Image Plane-bullet-holes.png cutline: Source: McGeddon, “Survivorship Bias,” Wikimedia Commons, 
https://commons.wikimedia.org/wiki/File:Survivorship-bias.png
A hypothetical diagram of the bullet hole patterns observed by Abraham Wald during his analysis of World War II war planes that returned safe from battle. Image from

By Scott Hamilton

Senior Expert Emerging Technologies

There have been many times in history that loads of data have been available to solve problems, but the problems were either not solved, or solved incorrectly. We have data available today for tracking diseases back to the point of origin, but the data does not usually help in stopping the spread of disease. Looking at the data from the spread of COVID-19, it would have seemed that locking down the population to prevent the spread of the disease would work. However, in practice, it does not seem to have worked very well. The disease seems to be spreading more rapidly in areas with tighter travel restrictions and lock-down than in areas with relaxed regulations.

See the status of the spread of COVID-19 reminded me of a story my Grandfather told me. He was one of the Sea-bees during World War II. They were the construction battalions, and one of the first teams on the shores of Normandy on D-Day. He told the stories of his buddies in the Air Force speaking about the engineering efforts being done to make the planes safer.

They were studying the bullet hole patterns on all the planes that returned from battle. They noticed that most of the planes had bullet holes concentrated on the wings and bodies of the planes. The engineers began drawing up plans on how to reinforce the body and wings without adding too much weight to the plane. The area they felt they needed to over was well over 70% of the surface area of the plane.

After several months of planning and studying the structure, a single Hungarian-Jewish statistician named Abraham Wald reviewed that data and their plans. He had a completely different interpretation of the data. He released very quickly that they were studying planes that made it back from the battles. This meant that planes shot in the wings and bodies usually survived the attack. He suggested that instead they focus on reinforcing the areas of the plane that had no signs of damage, as these were more than likely the critical components of the plane.

Following Wald’s interpretation of the data saved the lives of thousands of Air Force pilots as well as millions of dollars on materials and engineering to make the aircraft safer. The efforts were focused on reinforcing the cockpit, engines and tail components of the planes. The results of fewer fatalities, safer and more successful missions lead to applying the results of his study to military aircraft design up through the end of the Vietnam War.

To me this is one of the most influential stories when it comes to interpreting data and how it is important not only to look at the information you have, but also the information that is missing from the study before jumping to any conclusions. There are way too many times we make decisions in life based on assumptions rather than taking the time to gather all the necessary information before making a decision. I challenge you all this week to think about an area of your life that you have been working to improve and see if you are missing something important by focus in what you know rather than what you do not yet know. Stay safe an learn something new.

Scott Hamilton is a Senior Expert in Emerging Technologies at ATOS and can be reached with questions and comments via email to sc************@at**.net or through his website at https://www.techshepherd.org.

Share via
Copy link
Powered by Social Snap