Chapter 2
The Insight
Abraham Wald, a mathematician who had fled Austria after the Nazi invasion, was part of the Statistical Research Group at Columbia University. When he saw the damage map, he made a deceptively simple observation.
"You're looking at the wrong planes"
Wald pointed out that the data had a critical flaw: it only included planes that returned. The bombers that were shot down and never made it back were missing from the analysis entirely.
"The military," Wald explained, "is only examining the survivors."
Wald's logic was simple but profound: If a plane is hit in the wing and manages to return, it means the wing can survive being hit. But if almost no returning planes have bullet holes in the engine, it's not because the engine isn't being targeted.
It's because any plane hit in the engine crashes and never comes back to be counted.
Vital Organs vs. Surviving Parts
Wald's insight required understanding the difference between a hit that is "damage" and a hit that is "fatal."
The Wings and Tail
These are large surface areas. While they look bad when riddled with holes, a plane can often lose pieces of its wing or tail and still maintain enough lift to stay in the air.
The Engines and Cockpit
These are the "vital organs." A single bullet to the engine can cause a fire or seizure. A hit to the cockpit can incapacitate the crew. These hits are almost always fatal.
Therefore, the "missing" holes in the engines were actually the most important data points on the map. They represented the deaths that the military wasn't seeing.
The correct answer
Wald recommended adding armor to the areas with no red dots: the engines, the cockpit, and the fuel systems. These were the areas where a single hit was likely fatal.
The wings and tail, covered in red dots, were actually the safest places to be hit. Planes could take that damage and still limp home.
"The absence of evidence is not evidence of absence."
This is survivorship bias: drawing conclusions from only the survivors (or successes) while ignoring the failures that never made it into the data set.
Quick Check
Did you get the key idea?
What was the critical flaw in the military's damage data?
Abraham Wald recommended adding armor to the engines and cockpit — areas with almost NO bullet holes. Why?
Continue reading
Chapter 3: The Legacy →