Election Night (or Week, or Month…) Viewing Guide
The 2020 United States presidential election is almost here! While it will probably take longer than usual to know who won thanks to COVID-19 and the accompanying surge in voting by mail, we will likely have a pretty good idea of what is happening by late Tuesday night or early Wednesday morning. In a very close election, results could take considerably longer — perhaps as long as a few weeks. As fate (or 2020) would have it, the single most pivotal state (Pennsylvania) may be one of the slowest to report results. To use a boxing analogy, there are plenty of early opportunities for Biden to deliver the KO to Trump, but if Trump can survive the first few rounds, the fight has potential to go the distance. Still, Biden is a heavy favorite and Trump is a considerable underdog.
“But what about 2016? What if the polls are wrong again?”
In 2016, 538 gave Trump a 28% of winning and estimated he would lose the popular vote by 3.5 points (actual result: he lost by 2.1 points). As things currently stand, Trump’s odds have worsened considerably. 538 gives him just a 10% chance of winning and projects he will lose the popular vote by 8.0% — the largest margin since 1996. There are two main points here: first, Biden’s lead is considerably larger than Clinton’s lead was, which gives him far more room for error. Not only would the polls have to be wrong (unlikely) and wrong in Trump’s favor (unlikely), they would have to be significantly more wrong (you guessed it, unlikely!). With the current size of Biden’s lead, even if the polls were off by exactly as much as they were in 2016 and in the exact same direction, Biden would still win somewhat comfortably. Remember, no side has a monopoly on polling error. In 2008 and 2012, Obama outperformed polls in many states. While polls could miss in Trump’s direction again, it is equally likely they miss in Biden’s direction and we see the biggest blowout election since 1984. Second, there is a specific, structural mistake that accounted for a lot of the polling error in 2016 but has since been corrected. If you’re interested in learning more about it (how polls failed to weight for education level, which led to them underestimating turnout for white voters without a college degree) I would recommend this article. To keep this post at a somewhat normal length, I won’t get into it now. But this much is clear: while Biden is a heavy favorite, Trump still has a small shot. I’ll now take you through election night chronologically, giving you an idea of what to expect and how to be a particularly savvy election viewer.
7–8:00 PM EST
Polls closing in: Alabama, Georgia, Indiana, Kentucky, North Carolina, Ohio, South Carolina, Vermont, Virginia, and West Virginia
Races to watch: Georgia, North Carolina, Ohio, South Carolina, Indiana
If you’re relatively new to watching elections, you might ask yourself: “How did they call some of those states so quickly? They haven’t even counted votes yet!” This is based on a combination of prior knowledge and exit polls. For example, I can almost guarantee every single decision desk will immediately call Vermont for Biden and West Virginia for Trump. That will be due to the fact we know (based on voting history, demographics, polls) both states are extremely blue and red respectively, but it will also be due to exit polls taken over the course of the day. In general, decision desks need to be 99.5% sure of a result before they call it, so when a projection is made so quickly, you can be pretty sure the final result will not be close.
To be clear, South Carolina and Indiana are not swing states — at least not really (Biden has a 9% and 4% chance in those states respectively). But that doesn’t mean we can’t learn anything from them. On election night in 2016, decision desks were able to immediately project Indiana for Trump. If they are unable to do so this time around, it might be due to extra caution about the unique voting circumstances in this election. But it also might portend doom for Trump in other states where he will rely on similar demographics for support. None of the swing states in this slot are expected to report votes very quickly, so it is unlikely we will learn much in the opening minutes of the night. Pay special attention to North Carolina, though, where they should be able to report all of the early and mail votes shortly after polls close at 7:30. Note that these will skew heavily toward Biden, while the remaining in-person vote will take longer to count and will tilt things back toward Trump.
Go to bed if: Biden wins Ohio, Georgia, or North Carolina (if Biden wins any of these states, his win probability rises above 99%)
Don’t overreact if: Early results in North Carolina show Biden with a big lead (it will decline significantly and may even flip over the course of the night, depending on actual results)
8–9:00 PM EST
Polls closing in: Arkansas, Connecticut, Delaware, DC, Florida, Illinois, Maine, Maryland, Massachusetts, Mississippi, Missouri, New Hampshire, New Jersey, Oklahoma, Pennsylvania, Rhode Island, Tennessee
Races to watch: Florida, Maine, Pennsylvania
This will be the busiest hour of the night in terms of polls closing. This group of states includes many that are highly partisan and likely to be called immediately, but it also includes some of the most important swing states. To be clear, this election hinges very much on Pennsylvania; if Biden wins it, his chances of victory soar to 99%. If Trump wins it, his chances of winning the election rise from 10% to 61%. Polls in Pennsylvania have been pretty consistently in the good-not-great region for Biden. As such, 538 currently gives Biden an 86% chance to win the state. Pennsylvania is expected to count votes rather slowly (since they cannot start counting until election day, per state law) and the early results will skew heavily toward Trump. As a result, it is extremely likely that Trump will be “ahead” in Pennsylvania on election night, but that his lead will shrink and ultimately vanish once all early and mail votes are counted. Simply put: don’t expect Pennsylvania to be called on Tuesday, and if things get closer than expected, it could drag on for days after polls have closed.
