After months of unavoidable election ads, assassination attempts, and candidate swap-outs, it is finally time for the 2024 election! Now, a lot of a people are dreading this one, and for good reason, but the goal of each of these posts is to explain the WHY and HOW of this election. This post is specifically focused on the ‘HOW,’ meaning how each state, congressional district, and demographic is going to vote in comparison to the last election. This post is also only going to focus on statistic predictions and electoral analysis, meaning it’s focused solely on the cold, hard numbers. The other post (The 2024 Election Analysis) is going to be more subjective, where I give my take on each of the campaigns and what they did to win this election. To summarize:

This post: “Trump will win Pennsylvania by +2.3% because he will overperform amongst union workers and young men in comparison to 2020.” The other post: “There are two errors that the Harris campaign has committed since July, and it’ll potentially cost her the election: 1) she can never get to the point and 2) she’s played it too safe.” So, if you would rather read my subjective take on this election, the link is here:

Finally, a couple of disclaimers:

#1 (Easily the most important): I have spent the last three election years deep diving into data and I consider myself pretty well-versed and qualified to talk about a lot these things and to make an election prediction. But at the end of the day, this is little more than an educated guess. When I tell you that this is the closest election from a polling perspective I have ever seen take my word for it. If all of the polling averages were EXACTLY correct, Kamala Harris would win the election with exactly 270 electoral voters. She would win Pennsylvania, the state needed to get her there, by less than 0.1%. The top election modeler I know is giving Harris a 50.01% chance to win as of this morning. The closest I have ever seen a polling average get to the final result is 0.3% off. The polls will be wrong to some degree (they should never be taken as the gospel), and I’m going to give my best guess as to why, but I am not perfect. DO NOT take this post as a predictor of what is going to happen. Because I could very easily be wrong, and I will almost certainly miss at least one call.

#2: Do not take what I think is going to happen as to what I want to happen. Full disclaimer: I did not vote for the candidate I think is going to win. I have done my best to remain neutral in order to give an unbiased take on the election and why people are motivated to vote for their preferred candidates this year. Statistics couldn’t care less about what we feel, and I want to make sure that I’m giving an unbiased opinion so the people who read this can walk away having a better understanding of how our elections work and HOW candidates win states to win the election. Think of this as a being a blog version of the NBC guy with the map every election night and he’s rattling off numbers you don’t understand.

Okay, now that that is out of the way, here is my official prediction:

In the 2024 election, EVERYTHING SWITCHES:

Republicans will pick up 3 Senate seats and win a 52-48 majority;

Democrats will pick up 6 House seats and win a 220-215 majority;

and Donald Trump will win 280 Electoral votes and become the 47th President of the United States.

PRESIDENTIAL RACE:

Donald Trump: 280 Electoral Votes (Winning Wisconsin, Pennsylvania, Georgia, North Carolina)

Kamala Harris: 251 Electoral Votes (Winning New Hampshire, Nebraska 2nd CD, Michigan, Arizona, and Nevada)

Popular Vote: Kamala Harris, 48.7% (~77.2 Million Votes); Donald Trump, 48.3% (~76.4 Million Votes). Harris +0.4%

State Races:

Michigan: Harris +1.2%

Out of all of the swing states, Michigan represents the best place for Harris to hold on. With a high college-educated white population, as well as recent successes in the last midterm election, this state has been great for Democrats since 2018. But, there are warning signs. Black voter turnout in Wayne County could be a problem. Michigan has the highest population, as a percentage, of Arab Americans in the country, at 2.1%. And there is a lot of dissatisfaction about the Israel-Hamas war, which could lead to depressed turnout or swings to Trump. Although it is a small proportion of Michigan’s population, every vote counts in these swing states, and could be a problem for Harris. However, I think she performs well among women here and college-educated white voters that negates any major losses, and she wins the state regardless.

Georgia: Trump +1.4%

On the flip side of Michigan, Georgia is the best swing state out of the seven main ones for Trump. While I don’t generally take the polls as gospel in every other state, I do in Georgia. Every election the polls there have been extremely accurate, and this time will be no different. In comparison to 2020, less voters will turnout. But Georgia is a state that is not “swingy.” So, the theme of the race here is going to basically be “who can lose less voters” in comparison since 2020, which as a baseline, is as close to a tie as you can get, since Biden only won by 0.2%. Harris, I think, loses more voters. Turnout in Atlanta is going to decrease more (which are heavily Democratic), than the rural areas of the state (which are heavily Republican). Just by that math, Trump has a bit of an edge. What will be worth watching in Georgia will be the outlying suburbs of Atlanta, like Forsyth and Fayette counties, and if there’s a significant swing in the Asian American vote there (which is a small share of the population but has seen signs of shifting).

