We in the US are entering a very interesting period: “Election Day” is now only five days away, but there seems to be little chance that the night of the election will bring any sense of finality the way it has traditionally (2000 aside) — not with uncertainties about when mail-in and early votes will be counted in which states, compounded with uncertainties about the potential transition of power. I don’t know if we’re entering “Election Week”, or “Election Month”, or “Election Quarter”.

As such I thought it would be a suitable time to start recording my perspectives on these events as they unfold, to have something to look back upon. Since this is the first post in a planned series, I need to start by sharing the context of our times.

The generally accepted wisdom is that Biden is a heavy favorite to win the Presidency; not a prohibitive favorite like Reagan in 1984, but certainly in a better position than Clinton was against Trump four years earlier. On average, national polls show Biden with a 9-point lead, although there are outlier polls; for instance Rasmussen, which has consistently exhibited a ‘house edge’ towards Trump, has the national race as Biden +1 today versus Trump +1 three days ago. Due to clumping in how Biden’s vote is distributed, Trump is felt to have a similar structural advantage in the Electoral College to what he had four years ago, when a Trump -2 result in the nationwide vote converted to 306 electoral votes.

Moving to the Electoral College calculus, Biden appears to have two main paths to winning – what Nate Silver referred to in an article yesterday as Plan A1 and Plan A2. Plan A1, which has gotten more attention, is: flip Pennsylvania. Biden has had a consistent polling lead in his birth state, which went Trump +0.7 in 2016; the current polling average is Biden +5. If Biden wins Pennsylvania, it appears to be very difficult for Trump to get to 270 electoral votes.

Plan A2 is: flip Arizona, and also the Omaha congressional district (NE-2). Arizona went Trump +3.5 in 2016, but the state has been changing; the Democrats captured a Senate seat in 2018 and the current polling average is Biden +3.5. Omaha’s sole electoral vote (Nebraska being one of two states that is not winner-take-all) went Trump +2 in 2016, but what little polling has been done specific to that district has shown Biden leads of 7 to 11 points, and FiveThirtyEight’s current forecast for the district is Biden +6. One can construct a plausible map in which Trump wins Pennsylvania and otherwise has a good night, but Biden wins Arizona and Omaha to pull out a 270-268 victory. Which helps explain why Trump took the time two days ago to do one of his trademark rallies in Omaha, only 7 days before the election.

But of course there are plenty of other states in play. For starters, it seems increasingly likely that Biden will flip both Wisconsin and Michigan. Wisconsin had unexpectedly gone Trump +1 in 2016, but there was a credible poll yesterday that had it at Biden +17. While that may be an overbid – FiveThirtyEight’s forecast is currently Biden +8 – the reality is that Wisconsin is currently one of the very worst states in the country with respect to the COVID-19 pandemic, and that is surely hurting Trump’s chances there. That same poll had Michigan, which unexpectedly went Trump +0.25 in 2016, as Biden +7. This is very consistent with what other polls have shown throughout the campaign, and FiveThirtyEight’s current forecast for Michigan is Biden +8.

More interestingly, there are several states that Trump won in 2016 where things seem very tight: Florida, Georgia, Iowa, North Carolina, Ohio, and Texas. If any of these break for Biden then Trump’s chances would be slim. Silver recently referred to these as Biden’s “multiple Plan Cs”. While Trump has attempted to play some offense in states that went for Clinton in 2016 – Minnesota, Nevada, and New Hampshire – the contour of the campaign has largely been Biden on offense and Trump on defense. If everything breaks in Biden’s favor, it’s possible to imagine him getting over 400 electoral votes.

However all of this model-driven speculation, while fun, ignores some important dynamics. Early voting and mail-in voting are both at levels not previously seen, due to a combination of voter enthusiasm and concerns about in-person voting due to the pandemic. This raises the usual level of questions about whether pollsters have accurately captured this year’s electorate. But in addition, there will be tremendous inconsistency across states as to which type of votes will be tallied when; and this has the risk of impacting the public perception of what is actually happening in a given state’s election results.

It is easy to imagine a scenario where, given a state’s election procedures, and given differences between the parties’ electorates with respect to voting method, one candidate appears to be ahead on election night but the other candidate ultimately takes a lead that is outside the mandatory recount threshold. Would the public accept such a result as legitimate, or would they suspect that hanky-panky by electoral officials occurred? These public perception concerns have been exacerbated by comments already made by both President Trump and SCOTUS Justice Kavanaugh, suggesting that this type of situation inherently raises questions about the legitimacy of the electoral process.

So, we may be in for a long haul here, with no clear winner established on the night of the election, with results coming in over days if not weeks, and with considerable potential for judicial activity not just in one state (as in 2000) but in multiple states simultaneously.

And then there are also Senate and House elections to consider. It is generally agreed that the Democrats will retain control of the House, with most observers believing they have an opportunity to expand their majority. The Senate landscape in 2020 did not initially look promising for the Democrats, who need a net gain of 3 seats to get to 50. The conventional wisdom in mid-summer was that while there was a clear path for the Democrats to win the seats in Arizona (McSally), Colorado (Gardner), and Maine (Collins), it was very likely that the Democrats would lose their windfall seat (from the Roy Moore debacle) in Alabama (Jones), and it was far from clear what other seat could successfully be put in play.

However, much has happened since then; and while it could easily still all go wrong for the Democrats and they could end at 49 (or even lower, e.g., if Collins pulls off a comeback), right now the modal outcome of FiveThirtyEight’s forecast is a tie between 51 and 52 Democratic seats. Republican incumbents in Alaska (Sullivan), Georgia (Perdue), Iowa (Ernst), Montana (Daines), North Carolina (Tillis), and South Carolina (Graham) are all in much tighter-than-expected races. In addition there’s a close race in Kansas for an open seat, and the “jungle general” race in Georgia to fill the remainder of Loeffler’s term appears destined to go to a runoff between the Democrat Warnock and either Loeffler or Republican Congressman Collins. And even states like Kentucky (McConnell), Mississippi (Hyde-Smith), and Texas (Cornyn) no longer look completely safe for the incumbent, although the same could be said in the other direction for Michigan (Peters), where the African-American Republican challenger could create some ballot-splitting.

As such there will be quite a bit to watch for outside of the Presidential election, including the possibility that Senate control could hinge on a runoff election in Georgia. And there will also be the usual array of state-level actions to follow; of particular interest to me is the so-called “fair tax” amendment in Illinois, which seeks to overturn a clause in the state constitution that prohibits graduated tax rates.