meanwhile, the smallest declines — she lost at least some ground everywhere — came in four noncompetitive states: Utah, Washington, Oklahoma and Nebraska. A regression analysis suggests her margins were about 1.2 points better in the swing states, controlling for the 2020 vote — but the coefficient is not statistically significant.

Maybe there’s a little something there — a point or a point-and-a-half is within the bounds of plausible for campaign effects like having a better ground game. But even so, the mechanisms aren’t clear. Trump’s (in)famous “Kamala is for they/them” ad was airing in heavy rotation outside of the swing states — I saw it a lot in New York. Maybe that ad was particularly effective in states where it was the only piece of campaign messaging from either side that voters were consistently seeing. But Harris lost ground in all 50 states, whether she was going toe-to-toe with Trump in the ad wars or not.

The “fundamentals” weren’t that bad for Democrats

Plouffe: And I think it’s worth reminding everybody we saw in ’22, even though that was a pretty decent democratic year, we saw these shifts, we saw them in ’20, we saw them in ’16. You know, Trump, specifically, but Republicans generally improving their vote share amongst non-college voters, particularly non-college voters of color. And this was a surprising race because Kamala Harris actually did, I think, better with senior voters than I think a lot of people would have thought. So, margin of error race where we inherited a deficit, we got it to even, but the thing never moved. So, to Jen’s point, I think we were, you know, we were hopeful.

I don’t know how optimistic we were, but we thought, okay, this is tied, and if a couple things break our way, and listen, I’m naive in this way, I just thought at the end of the day, particularly because Trump did not close well, I thought, and I thought Kamala Harris closed well, Trump was reminding people some of the things they don’t like about him. That that might give us what we needed. But at the end of the day, I think the political atmosphere, the desire for change, all those fundamentals that you’ve spent a lot of time talking about, really presented huge challenges for us. So, you know, we got, we got there, but we didn’t get the breaks we needed on election day.

I suppose I mostly agree that the fundamentals were rough for Harris, and that was a fairly consistent theme of Silver Bulletin coverage throughout the election. But one big reason I thought that is because, outside of New York Times polling, the data was projecting more of an Electoral College-popular vote split than the almost nonexistent one we actually got. (In the end, the tipping-point state, Pennsylvania, was only about 0.2 points to the right of the national popular vote.) Meanwhile, the fundamentals calculation embedded in our model, based on our Economic Index, actually projected Harris to win the popular vote by roughly half a percentage point, which would have produced a photo-finish Electoral College race.

Plus, another big reason why the fundamentals were tough for Harris was because of Biden and his unpopularity. As we’ll talk about in Part II, Harris did bafflingly little to distance herself from her predecessor.

Meanwhile, the Biden team rallied around Harris despite her mediocre electoral track record rather than give serious consideration to Josh Shapiro or Gretchen Whitmer or Gavin Newsom or anybody else. In the five swing states with Senate races, Democrats actually went 4-1. I don’t buy that this outcome was inevitable by any means.

Dan Pfeiffer: How deep was the hole that she had to climb out of?

David Plouffe: Well, Dan, I mean, I think, listen, there was the Biden-Trump 1.0, which is obviously pretty catastrophic in terms of where the race stood. When we got in, my recollection is some of that snapped back, but you know, we were behind. I mean, I think it surprised people because there was these public polls that came out in late September, early October, showing us with leads that we never saw. You know, I mean, it was just basically a race that in the battlegrounds was 46-47, 47-48. So that’s not where we started. We started behind. She was able to climb out. I think even after the debate, we might have gained, what, .5-1? It wasn’t a race that moved a lot. And so I think when you think about our own internal analytics, you know, if you have Wisconsin at 47-47, or Pennsylvania 48-47 Trump, let’s say, which I think is where we had it at the end, you know, you’ve got to have undecideds break your way more than your opponents, and you’ve got to get a little benefit from turnout, which we weren’t able to do.

Small point, but I’d argue those Harris internals were actually fairly consistent with the public polls. At no point was Harris ever in a better position than a toss-up in our model. If you thought otherwise, that’s because you were smoking the hopium.

