After spending this past week making - and then this week implementing - plans with our team here to move our seminary completely online during the COVID-19 outbreak, I thought I would take a reprieve, brush off an old hobby, and do a little math with you. For we need to make sure we are getting the right message when we look at these graphs everyone is producing regarding this virus.
Recently I have been reading Humble Pi, a book where the author tells you what it is really about in the subtitle: When Math Goes Wrong in the Real World. From advertising agencies promising Pepsi points could buy a military jet to McDonald's wrongly promoting you could combine their eight menu items in 40,312 ways (actual answer is 247), Matt Parker writes engagingly about how often math is used wrongly. How true his premise is, as he demonstrates repeatedly.
The most recent example of this propensity was delivered to us by newscaster Brian Williams. He concluded with a guest on a news program that failed presidential candidate Michael Bloomberg spent over $1 million per American in his campaign. How? Because, as they read from a Tweet, Bloomberg spent $500 million on his campaign and there are 327 million Americans, hence over $1 million per American (actual answer is $1.53 per American). Wrong math can be gloriously humorous.
But not when it creates undue fear and panic. Though I believe there is good reason to be cautious and heed guidelines given to us by our authorities, we still need to make sure we understand a few things about math when looking at the predictive models regarding the Corona virus. Some mathematical models explaining concepts can be helpful, such as these models in the Washington Post showing how the disease spreads and how effective different forms of prevention can be. Yet other math I have seen is tentative at best.
Here then are three relatively simple concepts you should remember as you evaluate the information coming your way.
Extrapolation. This action is when mathematicians, statisticians, or scientists estimate or make a conclusion by assuming that existing trends will continue as they currently are behaving. For example, one can look at the data points on a graph, in this case the number of cases worldwide of those infected by the Corona virus on a given date, and see that it is curving up. From those data points, one can then follow the curve and predict that it will continue upward at the same rate.
However, one has to be careful extrapolating. As the video below shows, changes can occur that can greatly reduce the future growth. The efforts of governments and people currently to reduce contact will "flatten the curve", reducing the rate of growth, and, Lord willing, cause that rate to eventually decline. So be careful not to think that because, for instance, the death rate doubles in a country for a few days that it will just continue to do so. This type of wrong math can produce unwarranted fear.
Variables. Often mathematical models are built on simplistic measurements that only take into account certain variables. Most of the graphs you see simply have two variables. The graphs have the date on the horizontal access and the number of cases of the virus on the vertical axis. However, in the real world there are many variables these simple mathematical models do not take into account.
The video above shows more of the variables, such as the number of people an infected person is actually exposed to or the actual number of cases in a given day, and how that affects the growth rate. Other factors include age, health, and population behaviors. Many variables come into play when we are considering a pandemic. Sunshine and fresh air caused a strong reduction in cases in the 1918 pandemic of the Spanish Flu. Likewise, the CDC in the bar graph here shows the peak months of viruses, and that they virtually die out come April due to warm weather.
Seeing these things, we should pray that the Lord would not only help people avoid contracting the virus. We should also ask that He would mercifully bring warm weather and sunny days to help curtail this virus and that it will not continue unabated into the summer.
Human Behavior. If there is any variable that is truly unpredictable, it is that of human behavior. Math can model inanimate objects very accurately, such as the tides of the sea or the orbit of planets. But when it comes to human behavior, watch out. People are incredibly unpredictable anyway, and that becomes especially true for mathematical models seeking to project the spread of a disease.
Think of a hurricane for a moment. Do you recall how difficult it is for scientists and meteorologists to predict the path where a hurricane's landfall might be? Part of that is because sea currents, wind patterns, temperatures, barometric pressures, tides, ocean depths, other weather systems, etc., all factor into determining where the hurricane eventually lands.
Similarly, human behavior is generally quite unpredictable. Social pressures, personal desires, moods, sexual motivations, political ideology, work pressures, geographic culture, other illnesses, relationships, etc., can all factor into how a disease might spread. For instance, think of how just simply fear helps in curtailing this virus. Seeing a simple upward curve is, at best, just a simplistic predication of an otherwise unpredictable event.
Please hear me accurately. We should do what we can to help stop the spread of this awful virus. But as we do, let us remember that God's math formulas are often far more complex - and far more merciful - than we can ever hope to calculate.