Election Night Guide: Alabama through Arkansas

Which one of these men will be president?

All right, let's do this thing.  I will start out by saying that, for the first time in four years, I went back and reread what I wrote about the presidential race in 2016.  I remembered predicting Hillary Clinton-I didn't remember the lack of confidence, which was clear in certain races I wrote about that year (I hadn't remembered that I'd guessed correctly that Florida would go to Trump, and in fact I only got wrong the three states that everyone got wrong)-I assumed, based on my recapping, inaccurately, that I was being too conservative in predicting Clinton would lose so many races, when in reality I was predicting near her ceiling.

I say this because, obviously, elections predictions are scary, and while you have to read the data, you can be wrong.  I've done this every federal election cycle since 2004, and some years I was right enough that I didn't need to even look at the election night results (2014 midterms I called every Senate race), but I will admit that in 2016 I didn't see the Trump win coming (the only presidential race I've gotten wrong), and as a result I've got something to prove this year.

For anyone who is new to these recaps (welcome back anyone who has put up with my long-windedness before), there are a few notes I want to address straight-out:

1. I predict in every presidential, gubernatorial, and congressional race.  No one can vote Tossup (well, you kind of can in Nevada, but that's a conversation for a different day), and so unlike a lot of other political sites you'll read, I will make calls in races that are genuine coin flips.  Every year, no matter what, there are races that you just can't quite figure out the dynamics of the race or that are going to be so close that calling it confidently is impossible, and that's certainly true in 2020.  I went with polling data, historical elections, demographic changes, insights from experts, and gut instinct to make my predictions, but I promise you there will be races that I will get wrong below, and am just going with what I know.

2. These are written a week before the election (if you want exact dates, I finalized on October 25th).  If something truly bizarre happens in the next week, I won't be able to catch it.  That usually doesn't happen, but I'm putting it out there right now.  October surprises that might matter include any of the major candidates (for president or for the Senate) getting Covid, a surprise Supreme Court vacancy, and of course any international incident or scandal involving the candidates.  I will also not be able to catch any late-breaking polling change; this doesn't happen very often either (and in 2020 the polls have been remarkably stable, particularly at the presidential level), but it does on occasion.  The most notable example of this is 2016, when late-breaking polls showed that Hillary Clinton was losing ground in Wisconsin, perhaps the biggest alarm to her campaign, and one that I didn't catch when I wrote this four years ago.

3. Early voting is a brilliant idea (it's something that you should all do if you can, as you don't know if you'll wake up on Election Day feeling under-the-weather and have to make a difficult decision about whether to go to the polls and risk exposing potentially dozens of people to what's causing you to be under-the-weather (which, if it's Covid, obviously could be serious for them even if it's something that jut feels like a cold to you), or not voting in a pivotal presidential race).  However, because of President Trump's criticisms of mail-in balloting and early voting, I would strongly caution against reading into early voting metrics, especially those that favor Democrats.  It is extremely likely that Republicans will vote in large numbers on Election Day, and as a result we won't know how much they have countered (or perhaps beaten) those Democratic numbers until Election Night.  These numbers are amazing to watch from a statistics perspective, but over-interpreting them to see who will win is a perilous journey, so proceed with caution.

4. I voted already (I took the above advice), and am very proud of my political beliefs.  I will not pretend that there isn't a party I'm cheering for on November 3rd.  I try as hard as I might to not let that influence these predictions, however.  I have had years where I was too kind to the Democrats (2016 being a big one), other years where I was too kind to the Republicans (2008 comes to mind).  So below is a prediction of what I think will happen based on data & my own interpretation-I try really hard to not editorialize too much, but in an article this long, it's inevitable.  Wanted to put that plug out there ahead of time, especially if we disagree politically.

5. Four years ago, the polling industry took a severe hit.  There were polls that were definitely wrong (Michigan comes to mind, and certainly the Wisconsin Senate race)-there were other polls that we were ignoring in order to make the night fit better (Pennsylvania & Wisconsin presidential races).  Polling in 2018 did a much better job, and most polls I've seen have tried to correct some of the errors (specifically around education-levels) that we saw in 2016.  I feel better about the polls this year, but I'm not going to totally defend them before November 3rd, and I get if you're apprehensive about putting too much stock in them.  I do think that if Trump wins, we'll be in a situation where political polling may become something of a dinosaur, though, as they have painted a much more consistent margin for Biden this cycle than was ever the case for Hillary Clinton, and as a result a win for Trump would be harder to square with their data.

6. The last thing I want to plug before I begin is how to watch Election Night.  There's been a lot of concern around how it may take weeks to know who will win, and that's possible, particularly if one of the swing states where they need to be postmarked by election day rather than received is the tipping point state (Minnesota, Nevada, and Ohio all fall into this bucket).  However, we should have a very good guess of the presidency, at the very least, on election night.  Some swing states are notoriously slow at counting ballots (Pennsylvania & Arizona, I'm looking at you), but others are quite swift (North Carolina & Florida).  We may not know the end margin of who will win on election night (that's certain-one state will take a few days at least), but unless we have an election that comes down to a single state like in 2000, it's probable we'll know the victor of the presidency on election night.  The Senate, because it's more likely that it comes down to one state, could take longer, but I also think we'll have a decent idea on Election Day of how the vote will go, unless the Georgia runoffs are the key to the Senate majority (we'll get into that more when we talk about Georgia).

With that, we're going to dive in-states are discussed alphabetically, so we'll start with Alabama!

