online video : Sanders vs Trump head to head
I attended an event in Singapore recently where a panelist with an Australian accent proclaimed with confidence that Trump would get re-elected in the US in 2020. The stakes are high, especially considering Trump’s outsized influence on the willingness of international players to commit to more ambitious Nationally Declared Contributions to the Paris Agreement. So is there any merit to this Aussie’s prediction?
What are Trump’s chances for re-election in 2020?
Given the information we have now — I estimate 23% likelihood using this spreadsheet and I believe even those odds are optimistic for Trump as I will explain. His odds in online prediction markets and this Bloomberg article are not much better hovering around 30–40%. If there is anything we have learned from the past however, it is that when it comes to elections anywhere in the world — nothing is certain. Just ask Carrie Lam. In an earlier article I explain why I feel that a Trump aquittal in the impeachment is good news for Democrats in the 2020 election . As explained in this article, his prospects are not great for re-election and a conviction in the Senate would have allowed the GOP the chance to reposition themselves with a better candidate. So long as Trump is the favorite in the party no viable competition can get off the ground so they are locked-in to him, for better or for worse.
So why do I give him 23% odds?
Here are four explanaitons, and I will end with four “wildcard” factors which could dramatically change the results as they stand at the writing of this article.
Four factors working against Trump in 2020
- Demographic shifts
- Voter turnout
- Swing states prospects
- Impeachment stigma
Pew Research study of voter demographics shows that the pool of Republican support is consistently losing ground from long term demographic trends and they have not figured how to reverse this decline. Democrats have a clear advantage among all non-white races, women, millennials + Gen Z (voting in this election) and non-evanglicals. All of these demographics trends (except male-female) are growing over time and pointing to a shrinking pool of white old male evangelicals. These are long term trends and were already favoring Democrats in 2016 — Hillary did win the popular vote. Most notably is the replacement of the Silent generation voters by Gen Z. This is a predictable trend with clear supporting evidence showing up in 2018. The 2018 mid-terms flipped the House from “Red” to “Blue” and was a decisive factor enabling the Ukraine impeachment investigation to happen at all. A number of factors are different now. The media was relatively balanced and Trump was mostly given the benefit of the doubt compared to the current environment. More importantly in 2016 Independents favored Trump — a historically important predictive indicator. How do Independents feel about 2020? Independents are leaning Democrat. As a side note Independents seem to favor Bernie over Biden as a whole although the Bernie/Biden divide is more complicated, especially in the Gulf States.
So Trump’s chances weren’t great in 2016 and he did lose the popular vote to Hillary. So why did he win in 2016? Two explanatory narratives — voter turnout, and swing states — both with a common theme of disenfranchised working class voters. Voter turnout has historically been higher for Republicans than Democrats, and in particular young and minority voters tend to disproportionately show up less at the polls and that there was evidence that Clinton performed more poorly in the states she lost to a lack of votes from young and minority voters that would have been predicted otherwise by the Obama election in 2012. There are several new factors in the 2020 election that have the potential to motivate both young and minority voters to turnout in higher numbers this year
- Climate change and the Sunrise, Gen Z activist momentum
- Bernie’s ground campaign which is must more energizing for voters compared to Hillary in 2016 — even if the nomination eventually goes to another candidate.
- Impeachment — which has polarized both sides and may be a wash, but will favor the Democrats since they already have the numbers from the general public and Republicans already have high turnout to start with in 2016.
To win a US presidential election the candidate must win state delegates, not necessarily voters — 270 delegates to be precise. Most of the states fall into predictable “Red” or “Blue” states whereas others are “Purple” swing states and the outcomes of elections rests on these swing states. To say you can kind of ignore the other states doesn’t sound very politically correct but actually - it is politically correct. In the 2016 election the important swing states notably 3 “Rust belt” states flipped from “Blue” to “Red” : Michigan, Pennsylvania and Wisconsin. Rust belt is a term that refers to the loss of manufacturing jobs in those states which has resulted in a long term decline in economic activity in those areas.
What happened in the rust belt states that caused them to flip in 2016?
- Switch from swing voters in rural, suburban and independents
- Low voter turnout particular from young and minority voters
- Lack of campaigning in those states by the Hillary team
If we take the map provided by the website 270-to-win, and give all of the delegates to the “Red” and “Blue” we are left with 6 “swing states” and two states that split their delegates based on the vote count — Maine and Nebraska. Democrats must secure 38 out of the total 101 swing delegates from these states. Trump’s strongest support is in the Southern states —Arizona, Florida and North Carolina and gradually weakens the further north you go. Florida is a heavy-weight vs the others with 29 delegates, but is also less likely to swing compared to the others — unless the hurricane season wildcard factor kicks-in which in my opinion as a Floridian and a follower of the climate change story is a near certainty.
