2024 Election Predictions
Introduction
2024 Presidential electoral map based on a consensus forecast from composite ratings of several forecasters found at 270towin.com.
*Both Nebraska and Maine award electors based on percentage of the vote rather than winner take all. Thus, those states have stripes to indicate that the vote may be split.
For reference:
-Dark Blue: Safe Democrat
-Medium Blue: Likely Democrat
-Light Blue: Leans Democrat
-Brown: Toss-up
-Light Red: Leans Republican
-Medium Red: Likely Republican
-Dark Red: Safe Republican
As the name of the website 270towin.com suggests, a candidate must receive at least 270 electoral votes to win the general election. As was the case in 2016 and 2020, a candidate can win the popular vote without winning the electoral college. Thus, the electoral map can be a bit of a puzzle for a campaign as they try to ascertain which states may be within reach and, therefore, worth their money and attention; which states are already in the bag, so to speak; and which states are bound to be won by the opponent - all while in pursuit of at least 270 votes.
Much like in the 2020 race, the election will essentially come down to two clusters of states: The Rust Belt and The Sun Belt. Below is the map I created on the same website, 270towin.com, followed by an analysis of my predictions with regard to toss-ups/battleground states.
The Rust Belt
Background from my 2020 blog: In recent elections, many of the most coveted electoral votes have come from Rust Belt States like Michigan, Pennsylvania, and Ohio.
The Rust Belt has become an interesting region politically. As in my home state of Pennsylvania, there has been significant exodus from the Democratic Party across these states. I suspect this shift has been due to two significant factors: 1) the weakening of unions and disappearance of traditional union jobs in the steel mills, coal mines, etc. and 2) the leftward socio-cultural drift of the Democratic Party has pushed some traditional (now former) Rust Belt Democrats to a more comfortable home in the brasher, less politically-correct Republican Party. This shift has been occurring certainly over the past three decades, but has been most noticeable in through the past three election cycles: 2008, 2012, 2016.
In 2008, Indiana, Illinois, Michigan, Ohio, Pennsylvania, and Wisconsin all went blue. Obama lost Indiana in 2012 and, while still holding the other states, hemorrhaged support among traditional Rust Belt Democrats, for whom the excitement of electing the first Black president had worn off and the Party itself was continuing to lose its appeal. By 2016, Illinois was the only state from this list that remained blue, in large part due to Chicago and its suburbs. The maps below offer a direct comparison of the 2008, 2012, and 2016 (and now 2020) electoral results. Notice the shift along the Rust Belt (below the Great Lakes) from blue to red.
Update: As I’d predicted in 2020, Michigan, Pennsylvania, and Wisconsin all flipped blue, while Indiana and Ohio remained red. Not to pat myself on the back too much, but that prediction is one of my prouder moments.
Rust Belt Predictions for 2024
INDIANA will remain red for a fourth election in a row, and it’s not even close.
MICHIGAN will…Listen, the race is so close this year that I wouldn’t be surprised either way. Harris currently leads by the slimmest of margins and has been ahead of Trump in the polls here since the beginning of August. Even her greatest margin of the past three months was less than four points, and once RFK, Jr., dropped out of the race, Trump narrowed the gap. While polling shows this state is a virtual tie, I’m obligated to make a prediction. Not the ideal way to make the call, but my gut tells me Michigan goes to Harris.
OHIO will remain red. The days of Ohio being a swing state are miles back in the rearview.
PENNSYLVANIA flip red. Don’t get me wrong, I think it will be by a hair’s breadth. But there are a couple of factors that play into this prediction. 1) Pennsylvania has a generally well-liked Democratic governor and currently has two Democratic senators. It went for Biden in 2020. That’s a lot of blue for a state as purple as PA has become in the past 15 years. 2) Trump was almost assassinated 20 minutes down the road from my house. If the state’s general unease with its apparent leftward lean isn’t enough to get conservatives into the voting booths, the impact that the events of the Trump campaign has had on Western Pennsylvania will get it there. And Harris feels it. That’s why today it was announced she’d spend her election eve doing a last-minute Get Out The Vote event in Pittsburgh. No polls referenced in this one; I’m going wholly on vibes, baby.