The backstory of Biden dropping out is as bad as you might expect

Dan Pfeiffer: Obviously the defining event of this race was the candidate switch. And everything, every decision you guys had to make, everything you had to do was defined also by the compressed calendar in which you were operating in. Quentin, you were there when that switch happened. Were you able, there was a one month period between the debate and when the president actually dropped out. Obviously it seemed like dropping out could be a possibility. Were you able to do any thinking or planning in that one month period about what a race with the Vice President look like, or did you have to sort of start cold on that first day, the moment the, you know, you got the call or the statement went out?

Quentin Fulks: I mean, we started cold. We, there was no planning involved in any other candidates. I mean, we were honestly in, in crisis management mode of keeping President Biden in the race. You know, convincing, um, you know, Democratic allies, that he could still do this. And one of the things was trying to keep the President out on the road as much. We were still doing everything we could from a campaign, and he made the decision that he did not want to continue on.

OK, this passage is revealing. I’ve seen people criticize Fulks for saying “we started cold”. Shouldn’t they have been more prepared for the possibility that Harris or another candidate would have to take over?

Well, not necessarily if by“they”, you mean the people who were working for Biden. Because like Fulks says, trying to prop up Biden was surely an all-hands-on-deck operation — undoubtedly a total clusterfuck as Biden was constantly making new gaffes. Basically, the White House was trying to gaslight the political world about Biden being OK when he clearly wasn’t, even in their own internal polls. I imagine that being a lot of work.

But, yes, some group in the White House, such as Harris’s staff, should have been planning for this eventuality. Or if not them, outside Democratic groups should have been prepared.

I’m surprised, I guess, that Newsom — who was clearly interested in the nomination at earlier points in the cycle — didn’t explicitly signal he was ready to step in if required. This is where my “outside view” can only tell you so much. How did the White House fend off credible challengers to Biden, both earlier in the cycle and then at the bitter end? With persuasion — hopium after Democrats’ solid midterm? Or instead with threats of ostracization and retaliation? Dean Phillips, when he finally did challenge Biden, was blackballed by Dem-friendly but supposedly independent media organizations like MSNBC.

I understand the game theory here. Such a challenge, even the threat of one, would have weakened Biden. But Biden had become a likely loser even before the debate. Democrats like Nancy Pelosi eventually did find their courage, but the party’s behavior wasn’t consistent with viewing Trump as an existential threat. And it’s a little rich that the wing of the party that was standing behind Biden now blames the late date of the candidate switch for their problems.

Fulks: And he pulled some of the senior leadership together and said that he was going to be with the vice president. It also wasn’t anything that our team took for granted to just say, okay, she is the nominee. We knew that there was still a situation where we had to shore up delegates. And that’s where we started from. And then after that point, that is when we begin to say, okay, how can we define her? Also Trump’s favorability numbers were creeping up as Plouffe said, and we had to do something about that as well. And so it was a lot of walking and chewing gum at the same time, but there really was no sort of contingency planning to turn the race over to her right after that debate or at any point until President Biden definitively said he wasn’t going to continue on.

This is revealing, too. It sounds like Biden’s decision to endorse Harris — which, oddly, he made in a second tweet shortly after he announced he was dropping out — was not tipped off or even really hinted at in advance. This reflects poorly on Biden and frankly makes him seem like a narcissist. It shouldn't have been that hard to telegraph that he still saw himself as Option 1A, but Option 1B would have to be Harris, circumventing what at the time was a robust intraparty debate between Harris and some type of “mini-primary”.

However, one thing gives me real pause is that the White House reportedly was doing some internal polling on the Harris matchup before Biden dropped out. And it can’t have been a total shock: there was plenty of public discussion about Harris becoming the nominee. Once other candidates like Gretchen Whitmer signaled they weren’t interested, Harris was increasingly the only logical choice. So I’m not quite sure what to believe. But what the White House really needed to do is at least red-team out the scenario: designate a set of staffers who would start planning for a Harris takeover.

