Grappling with Statistics
The third post in my Blogging for the Holidays series: Grappling with Statistics. Hopefully not so heavy…
We ask a lot of the human brain these days. For a species that evolved on the savannah of Africa, hunting and gathering in small social groups, we’re now a population of 8 billion people living and working together, tackling ever large and more complex problems. Through application of our incredible minds, unmatched in the rest of the natural world, we’ve managed to spread to every corner of the planet, plumbed the depths of the earth, dived to the bottoms of the oceans, and extended our reach into the solar system.
And yet, through it all, we find ourselves grappling with ideas that seems beyond the reach of our comprehension.
Let’s try a little Stoic exercise. First, visualize yourself from above in the room you’re sitting in now. Next, pull back, and try to visualize your home and your position in it. Now pull back again, visualizing the block on which your house sits. And again, but your quadrant of the city. Now the city itself.
Now consider how quickly you lose the ability to truly reason about size and distance. For me, beyond my local neighbourhood, I start to lose a sense of scale. When just thinking about my city, I can only picture my position on the map because of how often I’ve punched my address into a GPS.
Next imagine the scale of your country or the planet. Already we’re beyond the reach of the human mind to reason. Instead, we have to rely on relatable metaphors that connect our ground truths to these types of scales.
And the solar system? Our galaxy? Good luck. The universe? Impossible!
The same can be said of many concepts. I’ve long believed that the inability for humanity to truly rally together to address climate change is, to a great degree, a consequence of the inability of the human mind to truly understand the magnitude of aggregate human impact on the planet, or the timescales over which we’re changing it. Humanity itself has gotten so large that our collective behaviour is beyond our intuition.
Probability and statistics are another example of this phenomenon. The human mind is simply not wired to think in these terms–I suspect this is one of the many reasons why Dr. Richard Feynman is famously quoted as saying “If you think you understand quantum mechanics, you don’t understand quantum mechanics.”1
Now, in the past, we mostly led our lives without spending a lot of time thinking about statistics. Rather, we’d use our intuition to make what we believed were sensible decisions every day; intuition that was often very wrong! But in the end, given the localized impacts of these decisions, that they were based on faulty reasoning didn’t matter all that much.
But, of course, the pandemic has upended things, turning normal uncertainty into life-or-death decision making. And we’re just not wired for it.
What does it mean for a vaccine to be 70% effective? Well, first of all, you have to ask: effective at what? At preventing contraction of illness upon a confirmed exposure? Preventing symptomatic infection? Preventing hospitalization? Preventing death?
How do I use those statistics to inform decision making?
If I’m vaccinated and the vaccine is 70% effective at preventing contracting the illness upon exposure, should I go to a restaurant? Attend a Christmas party?
How does that calculus change if 60% of the total population is also vaccinated? What about 80%?
Now what if I’m a caregiver for someone unvaccinated or vaccinated but immunocompromised? Should I go to the restaurant then?
A new variant has come out. Effectiveness is now 30%. What do I do now? How should I change my decision making?
The pandemic has forced us to ask these and a million other questions, and they’re novel and challenging for a number of reasons.
First, we’re dealing with aggregate risks and statistics, and the human brain is truly terrible at thinking in those terms.
Second, unlike typical risk-based decisions, our individual choices have an impact on not just our friends and loved ones but our community. Conversely, the decisions of others have a direct impact on the risks in my own life, which means I now need to consider not only my own behaviour but the behaviour of everyone else.
Now, oddly, these types of risks are not entirely new! Every day millions of people take to the roads. Each individual choice behind the wheel has the potential to impact others.
But where a poor driver has the ability to impact the lives of a handful of others, the decisions made by a carrier of COVID have the potential to impact the lives of thousands of people, a number that itself is difficult to comprehend.
Uncertainty is now all around us, constantly forcing us to consider decisions that, in the past, wouldn’t earn a second thought. Should I go to a theatre? To a restaurant? The company Christmas party? A hockey game? A family gathering?
Should I wear a cloth mask or an N95? Should I wash my hands right now or is hand sanitizer good enough? Do I really want to go inside the grocery store today or should I just do a pickup order?
In trying to answer these questions, individuals are now faced with the idea that our knowledge, itself, is uncertain. For example, right now, we find ourselves wondering how severe the Omicron variant will turn out to be. Will it be mild, as has been seen in South Africa? Or will it be comparable to Delta, as is suggested by early data in the UK? And how will public policy change as that our understanding changes? We don’t know! But as we learn more, our behaviours must change accordingly.
Medicine and associated health policy has always been built on a foundation of statistics. Anyone who’s had extended contact with the healthcare system quickly discovers how often medicine is a process of hypothesis based on a preponderance of evidence (diagnosis), followed by validating of that hypothesis (tests, proposed therapies), rather than certainties based on concrete knowledge.
Your average person is rarely exposed to this side of healthcare. Until, that is, COVID came to dominate our lives.
In this way, just as we now find ourselves grappling with viruses for the first time in a generation, we’re finding ourselves facing brand new kinds of risks and uncertainties the likes of which we’ve never had to deal with before. And the shift has been sudden and seismic.
Why, then, should we be surprised that folks are angry, frustrated, confused, or disbelieving?
There’s been a lot of talk, lately, about how weird the financial markets have gotten. GameStop and AMC; the incredibly frothy broader stock market; the hot IPO and M&A environment; the speculative mania that is crypto and NFTs; the endless DAO rug-pulls. Things have gotten legitimately a bit weird.
In parallel we’ve seen an absolute explosion in online gambling. The US has loosened up laws that have allowed for online sports betting, and we’ve seen an increase in both private and government-run gambling services in Canada as well. More broadly, in the gaming industry, there’s been an absolute plague of lootbox-style mechanics, and now companies blending gaming and the crypto “industry”.
It’s remarkable to me that these changes are happening at a time when our lives have become so incredibly uncertain; that, when things look riskier and scarier than ever, gambling itself is on the rise.
In the end I wonder: in a world filled with so much uncertainty of a kind we’re simply not equipped to handle, are we turning to gambling as a way to take risks that are familiar and understandable, thereby paradoxically giving us some illusion of control?
-
Admittedly there’s some dispute over the source and phrasing of this quote. ↩