The question we’ve all been asking!
????♂️ #chriswhitty pic.twitter.com/v9vzt9Lb1N— Spoofed (@spoofed_uk) December 18, 2020
Human beings have been striving to predict the future for as long as we’ve existed. Knowledge, after all, is power, and if we know what’s going to happen, then we can be ready for it. Whether you’re visiting a shaman or a vodou priestess, a local wise woman or an astrologer, you’re likely to be posing them the same kind of question. Where will I be in the future? What will come to pass? And while we’re at it, who’s going to win the Grand National?
While there’s always something to be gained from looking ahead, foresight is probably most quantifiably beneficial when it comes to betting – here, the advantage of foresight is measured in monetary terms. What’s more, one can bet on almost anything. As a hitherto unenthusiastic gambler, I was intrigued to discover that many bookies offer odds on much more than football or horse racing – political betting sites, for instance, permit wagers on everything from the next leader of the Tory party to the date of the next general election.
ArrayOthers go further still, allowing punters to request bets on practically anything they can think of – Bono for the next Pope, even ‘when will the world end?’ – but however outlandish the bet, it’s in calculating odds that things get really interesting. Basically, a betting provider has to balance their understanding of a specific outcome’s probability with an enticing potential return for the customer. The less likely an event, the higher the odds offered – as such, the biggest risks make for the most sizeable rewards.
Certainly, it’s a refreshing angle on uncertainty, especially now that dice-rolling has left the Craps table and become a way of life. The pandemic’s story is far from over, but as we hurtle past the anniversary of the UK’s first confirmed coronavirus death, we can benefit from a full year of (horrible) experience. As they say, hindsight is 20:20; looking ahead, with the best will in the world, tends to be rather less clear – not that that stops us trying.
Some with more success than others. There exists a category of forecasters so preternaturally accurate as to have been bestowed the prefix ‘super’ – one such superforecaster, Jonathon Kitson, explains how such a cohort came to be identified in the first place. Back in the 80s, Professor Philip Tetlock had just been given tenure at UC Berkeley at a time when liberals and conservatives alike were feverishly prognosticating on the Cold War.
Array‘So people would make very bold statements about what the Russians were doing […], and [Tetlock] collected thousands of these statements [from Sovietologists], and put them into something he called the Expert Political Judgement project’, explains Kitson. ‘What they found was that experts basically flip a coin’, he continues, describing the odds of a specialist forecasting a given question correctly. In short, when the predictions and outcomes were tallied, ‘being a subject level expert did not give you any predictive advantage whatsoever.’
Shocking, right? ‘That’s the word for it’ muses Kitson, and Tetlock must have thought so too: along with a team of academics, he decided to see if there existed people who could better the experts when it came to forecasting. In 2011, ‘they put together the Good Judgement Project, which was an online crowd sourced prediction market. Say the question is, who’s going to be Prime Minister of Japan – Mr X or Ms Y? Well, what they found was that the crowd itself was more accurate [than an expert].’ And what’s more, ‘they found a specific group of people, about 2%, who were very accurate – […] about 70% on any given question’: the superforecasters.
Astonishing, at first glance anyway, is the fact that these exceptional individuals weren’t just more likely to be right than experts – superforecasters did better than CIA agents, with access to classified information pertaining to the subjects they were musing and years of field experience. So what’s going on? Are superforecasters clairvoyant? Why are they so much better at predicting things than the rest of us, not least our most lauded experts? Well, according to Kitson, ‘we tend to think about a question differently’; superforecasting is an art as much as a science. The more I learn, the more it seems that maths is neutral – rather, people are what make things tricky. As Kitson says, a ‘super’ is able ‘to separate what you want to happen from what you think will happen’ – that might sound simple, but it’s easier said than done.
From pandemics to policing our bad habits, public health legislation is devised based on various calculations much like those ones churning in our heads – take the issue of alcohol, and you might start by counting how many deaths per year are attributable to its (mis)use in a given population. You could look at levels of consumption across different demographics, zoom in on what kind of drink an alcoholic is most likely to use, how many people in A&E are there because of drink driving or over indulgence, and so on.
Cast your mind back to summer, and Eat Out to Help Out. If you’re a (short-sighted) economist, it was a stroke of genius: pouring money back into businesses who had lost income over lockdown, the scheme boosted the treasury by £250 million in just a few months. If you’re an epidemiologist, on the other hand, Rishi Sunak’s plan for the nation to pile into crowded restaurants via packed public transport might have pushed you to the brink of tears.
Taking a wider view, and when it comes to locking down, the UK government has been staggeringly and consistently slow (doing far more economic damage than a few months out of cafés, but that’s another story). Far be it for me to ascribe their glacial pace to incompetence – what’s unquestionable is its ideological foundations. I might not agree with Boris’ reluctance to rip the plaster off, but that’s not the point: rather, people (Prime Ministers, scientists, economists) look at the same facts and figures, only to emerge with wildly divergent conclusions depending on their priorities. Hey, we’ve all got ’em – but it’s precisely those blinding biases that disqualify the majority of us from potential superforecaster status.
Just because Brexit seemed unimaginable to half the country back in 2016, didn’t stop it happening. Just because Boris would like to open schools doesn’t make it a good idea from an epidemiologist’s point of view; on the other hand, it’s not hard to find parents who are troubled by keeping them shut. All three of my imaginary complainants (Boris is real, I think) are projecting themselves into potential futures according to their personal priorities and populated by the characters which loom large in their psyches, fears and desires both.
Right on cue, here’s Kitson again: ‘Say I asked, in 2016, who’s going to be the next President of the United States – Donald Trump or Hillary Clinton? If I say Donald Trump to you, what do you think?’ Great question, Jonathon! Verbatim, and I have the recording, I said: ‘I think, oh no. I hate him very, very much.’ Having thoughtfully demonstrated an exemplar of how not to predict things, Kitson gently explains that a superforecaster takes a decidedly less hysterical stance.
‘It’s an election, so who tends to win the elections? […] Rather than it being Donald Trump, Hillary Clinton, it’s Republican versus Democrat’, he says. ‘And you’re also trying to establish a base rate. So once you’ve got the outside view, how often do the Republicans win, how often do Democrats win? How often do the changes happen?’ What had felt unbelievable in an almost bodily way is sounding more sensible by the second. ‘You tend to get eight-year blocks, Republican Democrat, Republican Democrat. And [the US had] just had eight years of a Democrat.’ Gulp.
When it comes to forecasting the Big Picture, then, emotion emerges as the anathema of accuracy. Nonetheless, Big Pictures continue to seem inextricably linked to the Small Ones which comprise them. What’s a crowd but a swarm of individuals, diverse as they are legion? The temptation to project what one wants onto the world, and onto the rest of the herd, is agonising – especially today when uncertainty has accrued to soaring heights and yet the end feels, dare I say it, almost in sight. Then again, I would say that, wouldn’t I? Bloody wishful thinking! Leave the future to the professionals (not necessarily the experts), and don’t place any bets on me.
–
Header image artwork is:
