When practicing marksmanship, do you first shoot at something only then to draw a target around the bullet holes (known as the Texas sharpshooter fallacy)? Hopefully the answer is no. Yet, this is something that would come naturally to us. Gary Klein informs us in A Data/Frame Model of Sensemaking, in Expertise Out of Context, that our brains automatically fit data into a frame and fit the frame around that data. We have the tendency to interpret patterns where none exist.
In the case of the coronavirus disease of the year 2019 (COVID-19), which is caused by a coronavirus, due to increasing cases, we seem to have a very serious issue on our hands in the U.S. Or do we? Nate Silver writes in Coronavirus Case Counts are Meaningless*,
“A country where the case count is increasing because it’s doing more testing, for instance, might actually be getting its epidemic under control. Alternatively, in a country where the reported number of new cases is declining, the situation could actually be getting worse, either because the system is too overwhelmed to do adequate testing or because it’s ramping down on testing for PR reasons.”
We have tested 0.8% of the population
According to The COVID Tracking Project, it appears as though some of the first COVID-19 tests in the U.S. began in late February (approximately February 28, 2020). As of April 12, the U.S. has had 551,826 people test positive for COVID-19 out of 2,805,892 total tests (approximately 20% of those tested). There are nearly 330 million people in the U.S. and we have only tested 0.8% of the population.
Does this tell us anything? It tells us one thing: we will continue to see an increase in the number of positive cases as we increase the number of people tested.
New York versus Kansas
The increase is largely dependent on where we test. For example, let’s examine one of the hardest-hit areas: New York. On March 11, there were only 308 people tested in New York, with 216 positive cases. On March 25, New York had tested 103,479 people with 30,811 positive cases. As of April 12, New York has tested 461,601 people, with 188,694 positive cases (41%).
Compare this with an area that has not been hit as hard: Kansas. The first tests in Kansas were conducted on March 6, 2020. On March 12, only 45 people in Kansas had been tested, with 4 positive cases. As of April 12, 13,253 people in Kansas have been tested with 1,337 positive cases (10%). The only thing consistent in both New York and Kansas is that we are testing more people every day and more positive cases are being confirmed.
Send in the Army!
If the goal is to stop the spread of the virus, then we might want to reframe how we are measuring the spread. With the increased role of the U.S. Military to fight the virus (not sure this is possible), we are adding yet another variable: media coverage.
Thus, the more we test, the more positive cases we report, the more we deploy the military to fight the virus, the more the media reports on the military fighting the virus, leading to an increased effort to test more people to lower the number people testing positive for the virus. See the problem here?
The Cobra Effect
This problem reminds me of something called the cobra effect, which is when an attempted solution to a problem makes the problem worse as a type of unintended consequence. The cobra effect originated during the time of the British rule of colonial India.
The British government was concerned about the number of venomous cobras in Delhi. The government, therefore, offered a bounty for every dead cobra. Initially, this was a successful strategy as large numbers of snakes were killed for the reward. Eventually, however, enterprising people began to breed cobras for the income. When the government became aware of this, the reward program was scrapped causing the cobra breeders to set the worthless snakes free; the wild cobra population further increased.
With COVID-19, the cobra effect can be summarized as:
Intention: Reduce the spread of COVID-19 in the U.S.
Action: Increase the number of tests for COVID-19.
Effect: Cases rise as the number of tests increase.
Essentially, we have fired a large number of bullets (COVID-19 tests) into a door only to then draw a target around those bullet holes.
A couple more ways to look at this are as follows:
Goodhart’s Law. When a measure becomes a target, it ceases to be a good measure. If we measure based on the number of tests, then we will see a daily increase in the number of positive COVID-19 cases.
Campbell’s Law. The more a metric counts for real decisions (number of COVID-19 tests), the greater the pressure for corruption (testing leads to more positive cases, thus we must test more), the more it distorts the situation it’s intended to monitor (the spread of the virus).
Therefore, the more we test for COVID-19, the more positive cases we will have. If we conclude that the pandemic is worsening because of this, then we are looking at and measuring the problem wrong.
Reframe the problem
I propose we reframe the problem. We should look at the data from a different angle and ask the following questions:
Question #1. What’s happening?
Answer: A daily increase in the number of tests and positive cases.
Question #2. Why is this happening now?
Answer: Because we are testing more people every day.
Question #3. What is the overall impact so far?
Answer: We have a false view of the actual problem.
Question #4. How should we look at this problem?
Answer: We can’t have our cake and eat it too… We must change how “Big Pharma” is incentivized. In Coronavirus: The Latest Problem Big Pharma Won’t Solve, Ellen ‘t Hoen writes,
“The global health emergency caused by the new coronavirus outbreak should bring countries, international organizations, NGOs, investors and industry together to agree on a new social contract where industry acknowledges it cannot have and eat the cake: If more public money is going to be spent directly on the development of new products and vaccines, the price of the products has to be closer to the marginal cost of production. Governments then need to keep their end of the bargain and invest in robust health systems to provide prevention, treatment, and care. Drug markets will not fix themselves.”
Question #5. What’s next?
Answer: Focus on changing incentives so that “Big Pharma” is incentivized to produce vaccines quicker — and proactively look to future diseases/pandemics.
Question #6. What are the implications if we fail to reframe?
Answer: This crisis will become so horrific that we will deploy the Space Force to miniaturize Dennis Quaid and inject him into people with COVID-19 to kill the virus.