The Odds of an Electoral Tie in 2020 are Really, Really High
It’s probably too early in the election cycle to be doing this sort of thing, but I don’t think I’ve seen anyone in the political media point out how shockingly close we are to an electoral tie in the 2020 election. Some variables are likely to change when the Democrats nominate a candidate, of course, but if the map remains the same and President Trump loses electoral votes in the three places he is most likely to lose, it would lead to a perfect 269–269 split in the electoral college. Moreover, the probability gap between the Democrat winning the third most likely pickup and the all-important 4th is not particularly close, meaning that the tie scenario has a relatively wide margin for error.
To sum up: in addition to preventing Trump from expanding the 2016 map, Democrats would need to win in Pennsylvania, Michigan and Maine’s second district to secure the tie. In all three places, I think the Democrats are heavy favorites, whether the nominee is Biden, Sanders or Warren.
Pennsylvania
If the Democrats are going to win any states lost by Hillary Clinton, the most likely state will be Pennsylvania. Referred to as Clinton’s “firewall,” Trump won the state by doing well enough in the suburbs, suppressing turnout in urban areas, and activating a certain kind of populist enthusiasm in rural areas, especially in the coal region of Appalachia. The coalition is unlikely to hold, however, as Trump’s support among suburban women has plummeted and all three Democratic frontrunners appeal better to disaffected rural white voters than Hillary Clinton, who largely ignored them in favor of turnout operations in the cities. Born in Scranton, Joe Biden’s blue collar style and call for a return to traditional Democratic labor politics should bring back a number of the Obama/Trump voters that made such a difference here in 2016. Similarly, Sanders’ or Warren’s left wing populism would appeal to disaffected whites who feel Trump has not delivered on his promise to rebuild their communities. If Trump loses Pennsylvania by less than 5 points, he is in for a great night.
Michigan
Michigan was the closest contest of 2016, wherein Trump won the state by less than 11,000 votes. Natural demographic change alone should be enough to make up the difference here, but if it isn’t, there are plenty of reasons for Democrats to be bullish about winning Michigan. Of all the surprising losses in the last election, Michigan was by far the most preventable. While Trump’s team silently built a grassroots campaign, Clinton refrained from investing time or resources into the state. Democratic voters, too, didn’t vote with the enthusiasm of what they failed to realize was a swing state. More voted 3rd party in Michigan than the national average and since there was no Senate race on the ballot a lot of people stayed home. That shouldn’t happen again in 2020, as a Democratic candidate will not take Michigan for granted. Also, Trump’s trade war has softened his support among agricultural communities in the midwest. Farmers will still support him, of course, but not by the same margins he enjoyed in 2016. Michigan shouldn’t be close.
Maine’s Second District
Maine is one of two states in the union that splits their electoral votes (the other is Nebraska, but Trump shouldn’t have a problem winning there). In 2016, Clinton won the larger 1st district with it’s 3 electoral votes but lost the single vote in the 2nd. I’ve yet to see any 2020 polling data that examines the split between the 1st and 2nd district, but based on what I have seen of statewide job approval and head to head numbers, I don’t think Trump will be in a strong position to defend the singular vote. This is in an area of the country where Trump’s nationalist rhetoric is least effective and there aren’t any proximate swing states to make crafting a regional economic message worthwhile. More likely, the internal polling will scare away the Trump campaign and they will write it off to focus on higher priorities. It is only one vote, after all…
For those interested in the data that I referenced, Clinton won the entirety of Maine by less than 3 percentage points (47.83–44.87). She took the 1st district by 15 and lost the 2nd by 10. The three pollsters that poll statewide presidential job approval (Gallup, Civiqs and Morning Consult) all show net disapproval for President Trump over 20 in Maine, and a trend that we have seen nationally has shown that job approval is always slightly higher than the percentage willing to vote for Trump. Again, without a split, there is no way to determine how much of that disapproval is coming from the 2nd district, but it is hard to overcome a margin that big. For comparison, in 2012, Obama won the state by 15 (56.27–40.98) and won the 2nd district by 8. All three pollsters would need to be off by about 10 points to even make this race competitive.
I would put the odds of the Democrat winning all 3 of these contests at about 70%. Oddsmakers seem to agree that Trump has an overall 60% chance of losing outright, meaning that there is a 10% chance of the Democrat winning enough votes to tie but not enough to win. This math excludes scenarios where Trump expands the map, because it is unlikely that he can do so while still losing PA, MI and ME2. I will look at that, as well as the Democrat’s overall likelihood to pick up more votes.
Trump Expanding the Map
Recently, Trump’s campaign team leaked plans to expand the map and compete in places like NH, VA, MN, CO, NV, and OR. Some journalists are characterizing the move as an act of desperation and others are saying this is just a natural function of a campaign that has more money than they know what to do with. There are probably elements of truth to both interpretations, but more likely it is a diversion to hide how invested they are in winning the midwest. Since Trump won all of the states currently in play and can’t afford to lose more than two of them, claiming that such a thing is possible seems inevitable, even if nobody takes it very seriously.