WISCONSIN will…stay…blue? Look, y’all. I don’t know. Maybe it will go red. It’s as tight as Michigan, and with a strikingly similar polling trend of Harris up by a couple points until mid-August and the margin closing to dead even as of a few days ago. At the moment, Harris is up by a little under a point. It’s a tough call, but here’s what I’ll hang my hat on: Tammy Baldwin is the Democratic incumbent running for senate, and she is slightly favored in most polls. Wisconsin’s other senator is Republican Ron Johnson. The last time Wisconsin had 2 Republican senators was in 1957. I think the pressure will be on Democrats to keep Baldwin in her senate seat, which will give Harris the edge. Additionally, Governor Tony Evers, a Democrat, overperformed his polling two years ago when he ran for reelection and won a second term. I suppose that will be enough for me to, with great trepidation, place Wisconsin in the blue column.
The Sun Belt
The Rust Belt is primarily a question of whether former blue states will settle into new voting patterns that leave them reliably red or continue to shift from one election to the next. In the Sun Belt, the question is whether traditionally red states will begin to go blue. In 2020, Arizona's polling leaned blue for the first time since 1996. Texas and Georgia were both considered toss-ups for the first time in modern history. North Carolina was and is a toss-up after going red in five out of the last six elections. And of course, there's the former perennial question mark: Florida.
Sun Belt Predictions
ARIZONA will remain blue. In discussing the upcoming election with my brother (shoutout Trevor), I settled on a theory that plays a role in several of my predictions, and now is a good time to introduce it. Kari Lake, the news anchor turned 2020 gubernatorial candidate turned election conspiracy theorist, is running for Senate against Ruben Gallego. While it would be shocking for Arizona to have two Democratic senators, I have to imagine that the straight-shooting Republican mavericks of The Grand Canyon State will once again be turned off by both Lake and Trump, as they were in 2020. I think lack of enthusiasm for the top of the ticket will lead a lot of would-be Trump voters to stay home, leave the President and Senator votes empty, or write in the ghost of Barry Goldwater. If it’s not clear, I love the political vibe of Arizonans. Possibly my favorite state and I’ve never even been there.
TEXAS will remain red. Democrats get eager every election cycle, like Charlie Brown lining up to kick a field goal before Lucy predictably and gleefully snatches it away. While Allred’s linebacker shoulders will continue to carry the Democratic Party apparatus forward in Texas, it’s not going to be enough to notch him, or Harris, the win.
FLORIDA will remain red. I made this call with some unease in 2020, but it feels predictable to make it this year. Growing up, I thought of Florida and Ohio as the swing states, but that is less true these days for either of them. While there’s an outside shot that it becomes “less red”, I can’t imagine we’ll get a surprise Harris victory in Trump’s home state.
GEORGIA will flip red! Georgia Democrats are exhausted. In 2020 alone they had three major races: president and two senate seats. After an exceptionally close election led to a recount by hand, Biden eked out the first Democratic presidential win in the state since Bill Clinton in 1992. Both of the races for Senate, David Perdue (R) versus Jon Ossoff (D), and Kelly Loeffler (R) versus Raphael Warnock (D), were so close they went to run-off. The Democrats also clinched both of those races. Words cannot describe the Herculean organizing effort it took to get out the vote for not one but two major elections in the span of 2 months. To top all of that off, Georgia’s governor and secretary of state both stood firm regarding the integrity of the election while many other Republicans cried foul, thus endearing them to many centrist Democrats and Republicans. After the chaos of 2020, the stakes just don’t feel as high locally this year. I don’t think Democrats can generate enough enthusiasm this time around to notch Harris a win. That said, Georgia was my one mistake last election cycle, so maybe I shouldn’t count them out.
NORTH CAROLINA will…go blue. The last time this happened was in 2008, and it’s not a sure thing, but for the same reason I think Arizona will be blue while Georgia and Pennsylvania flip red, I’m putting Carolina in the Democrat column. North Carolina’s current lieutenant governor, Republican Mark Robinson, has had a potpourri of scandals during his current bid for governor. Moderate Republicans who might otherwise hold their noses and vote for Trump have to contend with their options for governor as well, putting them in a similar position to like-minded voters in Arizona.
