Election Models, Election Dynamics and Early Voting Data

As it stands today, the Biden Campaign would appear quite likely but not guaranteed to win come November 3rd (or at some point thereafter). It could end early on November 3rd (if Florida appears to be trending toward Biden). Namely, it is hard to craft a scenario whereby Trump loses Florida and wins the White House. 538 has created an interactive where you can explore the inferential dynamics between the states (we learn about the likelihood in State B from the earlier results in State A). The interactive also highlights how results in early reporting states can reduce the remaining plausible paths to victory (there are only a few paths for Trump at this point).

Of course, it should be stated that remaining events or other issues could (potentially) change the dynamics or undermine the ability to leverage polls to make a proper inference. Here are few possibilities —

(1) Another October Surprise could drop between now and Election Day (there have already been several). However, it should be noted that one implication of all of this early voting is that the impact of a late October surprise is diminished.

(2) There could be systematic bias in polling (such as an unwillingness on behalf of voters to admit to pollsters their support for Trump). Alternatively, there could be a fundamental misunderstanding of the composition of the 2020 Electorate. As has been recently noted, the Trump Campaign has spent a significant amount of time on voter registration in several key battleground states. Will these newly registered folks actually vote ?

(3) Turnout dynamics associated with the cocktail of early voting (very large numbers so far), large scale absentee ballots (including rejection of ballots, delays in mail, etc.) or fear of turning up to the polls due to our latest COVID surge (the Trump campaign is counting on a Election Day surge). Any or all could impact the final outcome.

That said, if I had to bet I would bet on Biden to win (and give far better than even money).

We do have at least some information on the state of ongoing voting thanks to the Early Voting Tracking Project by Michael McDonald.

It is unprecedented turnout thus far.  On its face this would purport to favor the Biden Campaign. However, the question remains whether this is merely a cannibalization of the normal Early In Person Voting and/or Election Day In Person Voting.  In other words, how much will net turnout increase? Will it make a difference?    

Taking Pennsylvania as a highly probable Tipping Point State, it will be interesting to see what percentage of mail in ballots are returned in the days to come.  At the County level, there is significant variation in number of returned ballots thus far (even among those who have already requested a ballot).   

What A Difference 2 Percentage Points Makes and Nate Silver vs Huff Po Revisited

There is likely to be lots of recriminations in the pollster space but I think one thing is pretty clear — there was not enough uncertainty in the various methods of poll aggregation.

I will highlight the Huffington Post (Huff Po) model because they had such hubris about their approach.  Indeed, Ryan Grimm wrote an attempted attack piece on the 538 model which stated “Silver is making a mockery of the very forecasting industry that he popularized.”  (Turns out his organization was the one making the mockery)

Nate Silver responded quite correctly that this Huff Po article is “so fucking idiotic and irresponsible.”  And it was indeed.

Even after the election Huff Po is out there trying to characterize it as a black swan event.  It is *not* a black swan event.  Far from it … and among the major poll aggregators Five Thirty Eight was the closest because they had more uncertainty (which turns out was quite appropriate).  Specifically, the uncertainty that cascaded through 538’s model was truthful … and just because it resulted in big bounds didn’t mean that it was a bad model, because the reality is that the system in question was intrinsically unpredictable / stochastic.

From the 538 article cited above “Our finding, consistently, was that it was not very robust because of the challenges Clinton faced in the Electoral College, especially in the Midwest, and therefore our model gave a much better chance to Trump than other forecasts did.

Take a look again at the justification (explanation) from the Huffington Post:  “The model structure wasn’t the problem. The problem was that the data going into the model turned out to be wrong in several key places.”

Actually the model structure was the problem in so much as any aggregation model should try to characterize (in many ways) the level of uncertainty associated with a particular set of information that it is leveraging.

Poll aggregation (or any sort of crowd sourcing exercise) is susceptible to systemic bias. Without sufficient diversity of inputs, boosting and related methodological approaches are *not* able to remove systematic bias. However, one can build a meta-meta model whereby one attempts to address the systemic bias after undertaking the pure aggregation exercise.

So what is the chance that a set of polls have systematic error such that the true preferences of a group of voters is not reflected?  Could their be a Bradley type effect?  How much uncertainty should that possibility impose on our predictions?  These were the questions that needed better evaluation pre-election.

It is worth noting that folks were aware of the issue in theory but most of them discounted it to nearly zero.  Remember this piece in Vanity Fair which purported to debunk the Myth of the Secret Trump Voter (which is the exact systematic bias that appeared to undermine most polls)?

Let us look back to the dynamics of this election.  There was really no social stigma associated with voting for Hillary Clinton (in most social circles) and quite a bit (at least in certain social circles) with voting for Trump.

So while this is a set back for political science, I am hoping what comes from all of this is better science in this area (not a return to data free speculation (aka pure punditry)).

P.S. Here is one more gem from the pre-election coverage – Election Data Hero Isn’t Nate Silver. It’s Sam Wang (the Princeton Professor had HRC at more than a 99% chance of winning).  Turns out this was probably the worst performing model because it has basically zero model meta-uncertainty.

Econometrics (hereinafter Causal Inference) versus Machine Learning

Perhaps some hyperbolic language in here but the basic idea is still intact … for law+economics / empirical legal studies – the causal inference versus machine learning point is expressed in detail in this paper called “Quantitative Legal Prediction.”  Mike Bommarito and I have made this point in these slides, these slides, these slides, etc.   Mike and I also make this point on Day 1 of our Legal Analytics Class (which really could be called  “machine learning for lawyers”).