This also reflects poorly on Harris, however. The Biden staff frequently bad-mouthed her throughout her vice presidency and never had a high opinion of her. Why go with a Biden loyalist like O’Malley Dillon? Along with her running mate choice — something Pfeiffer should have asked about but didn’t — that seems like a clear mistake. Were the staffers working behind the scenes to ensure that if Biden dropped out, they’d stay on in their roles?

Finally, while I understand that staffers are expected to remain loyal to their boss, if O’Malley Dillon or Fulks had gone to the Washington Post and said “this isn’t working, our internal polls show him way behind, and it’s my duty to the nation to tell you this because the alternative is a second term for Donald Trump,” that would probably have set off dominoes that quickly ended Biden's bid. While they would have been ostracized by many Democrats, they’d be a hero to others, especially if the replacement candidate had eventually won. They’d get big book contacts and offers to host podcasts or cable shows. But the Democratic Party is dominated by classic Village types: people-pleasers who are afraid to rock the boat or take risks. Loyalty is not a virtue when it leads you to follow your boss off the cliff.

It’s not clear whether or how the short campaign hurt Harris

Dan Pfeiffer: Stephanie, I think what probably surprised a lot of people in politics was the vice president was a largely unknown quantity to much of the electorate. So you guys had, under a very short time frame, had to do two things, as Quentin said. You had to teach people about her and also make a case against Donald Trump, who would just come off, he was at an all time high, come off the assassination attempt, the debate against the president. In terms of messaging, how did you think about the balance between the two?

Stephanie Cutter: Well, the first thing we had to do is put on a convention. And we had about three weeks to flip a convention that was being built around Joe Biden. So, we were able to flip it, you know, to fit this very new character of a different generation, different experience, different background and looking at the data at the time, which Jen and Plouffe and Quentin have all talked about, there was, she had a huge deficit in favorability because either people didn’t know about her, or what they did know about her was based off of negative media. So, our first priority was to define her in that convention. Fill in her bio.

What’s weird is that Democrats actually did run a pretty good convention under difficult circumstances. Or at least it seemed that way to me: maybe the lack of a convention bounce suggests otherwise. And the first few weeks of the Harris campaign went well — that was the whole “Brat Summer” period. So the aides complaining that they got off to a cold start doesn’t fit the trajectory of the race. They got out of the starting blocks at a fast pace, only to lose steam and fail to switch gears later. To be fair, maybe any Democrat would have gotten that momentum boost just because Biden’s prospects had become so dire.

Cutter: As part of that, you know, we already knew how to do the negative on Trump. And we knew that there was a lot of Trump-nesia out there. People didn’t remember the four years of the Trump administration that badly because they had been through hell. They had been through COVID. Both under his watch and under President Biden’s. I’m putting aside a lot of the details of who’s at fault and what Biden did to dig us out and all of that. And then they had to deal with inflation. So they had been through hell. So looking back, you know, you remember a previous time much more fondly because you now think that you’ve gone through the worst.

Obviously it’s reasonable that campaign wanted to do some work to remind voters of how bad 2020 was under Trump, so I’m not blaming Cutter for this. Maybe they didn’t do enough of it, in fact. But the politics of 2020 are tricky. Many of the negative associations people have with 2020 — like prolonged COVID restrictions or the “racial reckoning” — are quite reasonably associated with the left. And also associated with Harris, who was picked in part because of her racial identity: the reporting was quite explicit about that both at the time and afterward.

Cutter: So we had to remind people what life was like. That was our second imperative. And then the third imperative as part of the convention and leading into the remaining days of the campaign is what’s that choice? What are the two very different visions between Trump and Kamala Harris? So the convention demonstrated a lot of enthusiasm for Kamala Harris, a lot of freshness, future oriented, bringing a variety of coalitions together. We had Independents, Republicans, Democrats, business leaders, sports figures, everybody coming together around a new way forward and finally turning the page.

