Surface Ripples

When Numbers Go Wrong

I recently finished a book called “Humble Pi” by mathematician Matt Parker.  The book is a fascinating series of examples of how “human error” can render complex calculations useless, resulting in some spectacular (and tragic, at times) disasters.

Subtitled A Comedy of Maths Errors, the essence of the book can be distilled down to “Rubbish in, rubbish out” (particularly as many of Parker’s examples involve our reliance on computers and the computational powers of Microsoft EXCEL but forget the “human error” factor.  Parker’s main point is, if you get the inputs, the sums or a combination of both wrong, you may end up with a problem.

This is not to deny the advances in human knowledge and scientific understanding, it’s just that sometimes we overstretch ourselves - we run before we’ve learnt to walk properly.

Much of our life is dominated by numbers: salaries, cost of goods, returns on investments, power consumption, trade deficits, exchange rates, interest rates, how our business is running and so on.  The more we come to rely on machines to crunch these numbers for us in different ways to inform decision making, the more we may be lulled into a false sense of security.  Many of the problems narrated by Parker arise due to the proverbial “one in a million” chance of something going wrong in that, the more something is used or run, the more likely “one in a million” event becomes.  As long as we’re not the millionth, things are OK.  

Take the problems of Cerebral Venous Sinus Thrombosis (CSVT) - rare brain clots - experienced by a tiny percentage of people who received the AstraZeneca COVID-19 vaccine.  This condition is said to affect c. 4-5 in 1 million people every year.  Apparently, it is also more common amongst younger women and may/may not be linked to birth control medication and to the under-30 age group.  

The regulator looked into the number of people in the UK who had developed rare blood clots after receiving the AstraZeneca vaccine.  It found that 79 people - two-thirds of them women - experienced the clots after receiving a first vaccine dose. Nineteen of them died.  

According to the UK medical regulator, 79 people (two-thirds - 53 - of them women) experienced clots after receiving a first dose of the AstraZeneca vaccine. 19 died.  More than 20 million (suspiciously nice round number) doses had been administered across the UK by the end of March.  

So, out of 20 million people vaccinated, 79 experienced clots.  Of the 79, 19 died.  One could infer that:

• Our chances of experiencing a clot are: (79/20,000,000) x 100 = .000395% BUT:  
• The chances of dying if we get a clot increase to (19/79) x 100 = almost 25%/1 in 4 

Another way of looking at it is to say chances of dying after receiving AstraZeneca are (19/20,000000) x 100 = 0.000095% or 1 in 1,900 (taking the percentage of total numbers vaccinated).  

In neither case are we necessarily correct and it is no comfort to us or our nearest and dearest if we are the “unlucky one”… 

There may or may not be a causal link, there is certainly a correlative one, and we’ve heard no reports of similar cases amongst this receiving (say) the Pfizer BioNTech vaccine.  The medical world is still unsure of whether the vaccine is what provokes the clots, but a tiny percentage of people (and only those receiving AstraZeneca) seem to have been fatally impacted.  

The point is, despite all the clinical trials, AstraZeneca could not have predicted this occurrence, given its rarity.  Trials involved a limited number of people over a relatively short period of time.  I suspect that there was also pressure to produce a working vaccine ASAP.  The medical world’s opinion is that the benefits of the AstraZeneca vaccine outweigh the risks of hospitalisation and death for the vast majority of people.  For younger age groups and those with a history of blood clots, it is more "finely balanced”, and they should be offered an alternative.

In short, it helps to be a little bit cynical about numbers (especially whole numbers) and to ensure that we understand how they were produced.  From Matt Parker, we understand that there is always a risk that “the numbers” may not work out in our favour for whatever reason.

I’ve spent more than half my life delivering change in different world markets from the most developed to “emerging” economies. With a wealth of international experience in international financial services around the world running different operations and lending businesses, I started my own Consultancy to provide solutions for improving performance, productivity and risk management.  I work with individuals, small businesses, charities, quoted companies and academic institutions across the world. An international speaker, trainer, author and fund-raiser, I can be contacted by email. My website  provides a full picture of my portfolio of services.  For strategic questions that you should be asking yourself, follow me at @wkm610

Labels: Financial, Productivity, Strategy

posted by William Martin @ 08:06