Climate change will eventually swing Florida blue, its only a matter of time.
The Republican Governor DeSantis has broken with his predecesor Rick Scott in being more open about the risks of climate change to the Sunshine State. Coral Gables has already declared a climate emergency. For those less familiar, at this point there is little hope of saving South Florida from returning to the sea. South Florida is a narrow strip of low lying limestone with the Everglades to the West, the Atlantic Ocean to the East, and a porous limestone foundation which would render sea-walls completely useless. As a kid I remember digging in the back yard and hitting the water table — just to give you an idea of how precarious things are. Absent of a miracle 180 turn in global politics towards a 1.5C warming pathway South Florida’s fate was decided in the 1982 ExxonMobil internal memo that predicted that sea level rise due to carbon emissions would eventually inundate the state. Floridians are only just now waking up to this reality. Trump’s base however isn’t necessarily in South Florida, but the rural and more high-ground Central, Panhandle, and to a lesser extent West Florida. However -as I mentioned the wildcard is the hurricane season which in 2018 devastated Trump stronghold — the Panhandle Bay County- Panama City with a Category 5 hurricane Michael. Hurricane’s aren’t necessarily a direct link to climate change, but Category 5 Hurricanes are the signature halmark in the era of a +1.0 C world. As a kid growing up in South Florida Category 5 Hurricanes were considered freaks of nature — like the 1992 “Andrew” that barely missed Miami and devasteated Homestead and Kendel — but never thought of as a real threat you would see in your lifetime — a once in 50 year storm as they say. They are now occurring at regular frequency almost annually. The statistical evidence from government data shows that major hurricanes (cat 3+) have increased by +90% since pre-industrial times.
Winning combinations for state delegates
The force of attitudes on climate change may be strong, but political ideology is notoriously stubburn. So let’s continue ignoring the climate change factor for Florida in this analysis. I was curious what the odds would be of securing these 38 delegates so I created a simple spreadsheet and ran 63 combinations of outcomes. By using just combinatorial odds without state-specific weights this works out to a 30% likelihood for a Trump re-election. The winning scenarios can be summarized as follows
- No Florida + 3 out of 5 of the other states
- Florida + 1 of the other states
However, the combinatorial odds are pessimistic given the state-by-state head-to-head matchup information we can estimate from polling data. In the polls, only Arizona is a true toss-up, and the rest are all leaning Democrat, with Biden scoring consistiently higher than other Democrat nominees (Sanders) and strength increasing with latitude. I ran the numbers again and added a bias proportional to the size of the spread lead — 60/40 odds for states with high spread like Michigan and 53/47 odds for states with a smaller spread. The result of this analysis is 23% likelihood for Trump to prevent Democrats from taking the 38 delegates. I assumed that the likelihood of Trump suprising and taking away a “Blue” state was equal to the upset of the Democrats from taking away a “Red” state so this effect was not considered.
spread of Biden vs Trump [# delegates] state
- +6.5  Michigan
- +0.3  Arizona
- +3.7  Wisconsin
- +7.0  Pennsylvania
- +3.4  North Carolina
- +2.7  Florida
Scandals — affairs, corruption — hurts a candidates favorability in ways that are not easy to reverse and forget about. It’s possible but it takes extra mental effort to overcome the cognitive dissonance. Scandals trigger disgust which is one of the most powerful emotive subconscious forces that may not be visible but could be a decisive in two ways 1) repetition — this issue is glaring and will be hard to keep the it out of the public attention because of opposition campaign talking point and may also appear in any head-to-head debates 2) Ballot box effects — for undecided and independent voters this kind of corruption controversy are things that can show up last minute as a subconscious instinct disgust reflex.
- Climate change linked natural disaster (almost certain)
- National security crisis — ex : terrorist attack
- Bernie Sanders nomination
Extreme natural disaster such as wildfire or Category 5 Gulf Coast Hurricane — the “Pearl Harbor” of climate change that crosses the tipping point for political mobilization.