NEVADA - a state with a section all to itself
I forgot to include Nevada in my 2020 write-up, and I wish I could forget to include it now. I genuinely have no idea which way this state will go. It’s long been a blue state, but it’s been inching toward red the last several election cycles. Maybe this will be the year it flips.
Conclusions
Ultimately my map shakes out to a Harris victory, 278 to 260. The races in Wisconsin, Michigan, Pennsylvania, North Carolina, Georgia, Arizona, and Nevada are so close and have been for at least the last 3 months. With the margin I’ve predicted, if Harris loses any one of the four swing states I’ve projected her to win, she will lose the race.
Polling error has been a topic of much discussion among politics nerds, and I imagine it’s broken through to the mainstream a bit as well. In 2016, very few people believed Hilary Clinton would lose based on aggregate polling which gave her, per Nate Silver, 71.4% to 28.6% odds over Trump. In part, that was a lesson in how to understand election forecasting. Nearly 30% odds are not bad, and do not mean that Clinton was predicted to have 71.4% of the vote. Essentially, Silver’s forecast ran a series of election scenarios based on polling data, and 70% of the time, Clinton came out on top. Trump certainly beat the odds, but Silver took undue heat for his FiveThirtyEight election forecast. That’s not to say there wasn’t an issue with polling itself. There certainly was. Final estimates of the outcome of the 2016 race showed that 88% of polls overrepresented Democratic voters (per the Pew Research Center). The polling error was even worse in 2020, where 93% of polls were skewed in Democrats favor, though in this instance, Biden’s victory was enough to overcome the margin of error. There are myriad explanations for these errors: “Shy Trump Voters,” Republican distrust of pollsters, a Democratic conspiracy. I’ll tell you what I believe.
In 2016, Trump brought legions of low-propensity voters off the sidelines - people who were disengaged from politics because they thought the system was broken and that no one represented their interests. Trump made them feel seen and gave them hope, he mobilized them. Pollsters traditionally poll “likely voters,” meaning individuals who vote consistently in election after election, because polling individuals who vote erratically could contaminate the pool with opinions that likely won’t appear in the actual election due to lack of turnout. Thus, pollsters entirely missed the Trump surge among low-propensity voters. Those same low-propensity voters now make up the largest, most enthusiastic portion of Trump’s base, while more traditional Republicans often find themselves on the outskirts of the party.
During the 2020 cycle, I kept hearing talk of “Shy Trump Voters.” The theory is that in 2016, people who supported Trump were embarrassed to say it to a pollster. While I can believe that there are parts of the country where people might be anxious about showing their support with a yard sign or a bumper sticker (the same way many Democrats are in places like where I live), it doesn’t make sense to me that those same people would be anxious to tell a stranger over the phone that they are voting for Trump in an anonymous poll. And when the polls were equally bad in 2020, some jumped to the conclusion that the “Shy Trump Voters” had struck again. And again, I did not think this was the case, but I didn’t have a solid theory as to what had actually happened until today. I was listening to The Remnant on my favorite podcast network, The Dispatch, and conservative thinkers/pundits Jonah Goldberg and Chris Stirewalt posited that the polling error in 2020 was likely caused at least in part by the COVID-19 pandemic. Democrats and Democrat-run areas were more significantly locked down than Republican-run areas. As a result, many Democrats spent months at home, craving social contact and growing in their willingness or even eagerness to answer a pollster’s call. I’ll add to this theory: Democrats also had enthusiasm on their side. Polling from the 2024 election has indicated that Democrats were far more willing to discuss the race with a pollster once Biden had dropped out. Similarly, in 2020, Democrats were fairly single-minded in their desire to defeat Trump in the election. On the other hand, many Republicans had continued with their daily activities to the greatest extent possible, silencing their phones as they buzzed in the check-out line or missing calls while visiting with friends. They also didn’t perceive Biden as a threat. He was doddering, rheumatic, dementia-addled. The enthusiasm from 2020 had waned in the perceived absence of a worthy adversary. The combination of these conditions resulted in a polling error even as pollsters made an effort to correct for their mistakes from 2016.
