A 'Pre-Mortem' for Both Campaigns

As we approach the end of this election season, the data are so close and contradictory that it is almost irresponsible to give a firm answer as to who will win. The true answer is that the data point to an extremely close election, well within any prediction error margin. We simply do not know who will win.

So instead, I’ve written up a sort of “pre-mortem” for both campaigns. These pre-mortems encapsulate what I think are the best arguments for a Trump win as well as the best arguments for a Harris win. They sort of explain what I think things will look like the day after a Trump or Harris victory.

In addition to the value added by exploring the best “pro” and “con” arguments for each candidate, these scenarios also provide what I think is the most important takeaway. There’s not a ton of differences between the scenarios I describe. The meaning for the parties, political coalitions, and so forth should be roughly the same. A narrow Harris win tells us little more about the future of the Republican Party than would a narrow Trump win. But that’s not how it will be interpreted, even if it should be.

Without further ado: 

****

Well, after one of the craziest presidential years in recent memory, Donald Trump has pulled off a feat no U.S. ex-president has pulled off since Grover Cleveland back in the 1800s: winning non-consecutive terms. Trump is now both the 45th and the 47th president of the United States. 

He brings with him a narrow House majority, accompanied by a surprisingly strong Senate majority. That Senate majority is qualitatively different than that from Trump’s first term, with more senators elected on pledges to support him and fewer finding themselves in total opposition to him. The Jeff Flakes, John McCains, and Mitt Romneys are out; the Josh Hawleys, Bernie Morenos, and Dave McCormicks are in.

In retrospect, however, the signs were there all along. Let’s start with some simple facts: Harris was the vice president for a deeply unpopular administration, who got into the race so late that she never really had a chance to define herself. In fact, given the opportunity to differentiate herself from her boss, she famously demurred. It helped that she was running against Donald Trump, but this was counterbalanced by the fact that his presidency was largely remembered fondly. Republican gains in registration more than told the story on enthusiasm.

Against that backdrop, the early voting numbers told us more than we expected. For years, prognosticators have parsed the early voting tea leaves and come to the wrong conclusions about what it was telling them. Then, in 2022, Democratic analysts suddenly got it right. This year, it was the Republicans’ turn, as historic Republican strength in early voting signaled true enthusiasm among the GOP and a pathway to victory. 

There was, of course, some cannibalization of Election Day voting, as many traditional Election Day voters turned out early. Election Day wasn’t as red as anticipated. It didn’t matter though. By pushing the early vote to a draw, even a slight red hue on Election Day was enough to push Trump over the finish line. Early voting also seemed to do its job, as the GOP was able to marshal its resources toward turning out low-propensity voters on Election Day. Moreover, the early vote correctly showed a diminution in black and Hispanic participation, which ate heavily into Democratic margins.

Does this mean early voting modeling is the wave of the future? Probably not. The fact that Republicans predicted a Republican win in 2024 and Democrats did so in 2022 is really mostly consistent with people correctly predicting that their political priors will win out. The real proof will come when Democrats start to predict good Republican years off of early voting numbers, and vice versa.

The other big story, of course, was pollsters underestimating Trump’s showing in the upper Midwest. In retrospect, the results make perfect sense in light of the data. Polls had shown a big pro-Trump movement in the Sun Belt and nationally. It never made much sense that the Rust Belt states would completely buck the national trend alone. 

Finally, Harris’ strong showing among seniors in polls should have been a flashing neon red sign that Trump wasn’t going to underperform in the Upper Midwest. After all, this was a tell in 2020 that the polls would be off. Those voters have always been among the strongest Trump supporters, and as expected, Trump ran well among them. It’s not clear why older, rural Trump voters are unwilling to talk to pollsters, but the most likely answer is that these voters are defined by extremely low levels of social trust, which means both that they aren’t willing to talk to pollsters and that there’s simply no way to weight around the issue. 

Overall, the simplest answer was the best one: This is Trump’s best polling year ever, and unsurprisingly he got his best election result ever. With a narrow popular vote victory, Trump stands as a political colossus, having twice won with almost every political and social institution arrayed against him (including a chunk of his own party) and having only lost (narrowly) in an almost impossible situation for an incumbent. He also won with a coalition less heavily dependent upon white voters than any Republican in history, setting up potential tension within the Republican Party about whether to expand upon this growing base or to try to win suburban whites back. He strides back into the White House promising big change, and with a better means to make it happen.

****

Well, after one of the craziest presidential years in recent memory, Kamala Harris has broken the ultimate glass ceiling, becoming the first female president of the United States. While Donald Trump has yet to concede, she is now on track to become the 47th President of the United States. 

She brings with her a narrow House majority. Although Republicans won a small Senate majority, Democrats feel like they dodged a bullet. There was a clear path at the beginning of the cycle for Republicans to reach as high as 57 seats without any real “reaches.” Holding their majority to 51, with a recount pending in Ohio, was no mean feat, and Democrats are hopeful they can retake the upper chamber in 2026.

In retrospect, the signs were there all along. Although President Biden was unpopular, Harris was not Biden. More importantly, Trump was still Trump. Exit polls suggested that Harris won big among the “double haters,” propelling her over the top.

Republicans did their best impression of Democratic forecasters, fell for the siren song of early voting numbers, and fell flat on their faces doing so. Although they won the early vote in states where they had never done so, they found themselves panicked when the voting booths in red precincts were largely empty on Election Day. Democrats narrowly won the Election Day vote, and that was enough. 

It was a mixed bag for the polling industry. First, polls likely overcorrected for their misses in 2016 and 2020. Terrified of underestimating Trump yet again, these pollsters adopted techniques like weighting by recalled vote. What that accomplished was to assure that they would all repeat the same outcome while missing important shifts in the electorate.

The second portion of the answer is that something Republicans had touted ended up having negative effects on their voting coalition that Democrats had long complained about: Demographic shifts in the coalitions resulted in inefficient vote distributions. Republicans did well among minority voters, moving the racially diverse Sun Belt states back into their column. Trump had the best showing for a Republican candidate in New York and California in recent memory.

But the problem was that those gains yielded no extra electoral votes. Trump then continued to bleed out among whites. Since the Rust Belt states are heavily white, this was felt disproportionately there; gains in Milwaukee were more than offset by losses in crucial Waukesha County, Wisconsin, while rural support for him continued to revert to mean. Again, the lack of Trump signs blanketing rural byways (and even a smattering of Harris signs) and empty seats at rallies should have been an indicator taken more seriously here. This drop-off was most pronounced among older voters, who reverted somewhat to ancestral Democratic voting. Ann Selzer retained some bragging rights as the final result in Iowa was closer to her prediction than to that of the competing Emerson College poll.

This inefficient result should have been no great surprise either: The same thing happened in 2022. Republicans actually performed quite well that year nationally in the House, improving their popular vote share substantially. But because the gains occurred in heavily blue majority-minority districts, it didn’t translate to seats.

Finally, no analysis of a Harris win would be complete without a discussion of gender. Harris’ campaign leaned in on their candidate’s gender, while Walz provided a contrast in masculine styles with Trump. With women constituting a disproportionate share of the electorate, as has traditionally been the case, the gender gap worked in her favor. That Trump had multiple missteps down the stretch did little to help his case trying to persuade female voters to support him; if nothing else, it reinforced their concerns.

Harris obviously made history, but much of the analytical focus has been on the Republican side. Some of this is because the Republican Senate is expected to stymie most of Harris’ agenda, but another part is that the Republicans have devolved into the proverbial circular firing squad, which is far more entertaining to watch. 

Diehard Trump supporters naturally pointed to voter fraud as a cause for his loss, but also pointed to Trump’s gains with non-white voters as showing the strength for a future GOP MAGA-aligned coalition. They also point out that Trump has run better three times than establishment Republicans like McCain and Romney, and that he’s still the only Republican to win the Upper Midwest since 1988.

On the other hand, since Trump’s narrow win in 2016 against a historically unpopular Democrat, Republicans have suffered disappointments in 2018, 2020, 2022, and now 2024. The party has only won the popular vote once since 1988 (although Trump came very close), and it is unclear whether a Republican not named Donald Trump can sustain the gains among non-white voters he enjoyed. Opponents of the Trump approach suggest trying to win back marginal suburbanites, rebuilding the historic GOP coalition while hopefully retaining at least some of the gains with black and Hispanic voters. Meanwhile, the guessing games begin as to who will be in the GOP field in Iowa, which is only 1,167 days away.