Early Election Results vs. the Final Forecast and a Quick Look Ahead

With the early results in, it is clear that the election is closer than the expected results from our forecast. It appears that polls broke in favor of Trump at the end, confirming signs we saw in some of the recent polls leading up to election day in key states.

What can these early results tell us about what is likely to happen?

We can use the simulations from our Presidential election forecast and condition on simulations where we observed results that we have so far. This allows us to focus on scenarios that have played out like reality so far while still allowing for random variation in line with the historical errors we have seen in past forecasts.

This allows us to make an updated probabilistic prediction on the election and guess at what might happen in the states that are yet to be called.

As of just past 2 am on the east coast, this is the current electoral map:

Image for post
Image for post

Florida is a state the forecasting model missed as it expected a slight lead for Biden based on fundamentals and polling averages of just under 1%, but Trump won the state and leads by +3% with 96% of the votes in.

As of now Biden needs 45 electoral votes to win the election while Trump needs 58. Assuming Trump wins Alaska as expected, we can look at what happens if he also wins Georgia and North Carolina, which is plausible based on current results.

That would give Trump 247 electoral votes. In scenarios where this happens, Trump is forecasted to win the election 74% of the time. If Trump loses Georgia though, that election win probability drops to 68%.

However, if we account for the fact that Biden currently leads in Arizona and condition on simulations where he wins the state, then Biden is actually likely to win the election 94% of the time.

All this is contingent on the midwest playing out as the polls suggested it would, meaning that in particular Wisconsin, Michigan, and Pennsylvania follow expected patterns and Biden is able to win them.

One concerning result today for Biden is that the margin in Ohio has shifted more towards Trump than the forecasting model expected. Trump seemed favored in Ohio by +1% but ended up winning the state by more. He currently holds a +8% margin in the state with 90% of votes in.

This suggests that Trump could be running ahead of his polling margins in the midwest, meaning that he might see his actual vote shares increase above expected polling averages anywhere from +4% to +8%.

Countering this though is the result in Minnesota where Biden is currently leading by +7% with 80% of the vote in, which is close to his polling average of +8%. Minnesota’s vote is historically less correlated with other midwestern states, but it is more demographically similar in some ways to those states compared to Ohio. The fact that the actual result is close in Minnesota suggests polling there, and potentially in Michigan and Wisconsin, was fairly accurate.

Biden was expected to win Pennsylvania by +5% and Michigan and Wisconsin by +8% each. We can see how Trump running ahead of his polling by large enough numbers in the midwest would shift these states towards his camp and therefore the election.

If that shift though is not large enough (whether due to late deciding voters, turnout, or polling errors), then Biden could still win Michigan and Wisconsin while losing Pennsylvania. That would make the election a nail biter as Trump would have 267 electoral votes, leaving the race to Arizona, Maine, and Nevada. However, given Biden is favored in these states, he would be slightly favored to win in this scenario.

The Trump camp has to feel a lot better about their chances than they did going into election day while the Biden camp is certainly feeling much more nervous.

It is going to come down to Pennsylvania, Wisconsin, and Michigan in the midwest. Overall, it looks like Trump is favored to win currently with anywhere from a 60% to 70% chance based on current results and voting patterns, but the odds could change very quickly pending how other key states, in particular Arizona and Georgia, turn out and how the remaining votes start to come in.

We will need to see whether the results in Pennsylvania, Wisconsin, and Michigan match the trend in Ohio or Minnesota more closely, especially since Pennsylvania mail in ballots have not really been counted yet.

Vinod B

Politics, Policy, Economics, & Data

Medium is an open platform where 170 million readers come to find insightful and dynamic thinking. Here, expert and undiscovered voices alike dive into the heart of any topic and bring new ideas to the surface. Learn more

Follow the writers, publications, and topics that matter to you, and you’ll see them on your homepage and in your inbox. Explore

If you have a story to tell, knowledge to share, or a perspective to offer — welcome home. It’s easy and free to post your thinking on any topic. Write on Medium

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store