STATISTICS VERSUS FACTS

We, human beings, are absolutely awful at understanding statistics and using them responsibly. This has of course been observed by many commentators over the centuries. Let’s start with everyone’s go-to-guy-for-fun-quotes: 

“Facts are stubborn things, but statistics are pliable.”

Mark Twain

The current Covid19 reaction is a clear example. Media (and their consumers) use statistics every day to hype the fear factor to new heights. Every death is highlighted, every infection turned into alarming heat maps showing the whole world turning red. Such statistics are, however, subjective choices by authors that at best, only partially reflect reality. They are certainly not “facts”. 

For example, how many of you- the readers of this article- are aware of the fact that today, on March 2nd, there are over ten thousand fewer people infected globally with the virus than two weeks ago?

We are bombarded with this ever-increasing number of total past infections which right now stands at about 85 thousand. But over 30 thousand of those have since recovered and no longer suffer from the virus. That number far outstrips the new cases that have been identified over that same period of time, decreasing the globally affected population by a substantial margin. This is a statistic that the media chooses not to focus on. 

It is a safe bet that many of the readers at this point intuitively rebel against this statistic and my- admittedly also selective- emphasis on it. Which brings us to the second quote of the day:

 “Most people use statistics like a drunk man uses a lamppost; more for support than illumination”. 

Andrew Lang

People tend to have an opinion on a matter and then look for statistical evidence to back up that opinion. Needless to say, the rational approach would be the opposite: look at the evidence and then form a perspective based on that. Unfortunately, COVID19 very much shows “support” rather than “illumination” being our main relationship with statistics.

For example, during the time period, you needed to read the above paragraphs, worldwide 20 people have died of a stroke. 15 thousand people each day. That is five times as many as all of the Covid19 fatalities thus far. In one day. 

I could easily paint a picture where I get an expert in to discuss the horrible process of having a stroke, the long-term effects for its survivors, etc. Any critical reader would then raise very good points as to why this only tells a partial story: it is a relatively stable number, it is not equally distributed among the population, etc. But the same applies to Covid19: currently, only those statistics that sound bad and scary are emphasised when it comes to the Coronavirus; not the whole story is being transmitted through media and word-of-mouth.

In case you argue that it is different because Covid19 is a virus and a stroke very much is not, well, yes. Firstly, in the above paragraph, I purposefully did not use the obvious common flu comparison, but it works with similar numbers. Secondly, so what? Do we care about genuine risks, or simply about the idea that viruses are like little aliens jumping from person to person? Fine if that gives you the creeps, but wouldn’t it make more sense to worry about statistically more dangerous threats?

Of course, there are some good arguments to pay special attention to new viruses such as this Corona variant. Crucially, if they are new, it means we do not have vaccines and our bodies are unlikely to have built up resistance yet. As a result, it is very difficult to know exactly how bad it will be. We just don’t have sufficient data (statistics) in the early stages. This is especially true for one that is relatively aggressive but does not display the same lethality as, say, Ebola. Uncertainty is always scarier than threats we know and understand well. That certainly justifies dedicating more emergency resources to an outbreak.

But the key is in the words “we don’t know”. That also means that we do not know that it will be worse than other threats, and all we can reasonably do is rationally follow statistical evidence as it trickles in to assess how serious the situation genuinely is. 

There are plenty of ways to look at the current crop of partial statistics about Covid19 to argue that we are horrendously overreacting. At the very least, there is no objective scientific consensus yet. How many articles in the media and how many conversations worldwide are currently making that case?

“I think of lotteries as a tax on the mathematically challenged.”  

Roger Jones

Perhaps the global economic downturn we are voluntarily creating in response to the virus outbreak is a more serious version of that: a tax on our global community for using Corona statistics like a drunk man uses a lamppost.