Then 2022 came around. It was not a presidential cycle. Just House seats, Senate seats, and some Governorships up for grabs apart from local elections. The restrictions from the pandemic had been lifted and the status was decidedly quo. And the polls were more accurate than they’d been in any election since 1998, according to FiveThirtyEight
Now what does that mean in the year of our Lord 2024? It’s impossible to know at this point. Pollsters have continued to try to correct for the errors of the previous two presidential election cycles, and the result, whether accurate or not, has been a dead heat in all the swing states. It certainly feels tied, regardless of the polling. However, it’s possible that the Republican base continues to elude polling calls. It’s also possible that a number of polls are overcompensating for this and incidentally juicing Republican numbers. There is literally no way to know until the results of the election are in.
Some parting thoughts as a politics junky and elections worker:
No state finishes counting their votes on election night. The official certified results don't come out for days or even weeks.
The media projects the winner of the election at the earliest possible time based on exit polls, early reporting from key precincts, and electoral math. Often they do this responsibly. Sometimes they do this irresponsibly. For example, in 2000 the election hinged on Florida. The media called Florida in favor of Gore, only to flip hours later and announce a Bush victory. Ultimately the election was too close to call that night, the margin so slight that it triggered a recount in several counties.
Speaking of the 2000 election, the results were not finalized for over a month after the election.
2020 was an unprecedented election with an unprecedented number of absentee and mail-in ballots that needed to be processed. This year there will likely be a drop-off in mail-in voting, though I’m not as bullish on that as some people are. In 2020, 46% of voters used mail-in ballots, which was a spike in the trend that saw about 7% in 1992 and up to about 23% in 2016. In election cycles prior to 2020, the majority of mail-in ballots were used by Republicans. You may recall that in 2020, Trump and others discouraged Republicans from using mail-in ballots, asserting that it was more secure to vote in-person. This year, however, Republicans are at worst neutral on mail-in ballots, and some are encouraging voters to use this method because mailing in ballots as early as possible can lessen the strain on local organizers who spend time calling and texting members of their party to get them to vote. I think there will be some decrease in mail-ins in 2024, but it won’t return to its pre-2020 trajectory.
Some states have already begun to process early and mail-in votes. Others, like Pennsylvania, will not begin to process until Election Day, hence the lengthy delay for results in 2020. If this makes you frustrated and you live in Pennsylvania, contact your state representative.
More Pennsylvania-focused stuff: The Supreme Court of Pennsylvania (SCOPA) recently ruled 4-3 that if someone mailed in a ballot that is ultimately rejected, but they voted provisionally in-person on Election Day, their provisional ballot should be counted. On October 28, The Republican National Committee, The Republican Party of Pennsylvania, and Butler County (where this case originated) petitioned the Supreme Court of the United States (SCOTUS) to stay that ruling, meaning they want SCOTUS to prevent provisional ballots completed by Pennsylvania voters whose mail-in ballots are rejected from being counted. The text of PA Act No. 2019-77 a/k/a SB421, which amended a 1937 act of the Pennsylvania State Legislature states “A provisional ballot shall not be counted if:…(F) the elector’s absentee ballot or mail-in ballot is timely received by a county board of electors.” Therefore, if the absentee or mail-in ballot was rejected due to a deficiency (e.g. failure to place in the ballot in the security envelope, failure to note the date on both the ballot and return envelope), the text of the Act states it cannot be cured (corrected) using a provisional ballot because the deficient ballot was received in a timely manner. I don’t suspect SCOTUS will take up the request for a stay, but even if they do, it might not be to Republicans’ benefit. As is often the case for any of us, it seems they are still fighting the last war in thinking that the majority of voters impacted would be Democrats since Democrats made up a significant majority of mail-in votes in 2020. Either way, I don’t imagine this would have a large enough impact to determine the results of the election in Pennsylvania. Just an interesting tidbit to discuss with your family at dinner tonight.
Ultimately, how to count votes, cast ballots, and determine electors are decisions made by the states. That's why these procedures vary from state to state, and why we'll likely receive results from some states quickly and others will take a while.
I WILL REPEAT WHAT I SAID IN 2020. Sit tight. Don't listen to early media projections, even if you agree with them. Don't listen to candidates declaring early victory, even if you agree with them. Stay the course and wait until we can be confident the votes have been counted. If one or both candidates jump the gun and we let ourselves be dragged along with them, it could be extremely destructive to our democracy and to our personal relationships. So sit tight, no matter which side you're on.