Survivorship bias is a logical error much more common than we think. When I first heard about it, I was fascinated by how often we are misled by seemingly solid research, so I discussed it briefly in this article.
A little bit of history
Out of the many examples and stories that could be used to illustrate survivorship bias, my favourite dates back to World War Two.
At the time, the American military was looking for ways to protect airplanes from being shot down. They knew that adding armor would help but they couldn't protect the whole plane because it would become too heavy to fly.
In the face of that constraint, they had an idea: to examine the planes returning from combat, see where they were hit the most and add armour to those areas.
That makes sense, right? Adding extra protection to the areas where the planes were hit more often would make them more likely to survive combat. Wrong.
Fortunately for the military, they had staffed statistician Abraham Wald to aid them with their task and he quickly realized the analysis run by the military had one fundamental flaw: they had only examined the planes that made it back. That means the bullet holes they were looking at actually indicated the areas where a plane could be hit and keep flying and, therefore, the areas that did not need reinforcing.
Survivorship bias
Survivorship bias occurs when researchers focus on individuals, groups, or cases that have passed some sort of selection process while ignoring those who did not.
In a nutshell, that means when a study or observation is affected by survivorship bias, we only pay attention to part of the data, which can lead to a number of issues such as overly optimistic conclusions, seeing cause-and-effect relationships where there isn't one and bad decision-making in general.
Survivorship bias affects our perception in many areas of life and there are countless examples of situations where our judgment can be clouded in our everyday life.
A typical example is the popular misconception around the phrase “they don’t make them like they used to.”
We tend to think that goods manufactured in previous decades or things that were built centuries ago are just better than what we have today but that analysis doesn’t take into account that only the sturdier items have survived into the present day. Things that were destroyed in the meantime are not visible to us and thus are completely forgotten.
Survivorship bias in investing
In the context of finance and investing, survivorship bias manifests in the tendency to view the performance of existing stocks or funds in the market as a representative comprehensive sample without regard for those that have gone bust.
That's why disclaimers such as "do your own research" and "past performance is not indicative of future results" are so common in the world of finance.
Investors are often misguided by survivor bias when the published investment fund return data is unrealistically high because a company's poorly performing funds are closed and their returns are not included in the data. In this case, the data specifically related to those funds has already been weeded out, producing an inaccurate and incomplete picture of a company’s overall fund performance.
The same also happens in more informal conversations. People love to talk about their moves that worked well but they are not as fond about discussing the times when they got absolutely crushed.
Final thoughts
Survivorship bias is a common thing that affects many areas of our everyday life, including our financial decisions.
Doing your own research is very important and can help you avoid being misled by incomplete or inaccurate data.
Whenever you are gathering data for any kind of research, try to think about what kind of data could have been omitted and why and, finally, before removing any outliers, make sure you understand what they mean.
Posted Using InLeo Alpha