US Election: Long-term indicators favour Biden for President

The series of events

Since I last analysed the race for Next President – before the primaries began – events have somewhat transformed the context and the betting has swung wildly.

Last autumn, I was betting on Donald Trump to be impeached by the House of Representatives and predicting the fallout to dominate election year. Whilst that bet landed, impeachment now feels an awful long time ago. The Russia scandal, Mueller report and imprisonment of several key Trump allies has been completely eclipsed by the coronavirus scandal.

Betting trends

When the Senate voted against the House verdict, acquitting Trump, his odds for re-election went into freefall – to a new low of $1.65. Elsewhere, I wrote that his odds were even more wrong than ever, adding to various anti-Trump positions taken during his tumultuous first term. For the record, my average lay price is around $2.39 – leaving aside a convoluted series of hedges between various markets that is panning out better.

Then coronavirus took centre stage and the stock market crashed. Amid a general sense in the media that Trump was bungling the crisis, the market flipped back. In recent weeks, he’s drifted out to $2.30 and today the price is $2.16.

You may think I’m feeling more confident. Obviously it is a relief to see the odds move my way but, in fact, that isn’t the case. My view is that, contrary to what many of us once said about rational political betting markets, this one has become highly irrational, driven by dubious media narratives in a changing world. Let me explain.

Why Trump won in 2016

First, my analysis of 2016 has not changed. Rather than the result representing a vindication of Trumpism, it was a rejection of Hillary Clinton. Trump underperformed Mitt Romney’s losing share from 2016 and the national GOP share. His victory was a geographical, statistical fluke, reliant on three factors.

First, a pair of successful ‘third party’ campaigns from Gary Johnson and Jill Stein, which disproportionately hurt Clinton. Second, a Kremlin-backed fake news campaign (later proven), that reinforced a quarter-century of negativity towards Clinton. Third, depressed turnout on the Democrat side, critically among black voters. This meant that, whilst she won the popular vote by 2.9M, the electoral college was lost by around 70K votes, spread across three states.

What has changed is my analysis of how likely voters are to switch and what issues would drive them to do so – in any election but particularly the USA. Back in 2017, I still believed votes shifted on the old fundamentals – events, scandals, mistakes, the economy.

Micro-targeting has changed the game

What we have learnt since – from the Mueller Report, the UK’s ‘Fake News Inquiry’, films such as The Great Hack, hard data from polls and elections – suggests the game has fundamentally changed. Those conventional indicators now barely move the needle.

Whereas political information was once dominated by mainstream media, now it is decentralised and anarchic. A relentless blizzard of information – political or otherwise – has fractured our attention spans. Unless committed to reading a range of sources, we are all vulnerable to one-sided news or disinformation.

Micro-targeting of voters, based on their data and psychological profiles – whether from conventional parties, lobbies or foreign intelligence agencies – is now king. The upshot is we are all doubling down on our existing opinions or prejudices. Consequently, the issues that once moved polls no longer seem to.

Coronavirus may not hurt Trump’s chances

A good example can be found in very recent events. Last week, the stock market was crashing and Trump’s narrative that coronavirus was a Democrat hoax imploded in full view. Commentators predicted his demise. The betting moved against him but polls didn’t. According to NBC-Marist numbers, Trump’s ‘strong approvals’ actually rose to a new high.

On the other side, at the mid-terms, exit polls showed 64% were positive about the economy. Did that translate into Republican success? Quite the reverse – Democrats enjoyed their best result in 40 years.

Mid-term polls proved spot-on, confirming a Democrat advantage that had been evident from almost the moment Trump took office, translating into a series of stunning wins in special elections. There is little or no evidence of change since. Polarisation seems entrenched.

Where does that leave us? Well, polls between Trump and Joe Biden have been taken frequently since 2015. They make terrible reading for the president. Although Biden’s lead has fallen from the mid-teens last summer, (probably due to divisiveness during the primaries, notably from Sanders supporters), it remains substantial. Five of the last six show him at least 9% ahead.

Trump approval remains low at an average sub-43% – down since taking office and remarkably consistent. Perhaps the clearest evidence of swing since 2016 lies in the generic congressional ballot. The best real-world measure of this is the House vote – which Democrats lost in 2016. They now consistently lead, averaging around 8%.

If these trends persist, I don’t see how Trump can win. He’ll need extra voters this time – 46.1% won’t be enough and turnout on the Democrat side has risen sharply. Biden showed in the primaries that he can get that critical African American vote out much better than Clinton.

I’m therefore happy with my position but perversely, rate his chance higher than probably any other time. Why? Because the narrative I thought would play out – Trump on trial, ties to Putin and organised crime exposed, ongoing legal battles – is yesterday’s news.

Instead, coronavirus offers an opportunity to show leadership. I doubt it will shift entrenched opinion but it at least motivates his base more than ever. Nevertheless, I still say he’ll lose by eight to ten points.