Arizona: Harris +0.5%

Out of all of the polling I’ve seen in this election, I am easily the most skeptical of Arizona. Out of all of the calls I’m making, this one represents the biggest “going out on a limb here” gut feeling. I don’t know what it is about the polling methods, the perceptions I have about the state, but I am not buying that Kari Lake wins the Senate race, or Trump gets this big win. I have to see it to believe it if Trump ends up winning. And yes, polling shows he’s roughly a 2-1 favorite. But there is a sizable population of Republicans who loved John McCain and hate Trump that isn’t factored here. They were the main reason he lost Arizona in 2020 and I believe could be the case again in 2024. While a lot of motivation to vote in the U.S. today is based off of negative partisanship (i.e. you vote for a Republican because you hate Democrats more), this is even more pronounced in Arizona, which have nominated a LOT of incredibly unpopular Republicans late for statewide office recently, who then simultaneously lost. I don’t put much stock in a “reverse coattails” theory personally, but Kari Lake and some of the congressmen will be a drag on the ticket. Expect all the talk about Arizona to be about Maricopa County (Pheonix). If you win Maricopa, you win Arizona.

Nevada: Harris +0.2%

Nevada sucks. I hate trying to analyze Nevada because it is a tricky state to poll and to predict because it’s population is all over the place. Literally. It has one of the highest transient worker population (workers that move around the state and out of state who take seasonal/multiple jobs). This voter group is a wild card because it’s a special demographic that’s hard to get a read on and they’re typically not engaged in politics. Not to mention, I’m still extremely frustrated at Nevada for taking SO LONG to process their votes in 2020 and they will likely take forever again. But, Nevada will be a split ticket state, I believe. In this election cycle, Trump appears to have gained ground with working-class voters of color, which make up a higher percentage of the population than the U.S. average in Nevada. In comparison to Biden winning the state by 2% in 2020, Trump is likely going to see some sort of gains among this group which I think will be enough to propel him to win. I don’t generally trust early voting data either, but Nevada’s is surprisingly accurate, and at the time of this posts writing (October 31), there appears to be a bit of a deficit that Harris needs to overcome from early votes. But, there’ll be a lot of votes to count on and after Tuesday. If the race gets down to the wire, expect Steve Kornacki or John King to be obsessing over the Washoe county suburbs (Reno + Carson City) on Wednesday/Thursday, because that will be the last of the vote to be counted and will tip the state in either direction. I think after all the late ballots are counted, Harris does just enough to win here, but this will take days to know the end result.

North Carolina: Trump +2.0%

North Carolina is, in principle, similar to Georgia. Demographically, there is high level of correlation between the two. It politically has been been somewhat of a tease to Democrats in statewide races. There has been a Democratic governor since 2016 and that trend will almost certainly continue in 2024 when Josh Stein wins by 11ish points. That is certainly a quirk in a state that has been reliably-ish Republican for over a generation. North Carolina is a state that has seen a growing population as well; it’s similar to the other Southeastern states in that regard. But, it has some important differences that distinguish it from Georgia. It has more than one major population center. Where Georgia has a pretty strict urban/rural divide, North Carolina is geographically a bit more complex. There are a lot of exurban areas that put somewhat of a wrinkle in things. Would white voters who live in these areas swing against Trump in comparison to 2020? I think that’s plausible. That’s something that could widen the margin of difference in vote share in comparison to Georgia. Harris will lose a little more voters than Trump will, but Trump will hold on to his narrow advantage (he won North Carolina in 2020 by 1.4 points) and keep the state red, even as the Republican gubernatorial candidate loses big.

Wisconsin: Trump +1.1%

While I believe that almost every poll in Georgia should be taken as gospel due to its historically high accuracy, the opposite is true in Wisconsin. Polling data in Wisconsin should be taken with a tablespoon of salt. It underestimated Trump by an average of NINE points in 2020. When it comes to underestimating Trump, as almost all polls have done the last two election cycles, Wisconsin is the prime reason why. In 2022, the polls underestimated the Democratic candidates by an average of FIVE. That’s a fourteen point swing in one election cycle so the margin of error is incredibly high here. The current consensus is that Harris has the slightest edge over Trump, and could very easily pull off a win here. Why I think Trump will win Wisconsin by +1.1% is because he will slightly overperform amongst white voters without a college degree comparison to 2020, which are groups that are harder to track with polling than other demographic group. Abortion, as an issue, will surely be very salient in Wisconsin and Harris will perform well in places like Dane County, a spot with a high proportion of college-educated white voters, but Trump gains in the Southwestern parts of the state that are heavily rural will offset that, among other factors. This prediction is without a doubt the biggest shot in the dark out of all the calls I am making.