Trump’s favorability rating in New Hampshire is abysmal, as his support in the northeast has fallen harder than anywhere else in the country. With only three electoral votes, it really doesn’t make sense to invest money here as the issues that are important to New Hampshire voters (healthcare, environment and fighting the opioid epidemic) do not play into President Trump’s strongest hand. Some speculate that a series of embarrassments among Democratic leaders in Virginia’s state government has created an opening for the Republican party to make headways, but I have seen no evidence in polling or state elections that suggests that even a moderate Republican could compete statewide in a place that becomes suburban DC a little bit more every year. The mountain west is also difficult terrain for Trump as demographic change moves quickly in Colorado and Nevada. In both states, young people are moving into the cities while rural communities shrink. The trend is even worse for Trump in Oregon.
Of all the states on the list that Trump would like to bring into play, I think Minnesota is probably the most interesting. If Trump overperforms with rural whites in the midwest, if his tariffs don’t lose him support in agriculture and he somehow reverses his trend in the suburbs and among minority populations, he could win here. But he couldn’t win Minnesota while losing Michigan.
Trump Losing Outright
There are obviously plenty of scenarios where the Democrat wins more than 270 electoral votes and this outcome has the most room to run. An amazing night for the Democratic candidate could win as many as 412 electoral votes and I think a median outcome is probably in the neighborhood of 300. Again, though, I am more interested in a scenario where Trump performs slightly better than expected, but still does not win. The following is the state of play for the rest of the field.
Beyond the three places already mentioned, the only state that is better than a coin flip for the Democrats is Wisconsin. Both sides are aware of this and will no doubt pump tens of millions of dollars worth of advertising and campaign infrastructure into the state. Polling and demographic trends should be encouraging to the Democrats, but they have to be disappointed in the 2018 midterms where big leads in the polls translated to only a 1 point victory for the the Democratic gubernatorial candidate. If polling is wrong (again), Trump’s brand of economic populism plays as well in Wisconsin as anywhere in the country. And the demographic composition of the state is not as friendly for the Democrats as Pennsylvania or Michigan. A better than expected performance from Trump could win him Wisconsin pretty easily.
There are plenty of other states the Democrats can win that are not necessarily tied to success in the midwest. A Democrat (especially Kamala Harris) can win in North Carolina, Florida or even Arizona and lose in Wisconsin. Political messaging plays differently in different parts of the country, which accounts for the differential between the likelihood of winning Wisconsin (around 52%) and the overall chance of a Democrat victory at 60%. But it isn’t terribly likely. I would consider Trump’s advantage in FL, AZ, and NC around 55%. Ohio and Iowa are a bit higher at 60%. A Democratic blowout could deliver Texas and Georgia.
What does a tie mean, anyway? There are provisions in the Constitution, of course, but the framers probably didn’t imagine that the country could ever be as politically polarized as it is today. The first place the election could be decided would be in the electoral college itself. As we all know, the electoral college is a crock of shit, and one of the reasons is that a state’s electors do not necessarily have to vote for the candidate who won the state. Most historical examples of a “faithless elector” involve an elector making an inconsequential mistake, are a result of the confusion surrounding a candidate’s death or are protests against both parties or the electoral college itself. There are no examples of a faithless elector voting for the pledged candidate’s opponent and such a thing wouldn’t go over well. Legal processes exist for invalidating these votes and deciding an election in this way would no doubt lead to an unending and contentious court battle. But so would the the only other outcome.
Faithless electors refusing to vote or voting for a non-candidate would not be an issue in this scenario, because a candidate needs 270 to win and a 269–268 outcome (for instance) wouldn’t qualify. Once the electoral college records that neither candidate reached 270, the decision goes to the House of Representatives. I have’t been able to find whether it is the incoming or outgoing class that would chose the President in the event of a tie, but it actually doesn’t matter. Rather than each member of the House having one vote, each state’s delegation gets a vote. In other words, the singular vote from Wyoming’s at-large representative is worth the 53 votes in California. If representatives vote along party lines, Trump wins easily.
So what we will have is a situation where the Democrat has won the popular vote (probably by around 5%) and won the House of Representatives, yet the House ultimately determines that Trump has won the election. Anyone assuming that Trump will do the magnanimous thing and concede (lol) or that Republicans will put the good of the country beyond their own political advantage is in for a surprise. Similarly, the chances that Democrats submit to four more years of Trump as a result of what is objectively an unfair process are pretty close to zero. Both sides will accuse the other of fraud and turn to the courts for resolution. Violence will probably erupt on the street and the country will be without a nationally recognized head of state for as long as it takes to litigate.
The only two times that the House of Representatives has elected a president were in the early 1800s. Neither of which had the sort of social climate that we have today. A better comparison would probably be the election of 1860, in which Abraham Lincoln won with less than 40% of the vote through what his opponents credibly called a rigged process. Under the Pennsylvania, Michigan and Maine-2 only scenario, Trump would have slightly more of the vote share, though with fewer plausible candidates on the ballot. Anyhow, let’s hope that it goes better this time around.