The earlier discussion about climate change was specific to Florida, but the story of climate change momentum is a national phenomenon that is picking up pace fast since the October 2018 IPCC report on 1.5C. 58% of Americans are either “Alarmed” or “Concerned” about climate change and that is a very big deal in politics. “Alarmed” or the politically active group represents 30%, a very large number for active political participation. Normally active political participation on any given issue — abortion, gun rights, healthcare only represents 5–10% of the population on either side, but for climate change this number has been growing fast, particular with the youth activists — XR, Sunrise, FridaysForFuture and this is likely show up in driving the direction of both the Democratic nomination and the presidential election. Climate change activists momentum has the potential to tip the election in two ways — 1) to push independents and less politically engaged towards the left and 2) to ensure that young and minority voters turnout to vote. Canvassing does have an impact and these highly energized climate activists, door knockers and phonebankers have a stake in moving the political needle in ways that may be difficult to counter with traditional paid advertisements. A “Pearl Harbor” like natural disaster would only hurt Trump if it is clearly linked to climate change such as the Australian bushfires. An ambigous natural disaster however could back-fire, and instead be viewed as a general National Security crisis and a victory for Trump if he is able to capitalize on the crisis relief response. Given the heavy media bias against Trump though and the real dynamics of the climate in which we are in uncharted territory, this is very unlikely to not be linked to climate change.
National Security crisis
Terrorist attack, Wuhan virus outbreak — these are all wildcard factors that would favor Trump in a flock to safety and a chance to flex the muscle of the powers of the Executive Branch and the military. Case in point the 9/11 attacks effect on Bush’s approval rating in 2001. Trump should be on-guard and aware that of all of the factors in his favor — this is the one that he has the most control of how he responds. If he is wise he would be especially on-guard to be quick and strong to mobilize his executive powers of the military to look competent and assure his role as provider of security. Anything that creates fear of existential issues (not climate change) could potentially help Trump — of course he also has to not blow it. I drew the inspiration for this scenario from a video clip by Naomi Klein interview by Democracy Now from 2019.
The US is likely to see some sort of slowdown which has little to do with Trump but instead is part of the long term business cycle. The run-up of economic growth also can hardly be attributed to anything done by Trump. Even the policies that accelerated the shale boom were from the Obama Administration. A receission is likely to be blamed on Trump for 3 reasons 1) The media is biased against Trump 2) He brought it on himself by claiming to be responsible for the economic growth, it will be difficult for him to walk that back if it starts to turn the other direction 3) historically presidential results and approval ratings have followed the economy — unemployment figures — despite the relatively little direct influnce that the president has on the figures during his term. Most forecasts at this point suggest a moderate slide and flattening out or slight decline in growth. In recent times however changes downwards are more often coming in crashes rather than gradual declines led by crashes in a hot asset bubble sector that was masked during the “good times”.
What could be the next big bubble? — Some are expressing concern that the next big bubble is oil and gas stocks AKA “the carbon bubble” Oil and Gas major stocks have been tanking (ExxonMobil is trading at below $60 at 10-year low) from a combination of factors — flood of cheap energy from the shale gas boom + divestment pressure from climate activists. The signal from Jim Cramer “I’m done with fossil fuels, they are in the death knell phase, they are the new tobacco”. There is currently ~ $22 trillion of unburnable reserves, stranded assets, that is incompatible with +1.5C warming pathway globally. To put in perspective the sub-prime mortage crisis was at best $0.5 trillion so if and when the carbon bubble pops — it’s effects will send shockwaves globally and one can safely bet that the word “depression” may be a closer fit to the fallout than “recession”. No matter the magnitude any carbon-bubble triggered recession would most certainly be a death blow to Trump’s campaign, and in general any loss of confidence in the economy may similarly wound him beyond recovery.
All of the predictions from the spreadsheet were based on Biden vs Trump, and Sanders right now is not performing as well in “Red” states. He performs slightly better in the “Rust belt” but in general Bernie is much less competitive in “Red” states compared to Biden. It is unclear at this point how much of Biden’s lead over Trump is general Democrat loyalty and solidarity against Trump vs persona appeal between the two candidates. It is not clear how well Bernie will perform in swaying independents and undecided voters away from Trump, and this may weigh heavily on his running mate choice if Bernie is considered for the nomination.
Michael Moore has promoted a theory that Bernie would actually perform better than Biden because he is a more favorable, inspiring candidate that appeals to both populism to peel away disenfranchised Michigan and Wisconsin “Rust belt” voters and mobilizes the youth and minorities in ensuring high voter turnout. While it is an interesting theory, it’s not clear yet from the polling data, but part of this theory is referring to outside-polls effects so its difficult to refute.
Putting it alltogether, given what we know now it really would be a long shot for Trump to win the nomination, but as we have seen recently with the Iowa caucus mixup and the historical butterfly ballot disaster in the 2000 elections which as a young citizen living in Palm Beach County at the time I will never forget — anything is possible in US presidential elections.