Pennsylvania: Trump +2.3%

Last on the core swing state list is Pennsylvania. First off, if you know someone that lives in Pennsylvania, you should give them a phone call and see if they’re doing okay. Election season is full of unwanted ads and Pennsylvania is at the eye of the storm. But at the end of the day, I believe that Pennsylvania goes for Trump. And I’m getting tired of writing a unique explanation for each of these states, but my main reasons for this are: 1) level of turnout decrease (like the other swing states), and 2) the Butler assassination attempt. Like everyone else, when the first assassination attempt happened, I thought Trump had the election in the bag. Especially against Biden, that shooting led me to believe it was going to be a landslide and he’d win every swing state by a healthy margin. Since then though, I’ve walked back that theory due to a number of reasons (but mostly because of the switch from Biden to Harris). But, I still think that he wins Pennsylvania by a decently healthy margin. A lot of this is because a heavy swing toward Trump in western PA. A huge swing toward Trump amongst white voters without college degrees, low propensity voters specifically, will the primary reason why.

A Larger Overview:

Overall, my prediction is fairly in line with a lot of election models are predicting as well, with a couple of quirks. Michigan, Arizona, and Nevada being the swing states that vote for Harris is a bit of an odd prediction to make. For reference, according to Silver Bulletin, an election forecast model I follow pretty closely, that combination of states is the not the most likely outcome to happen in this election. But predictions are an partly based on data with a little bit of intuition and hopefully some luck mixed in, and I feel confident in the fact that my arguments have merit to them, even if some calls wind up being incorrect in the end. The intuition part boils down to this condensed argument:

Elections come down to a relatively uniform swing that moves with the country as candidates win. The country is going to swing nationally about 4 points to the right I think, and a lot of the swing states are going to shift to the right by 2ish points to reflect that. The reason that it won’t swing by 4 is Trump is going to do REALLY well (in comparison to 2020 and 2016, even though it means he still loses by 20 points) in California and New York. That shift will mean the popular vote is going to be very close, and Harris still wins it, but Trump wins the Electoral College. I think Arizona, Wisconsin, and Michigan are going to be the exceptions to the rule, however. Arizona has progressively moved to the left in the past two decades. Even though I predict Harris’s Arizona victory margin to be nearly identical to Biden’s victory four years ago, in actuality, that is a big swing to the left because the national environment moved to the right, and would be in line with the rest of the nation.

I have a somewhat similar prediction for Michigan, although it’s staying roughly neutral in comparison to 2020. In an election, every state is susceptible to the national environment and will inevitably move around in terms of each percentages. But at this point, the country is so polarized that we only focus on the swing states because those shifts are getting smaller and a lot of states hardly ever switch. These small variations are why I’m predicting those three states are to be the exceptions to the rule, although North Carolina or Georgia could be the states that do it too.

A few, other states I’ll mention quickly before I move on to the congressional races:

New Hampshire: Will only be in play if Trump has a REALLY good night. It’s possible for it to go red.

Iowa: There was a big shock poll that had Harris winning by three just a couple of days ago. That was… really strange. I still think it goes for Trump by 6ish. But a good indicator of a potential swing to Harris.

New Mexico: Maybe Latino voters turn out huge for Trump and he wins? It’s possible. If this happens though, Arizona is guaranteed voting for Trump too.

Nebraska’s 2nd Congressional District: “The Blue Dot.” Harris probably. Not a lot of indication it’s swinging toward Trump.

Minnesota & Virginia: Possible it goes for Trump, but a lot of things would have to go right.

Texas: I believe the polls are underestimating Harris here, not Trump. But not by nearly enough to give her a realistic shot of flipping the Lone Star state blue in ’24.

Florida: Has a good shot of being farther to the right than Texas.

Alaska: There has not been nearly enough attention given to Alaska. Ranked choice voting has made it way more unpredictable since 2022. The last frontier could potentially throw a blue wrench in the works for Trump.

Maine: Maine as a whole state goes red? Weird. It would take a lot to convince me, but anything’s possible!

If there are any other races that you are wondering about that I didn’t put, it’s because the scenario is so unlikely that I’m not considering the possibility and I would be shocked if it were to vote against what it is expected.