Small thing, and it isn’t nearly as bad as, say, the Joy Reid version of the same argument. But the idea that all these celebrity endorsements or Liz Cheney’s endorsement or what have you were really going to move the needle speaks to what fundamentally seems like an incorrect theory of the case. To borrow Chris Arnade’s terminology, Harris needed to speak to “back row” voters, not the “front row” teachers’ pet types. Even with the benefit of hindsight, only Fulks seems to recognize (in some passages I’ll get to in Part III) that the campaign was aiming at the wrong target.

Cutter: So, you know, through the rest of that campaign, through the, our next thing was the debate just a few weeks later, and it was boom, boom, boom, all the way through probably early October after the Walz debate that we had to move through these things so quickly. Once we got through all of that, then the race started to gel.

And to the extent people were open to remembering what Trump– what life was like under Trump, we were trying to fill that in. To the extent people had questions about Kamala Harris, we were still trying to fill that in. So in 107 days, you know, what typically takes us a year and a half, two years in a presidential campaign, we were defining someone who was wholly undefined from the start, trying to remind people about the opponent and what life was like underneath him, and also take into account what the political environment was and the realities that we had to deal with which, you know, she was the incumbent, but she really wasn’t the incumbent. People didn’t know that much about her. The economy was still, slightly getting better, but we couldn’t really take credit for it. So we were in a bit of a crossroads trying to figure out what that October messaging and closing messaging would look like.

This the first of eight references in the transcript to “107 days”: the interval between Biden dropping out and Election Day. Each of Cutter, O’Malley Dillon, Fulks and Plouffe refer to this precise number at least once. When you see that sort of phrasing repeated, you can bet that it was part of the foundational narrative — the series of excuses that the campaign claims aren’t “excuses” — that the campaign was indulging in.

And there are a few things wrong with it. The most glaring issue — Cutter and Plouffe get a partial pass on this because they were brought in later, but O’Malley Dillon and Fulks certainly don’t — is the one I already pointed out: one of the main reasons they had only 107 days is because the White House insisted on propping Biden up, including for three-and-a-half weeks after the debate disaster.

Also, it’s at least theoretically possible that the short campaign would actually be an advantage for Harris — that she wouldn’t face the same scrutiny and could coast to a win off of vibes. This was, in fact, the conventional wisdom for parts of the race — even I bought into it at one point. But after a strong start, the campaign eventually ran out of things to say. With 14 days to go, the lead message on the Harris website was literally just that there were 14 days left:

By this point, the campaign ought to have had a few more tools in their arsenal. It’s not like a 107-day campaign is that short: it's longer than the “snap elections” that sometimes take place in the UK and other countries. Or necessarily even shorter than other general election campaigns in the US. Obama didn’t wrap up the Democratic nomination against Hillary Clinton until June 3, 2008, had a lot of intraparty healing to do (his leads against John McCain were initially relatively narrow) and a convention that wasn’t until late August.

Of course, I usually advocate for people in decision-making positions to have a realistic conception of the landscape. Harris had her work cut out for her, no doubt. At the time she took over for Biden, prediction markets and most observers (including me) had her as the underdog.

But if the campaign thought that Harris had only, say, a 40 percent chance, it wasn’t adapting its behavior accordingly. Instead, they were kind of nits:

• They went with the safer pick in Tim Walz instead of swinging for the fences with Josh Shapiro.

• They were very averse about doing traditional media, at first. (We’ll cover this more in Part II).

• They could have been more aggressive about challenging Trump to a second debate on Fox News (rather than just CNN) and then mocking him when he declined.

• They did attempt one sort-of Hail Mary, with Harris promising full marijuana legalization in October — maybe a good issue with “back row” voters — but couldn’t really drive a news cycle out of it.

• They never ruffled feathers or threw Biden or “the groups” under the bus.

• They constantly worked to tamp down expectations, which makes them look smarter given the result, but possibly also tamped down enthusiasm.

So how much of this was realism and how much of was learned helplessness: feeling like you have no responsibility for your predicament and no way out of it?

We’ll pick up the thread in Part II.

1. The second post is going to be paywalled and this one isn’t, and paywalled posts of course usually generate more revenues, so I don’t want to shortchange GiveDirectly by only having this offer on a free post.

2. There’s no Roman numeral for zero, apparently!