Florida is likely Biden’s best opportunity to all but wrap things up on election night. It falls into the same bucket as Ohio, Georgia, and North Carolina — it is by no means a must-win for Biden, but it is a must-win for Trump. 538 currently has Biden with a 68% chance to win in Florida, and if he does win Florida, his odds of victory will exceed 99%. That said, if Trump does hold on in Florida, his chances would increase from 10% to 32% — or slightly better than his odds from 2016. Either way, the results from Florida should tell us a lot about where the race stands, and unless the outcome is very close, there is a good chance we will know the outcome on Tuesday.
Maine is not a particularly interesting election as a whole, but it is one of two states (along with Nebraska) that does not award all of its electoral votes to the winner of the state. Maine will award 2 electoral votes to the statewide winner (likely Biden) and each of the other votes to the winner of each of its two congressional districts. While Maine’s 1st is a strong blue district, the 2nd is a tossup. One electoral vote does not count for much, but it will give us a good hint about how Biden may be performing elsewhere (Maine’s 2nd has similar demographics to some important states in the Midwest) and would also likely preclude the nightmare 269–269 tie scenario where Biden only flips Wisconsin, Michigan, and Arizona.
Go to bed if: Biden wins Florida
Don’t overreact if: Early Pennsylvania results show Trump with a big lead
9–10:00 PM EST
Polls closing in: Arizona, Colorado, Kansas, Louisiana, Michigan, Minnesota, Nebraska, New Mexico, New York, North Dakota, South Dakota, Texas, Wisconsin, and Wyoming
Races to watch: Arizona, Michigan, Nebraska, Texas, Wisconsin
Perhaps the sexiest race of the night is Texas: can Democrats flip those 38 electoral votes blue for the first time since 1976? Maybe — 538 gives Biden a 39% chance of doing so. But it is almost beside the point, as Texas does not figure to play a pivotal role in deciding the winner; it would really be more like adding insult to injury for Trump. Put differently, if Biden wins Texas, it means he has probably won nearly all of the swing states and earned a true blowout victory over Trump.
Wisconsin and Michigan were the two most surprising results in 2016, but both figure to be back in the blue with room to spare this time around (Biden has a 94% and 95% chance in those states respectively). If Trump is able to pull off a true shocker in either one of those states, he would become the favorite to win the election; if he wins both, there is likely no remaining path for Biden. That said, polling has been relatively consistent and not particularly close in these two states. Because Wisconsin is not accepting any votes in the mail after Tuesday, they will likely have all results counted by Wednesday morning. In a big enough blowout, we could plausibly see a Wisconsin call Tuesday night.
Arizona and Nebraska’s 2nd congressional district are the best “Plan B” for the Biden campaign if they lose Pennsylvania. Biden is a moderate favorite in both (70% and 74% respectively) and any polling error is unlikely to be correlated with polling error in Pennsylvania. What I mean by that is if someone was able to tell you with 100% certainty that Trump was going to win Pennsylvania, that would tell you a lot about Trump’s chances in Wisconsin and Michigan because all three are demographically and geographically similar states. The same isn’t really true for the relationship between Pennsylvania and Arizona, so Arizona is a reasonably diversified backup plan for Biden — he could very reasonably still win one while losing the other.
Go to bed if: Biden wins Texas OR Biden wins Arizona and Nebraska’s 2nd
Don’t overreact if: Michigan or Arizona show big early leads for Trump or Texas shows a big early lead for Biden; these will mostly be due to the order in which different kinds of votes are counted in these states
10:00-? PM EST
Polls closing in: Iowa, Montana, Nevada, Utah, California (11 PM), Idaho (11 PM), Oregon (11 PM), Washington (11 PM), Hawaii (12 AM), Alaska (1 AM)
Races to watch: Iowa, Nevada
None of these races will be nearly as important as ones from earlier, so you’ll still mainly want to keep an eye on those vote totals. Still, there are a couple interesting races ending later in the night. If Biden appears to be having a better-than-expected night, he could manage to flip Iowa. Nevada could be interesting in the opposite direction because most polls this cycle actually show Trump doing a bit better among Hispanic voters than he did in 2016. If Trump is having a stronger night than expected, Nevada could be close again just like it was 4 years ago. Still, because neither state figures to count all their votes very quickly, neither development would be groundbreaking at this point in the night. A Florida call may come around this time. Either way, by 10 PM, I would imagine we will have a pretty good idea whether Biden has won emphatically like expected or if Trump managed to improve upon his standings in the polls enough to make things interesting.
Go to bed if: Biden wins Iowa
Don’t overreact if: Many of the swing states remain uncalled. Major decision desks are typically very conservative in making these kinds of calls, and perhaps they will be even more conservative with all the added procedural question marks this time around. While we’d likely hear about a Biden blowout at this point in the night, there is still a distinct possibility that Biden wins by a comfortable margin in the end even if things look uncertain late on Tuesday.
To summarize:
Biden is a heavy favorite and Trump is a big underdog, but Trump still has a legitimate (albeit narrow) path to victory. His most realistic path to victory is if Biden only manages to flip Wisconsin and Michigan — in that case, Trump would win with 280 electoral votes. However, that means he must win all the other swing states — and that includes states where he has a small lead (Texas), states where he is effectively tied (Ohio), and states where he is trailing (Florida, North Carolina, Arizona, and Pennsylvania). Biden could all but lock things up early in the night if he wins Florida, or later in the night if he wins North Carolina. But if those states are too close to call or Trump wins them, we could be in for a long night… or a long week.
My Final Ratings:
My Picks:
