The 2024 Election, Six Months Out

Nothing is certain about the outcome at this point.

This article was originally published on May 3rd, 2024.

—

"Trump & Biden" by ekaden is licensed under CC BY-SA 2.0 .

I recently used an intro for a speech answering questions about Donald Trump’s vice presidential selection in forensics tournaments that goes something like this: Despite being "x" months away from election day, we know pretty much everything there is to know about this race, and a lot more than we typically would at this point in the cycle. We already know that the two major-party nominees are, of course, President Biden for the Democrats and former President Trump (again) for the Republicans, a historic 2020 rematch. We already know what the main issues of the campaign will be: abortion, the economy, the conflicts in Ukraine and Israel, and immigration. And we already know (or are at least pretty certain) that this election will be close. Very close.  

Now, this would be where I transition to the question the speech will answer and say something like, “But we don’t know who Trump’s running mate will be,"  etc. But I would argue that there's a much more important unknown. Despite knowing so much about this election already, we don’t know who will win, or even who is favored. 

So today, exactly six months out from election day, let’s take a look at where things stand. And no, I can’t say who will win, at least not statistically. I’ll get this out of the way right now by saying that if I personally had to make a wager today, I’d probably say Biden would narrowly win. That’s due to Trump’s trials, unpopularity, lack of campaign funding (in part because the money is going to legal defenses), and lack of campaign organization. His daughter-in-law, who arguably has no applicable qualifications, is currently RNC chair, and his campaign has virtually no field offices in battleground states, while Biden has many. But despite all of the issues Trump faces, there’s a lot of personal uncertainty for me picking Biden, and thankfully, that uncertainty is something we can quantify statistically. 

As of writing this article, I have yet to release (or finish constructing) my 2024 forecasts for the Presidential, Senate, House, and Governor races. They will most likely be published in that order. However, I have been running the presidential and Senate nowcasts, and we do have polling averages at HDRAnalytics.com/Polls that we can examine. And with that, let’s begin with the topline number. If the election were today, Donald Trump would be likely to win a second, non-consecutive, incumbent-removing term. That would be historic. Our nowcast gives him an 81% chance of winning (again, if the election were today) based on current polls at both the state and national levels. And it’s not hard to see why this is the output we’re getting. 

Trump currently leads in all six (or seven, depending on who you ask) key battleground polling averages. It’s pretty much universally agreed upon that this race will come down to the Sun Belt (Arizona, Georgia, and Nevada) and the Rust Belt/Blue Wall (Wisconsin, Michigan, and Pennsylvania). Some say North Carolina could still be in play, and we will discuss it today, as Biden’s actually performing better in NC polls than in Nevada, a state that both he and Hillary Clinton carried. (Trump won North Carolina in both 2016 and 2020, by a much closer margin in the latter.)

Current nowcast margins from the 7 key swing states (Biden is on the left, Trump on the right)

It’s worth noting that Biden won six of these seven states (the exception being North Carolina in 2020) and won the election, but Trump won six as well (excluding Nevada) in 2016 and won the election. All together, the seven total 93 electoral votes, or about 17% of the Electoral College, are certainly enough to decide the election when combined with safer states. So the mere fact that these states are competitive (which was expected) means a Biden re-election is very far from guaranteed. In fact, Biden was leading in the polls in these states by more than five points and yet won some by less than one percent. The polling error was fairly correlated between competitive states, and polls overestimated Biden by a noticeable margin, but much less so nationally.

2020 Polling error in each of the seven key swing states (from FiveThirtyEight)

And there are worrying signs for Biden at the national level as well. Currently, he trails Trump in our national polling average by just under a point. (43.7% for Trump, 43% for Biden, 5.1% for third parties, and 8.2% undecided) Now, to be completely honest, I fully expect Biden to reclaim his lead here in the national polls as we get closer to November, but he doesn’t just need a lead. He has to be up by about six or so points for there to be a fairly high degree of certainty. A typical national polling error in a presidential election is two-to-three points. Even in 2016, when state polls performed significantly worse than usual, national polls actually did pretty well, with FiveThirtyEight’s national average only missing by just under two points, at a 1.8% error. But of course, we don’t elect a president with the popular vote. And the general rule of thumb with the Electoral College is that the Democrat needs to win the popular vote by around 4-5% to secure 270 electoral votes. Look to 2016 and 2000, where both Clinton and Al Gore lost the election despite winning more votes nationally, as their margins were both less than three percentage points. So, if we assume a two-to-three-point error is realistic and Democrats need to win by over three points, things aren’t really safe at all until there is a six-point lead for Biden.

So in the interest of keeping this whole thing relatively short, I’ll discuss two more things; each candidate’s easiest path to the presidency, and my personal take on where the electoral map ends up in November. For Biden, he needs to carry Minnesota, Maine, Nebraska’s 2nd congressional district, and the Rust Belt, so Pennsylvania, Wisconsin, and Michigan. He won all of that in 2020, and I’d say his weakest states right now are Wisconsin and Michigan. Trump, on the other hand, just needs a bit of 2020-subtraction from Biden's winning map. His best bet is to hold North Carolina and Florida, flip the more favorable Sun Belt (Arizona, Georgia, Nevada) which seems like it has a pretty good chance of flipping to the GOP.  From there, he just needs to flip one of Biden’s Rust Belt states and he wins. Given how the polls stand now, it’s easier to see a Trump win than a Biden one. But we’ll explore this more with our forecast in the coming weeks, once that launches.

Finally, putting all data aside, as I said, I’d prefer a wager on Biden over Trump (slightly) right now. As for the map I’m expecting, I see Trump holding everything he won in 2020, and flipping Arizona, Georgia, and Nevada. I do believe Biden will hold Pennsylvania, given his Scranton background and strength in the Philadelphia suburbs, and I think he holds Wisconsin as well, especially with Senator Tammy Baldwin running down-ballot for another term. Frankly, Michigan was to the left of both PA and WI in 2020, so I'd say he'll pick that up too. But again, all of this will almost certainly change in the coming months, so stay tuned.

So, for now, that’s that. Trump would be favored to win in a hypothetical election held today, and is probably favored if you look at the polls, but things will almost certainly tighten and improve for Biden over the next few months. I plan on writing much more in detail about the race and the fight for Congress as well here at HDR Analytics. Thanks for reading.