(Note: I'll be doing commentary on every race for President, Governor, & Senate regardless of level-of-competitiveness.  I'll only do mentions for the House if I assume it'll be competitive in some way-if the House race isn't listed, I'm assuming an easy hold for the incumbent party)

Sen. Doug Jones (D-AL)

Alabama

President: Alphabetically, we start off with an easy victory for Donald Trump, who has shown a resilience amongst white voters in the South that has not been the case in the rest of the country (according to polling).
Senate: Doug Jones' victory in 2017 was a political earthquake, and likely saved the Affordable Care Act (Sen. McConnell would've had the votes when Jon Kyl replaced John McCain for a few months before the 2018 midterms to overturn it were it not for Jones), and he'll likely be rewarded for that defense of arguably the most recognizable achievement of the Obama-Biden administration by joining a future Biden administration (as Attorney General) if Biden wins.  But Jones victory was a fluke, and it won't be duplicated.  Jones could outrun Biden by 15-points in Alabama & still lose to Tommy Tuberville, and if polls bear out, he won't be able to do even that. R+1

Dr. Al Gross (I-AK)

Alaska

President: Alaska has shown some willingness this year to look at the Democrats, and Biden will do better than Hillary Clinton did four years ago (a refrain you're going to hear a lot over the next week), but I don't think it's going to be enough.  There's evidence of a landslide, but if Biden can't put away Georgia & Ohio, he's not going to get a surprise in Alaska.

Senate: Of the three competitive races in Alaska, I think the one that would be most likely to flip is the Senate seat.  Sen. Dan Sullivan (R) hasn't made enough of a mark in his first-term, and this is a state that loves gutsy independents, which his opponent Al Gross (I) is (he is running as an independent, but has stated that he'll caucus with the Democrats).  Polling shows a tight race, and this is one of seven races where the Democrats have potential, even if the polling doesn't show that they can win yet without the polls being slightly off.  I'm not ready to start calling upsets this early in the game, and I think it'll be Sullivan as Alaska polls oftentimes underestimate Republicans.

House: Every two years Democrats come to the well and insist this is the year they can finally beat Rep. Don Young, the longest-serving person currently in Congress (he's been there since 1973), but it never works out.  Alyse Galvin (another Independent who has promised to caucus with the Democrats) has shown some strong polls, but Young always out-performs polls since Alaska is so difficult to survey (particularly Alaska's large Native American population, with whom Young has consistently done extremely well).  Young wins again, and makes it to 50 years in Congress.  If Gross or Galvin do win, the strategy of running an Independent who will caucus with your party could become a popular trend in other states where one of the two parties enjoys a bad image with voters.

Sen. Martha McSally (R-AZ)

Arizona

President: The first of our "Big 6" states-the six states (NC/FL/MI/WI/PA/AZ) likely to decide who reaches 270 electoral votes.  Arizona was, just a few years ago, a stalwart red state, but a huge swing to the left in the state's populous Maricopa County delivered Democrats their first Senate victory in thirty years in 2018, and is what has helped keep their state largely in the Biden column when it comes to polling.  Arizona is notoriously slow in counting votes, so if it comes down to Arizona don't expect to get a result on Election Day, but the polling evidence has been overwhelming all year-it shows Biden up by varying degrees, and while Trump could over-perform that (and if he's going to win, he might have to), there's no evidence that he will other than Republicans' historically doing well here.  My money is this is a flip. D+1

Senate: A much easier call happens in the Senate race.  Sen. Martha McSally lost the 2018 race to now-Sen. Kyrsten Sinema, but due to Jon Kyl's resignation McSally was appointed to this seat anyway.  This ended up being a foolish move by the Republicans, who assumed McSally's loss in 2018 was just due to Sinema being a good candidate, but it turns out that McSally is not popular statewide, and many voters didn't like the way she got their Senate seat after being rejected.  It doesn't help her that Mark Kelly (D), an astronaut & the husband of former Rep. Gabrielle Giffords, has run a near-perfect campaign this cycle.  Consider this the quick volley back from the Democrats after losing Alabama. D+1

House: The House, because it's unlikely to flip, has not attracted the same level of attention as the Senate & presidency this cycle.  But if you're looking for a trend throughout this analysis, the biggest question is-can the Democrats complete the suburban transformation that they started in 2018 (which led to their winning back the majority)?  The first suburban seat they can make a play at is in Arizona-6, a district that includes the sprawling northeastern suburbs of Phoenix.  This is an historically Republican seat (Mitt Romney won it by over 20-points), but Trump cut that in half, and Kyrsten Sinema lost it by only three.  Rep. David Schweikert (R) has been dogged by scandal all year, and had to pay fines in July due to misuse of campaign contributions, and he's running against a doctor during a pandemic, Hiral Tipirneni, who has walloped him in fundraising.  I don't think it'll be by much (again, just a few years ago a win here would've been unthinkable for a Democrat), but my gut says that we'll see the Democrats score three big pickups in the Grand Canyon State on November 3rd. D+1

Arkansas

President: Arkansas gave their former First Lady a 27-point loss in 2016-I seriously doubt they do much better for Biden.

Senate: You can't win if you don't play.  Despite being a frontrunner for the GOP nomination in 2024, Sen. Tom Cotton faces no Democratic opposition and will win a second term.

House: Democratic State Sen. Joyce Elliott has run a helluva race here, making national Republicans spend money on a seat I wouldn't have even mentioned in this article six months ago.  But Elliott is still running in a district that'll be tough for Biden to win, and while I could see her getting close, I just can't picture Rep. French Hill (R) losing here, so I'm going with the Republicans.  I will say that if you notice me starting to be more wrong in predicting too many Republicans, this is the sort of district that could surprise since it does encompass a large metropolitan area (Little Rock & its surrounding suburbs).