Survivorship Bias

During World War II the RAF lost a ton of planes to the German anti-aircraft fire. Everyone was searching for ideas on how to armor the aircraft so they would withstand the artillery. When a flight returned safely, they would examine the bullet holes then configure re-enforcement armor in those places. Sounds like a good idea, right? Not to Hungarian born mathematician Abraham Wald. Wald determined to examine the planes that return safely with bullet holes was not going to provide accurate information. It would be misleading. Wald wanted to inspect the aircraft that did not come back because those planes would give him a better indication of the aircraft's vulnerable spots.

In Jordan Ellenberg's book, How Not to Be Wrong, Wald's insight was to ask: where are the missing holes? "The ones that would have been all over the engine casing, if the damage had been spread equally across the body of the aircraft. Abraham Wald was pretty sure he knew. The missing bullet holes were on the missing planes. The reason planes were coming back with fewer hits to the engine is because the planes that got hit in the engine weren't returning. It seems so obvious, yet only one person seems to understand."

What Abraham Wald was able to determine about the planes was called "Survivorship Bias," which focuses only on the positive results or data points. The best way to explain this bias comes from former CBS Sports Analyst Jimmy "The Greek" Snyder. Snyder would predict three NFL games a week on his show, and a half-hour later on his radio show select the opposite teams. To half of the audience, Snyder looked like a genius. The other half the Greek didn't care. We tend to focus on just one half of the problem and give no thought to the other half.

We allow our egos to focus on the simple solution and the easy path with less resistance. Author Jordan Ellenberg believes using math as an extension of common sense will allow us not to avoid the obvious. While you and I might not be as intellectually gifted as Abraham Wald, we can set up procedures and systems to ensure we examine every side of the problem. Always double-check the information that does not seem relevant.

Today before making any decisions spend time looking at the disregarded data. You might find a solution in the none obvious.

Please forward and share this email with your friends and family.