Senate Races:

The Senate & House races are of equal importance to the presidential race, as congressional majorities are so slim these days that one seat can quite literally be a deal breaker in some cases. However, there’s a lot less to explain here besides the three Senate races that will flip, so I’m not going to near in-depth with all of them.

For the rest of the Senate races that aren’t West Virginia, Ohio, and Montana, the incumbent will be re-elected. Even though it’s a smaller advantage than it used to be, incumbency still matters in the Senate/House. That 2-3 point boost, on average, will be enough to keep Democrats alive in Wisconsin, Pennsylvania, and Nevada. The Democrats have ran decent (Michigan) to fantastic (Arizona) campaigns elsewhere where I believe the new candidate will outperform Harris in each of these states. Texas will be relatively close (a lot of people despise Ted Cruz, myself included), but he’ll ride Trump’s coattails enough to win re-election. It could flip, but it isn’t likely. Same with Florida. Also with Nebraska, oddly enough. Alright, let’s talk about the three Senate races that matter, and Democrats start with 51 seats:

West Virginia: Democrat-turned-Independent Joe Manchin is retiring and the Republican candidate is up by 50 points. Republican pick up. 50 seats-50 seats.

Montana: Democrat Joe Tester has held on in Montana, but it’s not looking great for him here. I like to compare trying to win a Senate race like an actual race, so bear with me on this analogy for a second. The state’s partisan lean (i.e. how many Democrats and Republicans there are roughly in each state) determines how much of an advantage/disadvantage you have, especially when you’re running in a presidential election year. In this analogy, Tester and Sheehy are running a mile long race against each other, but Sheehy gets a full lap of a head start and Tester has to run the race carrying a 35 pound weight. EVEN IF Tester had been training for a long time and was an Olympic-level athlete, the disadvantage is so overwhelming that he’s going to lose the race and there’s nothing he can do to reasonably catch up. Sheehy wins by about 10 points, just for having a huge advantage built in for him. 51 Republican seats.

Ohio: This is the only prediction I have switched since I wrote it down a week or so ago. If Tester was down by a full lap in Montana, Sherrod Brown is down by about half of that and carrying a 10 pound weight in the Ohio race. The fundamentals are not on his side, but his race is a bit more winnable, and Brown is the political equivalent of an Olympic runner. This race could be very close and Brown could train really hard to win this race, but it would also take every single thing to go right for him to win under these circumstances. If one thing were to go wrong, like slipping at the last minute or a sudden change in the weather, he’ll lose his chance. While it is doable, I believe the fundamentals are too much to overcome and he loses re-election to Bernie Moreno by a point. 52 Republican seats.

The House

At first glance, I can understand how my House prediction doesn’t make sense. In an election where I think Trump wins, Democrats winning the House while losing the popular vote doesn’t seem plausible, but hear me out. Republicans in the House, on average, will always do better than the presidential candidate in the popular vote by about 2-3 points. I think Republicans will wind up winning the popular vote here by a total of 1.7%, while Democrats win 220 House Seats. The main reason is that more Republicans are running uncontested or against an Independent than there are Democrats running uncontested or against an Independent. In most of these cases, it’s because there’s no way the other party would have a shot, but that 25-40% that would go to the losing candidate is suddenly null and void. For each House race, this roughly translates to like 0.08% of the total national popular vote but that adds up. And in a lot of close races, I think Democrats will do just enough to win and scrape together a majority.

I don’t have the margins of victory on any of the House races, so I’m just putting them into three categories: 1) seats that flip, 2) seats worth keeping an eye on, in my opinion, and 3) the rest which I do not care about.

Seats that flip:

OR-05 (R to D)

CA-27 (R to D)

AZ-06 (R to D)

AK-AL (D to R)

NE-02 (R to D)

TX-34 (D to R)

FL-13 (R to D)

VA-02 (R to D)

NY-04 (R to D)

NY-17 (R to D)

NY-22 (R to D)

There are a few seats that will be pickups for both parties due to mid-decade redistricting as well

Seats worth watching that could very well flip:

WA-3, CA-13, 22, 47, 41

AZ-1, NM-2, CO-8, MT-1, IA-1, WI-3, MI-7 & 8, OH- 1 & 13, PA- 7, 8, 10, 17, NY – 19, ME- 2, NJ-7, TX-28

Thanks for following along! Hopefully I’m right on a lot of these or close to it. I will likely have a results post where I go through my calls and how off I was on each of them. And if you haven’t voted yet, please do! You have the right to vote, use your voice and make it heard!

Signing off,

Colby