Diogenes was asked concerning paintings of those who had escaped shipwreck: “Look, you who think the gods have no care of human things, what do you say to so many persons preserved from death by their especial favor?”, to which Diogenes replied: “Why, I say that their pictures are not here who were cast away, who are by much the greater number.”
Cicero, De Natura Deor., i. 37.
Survivorship bias is a cognitive bias which occurs in data analysis. The fallacy refers to our tendency to focus only on the winners and neglect the losers. It is a form of selection bias. To put it simply, most analyses do not include failed instances, and thus, they usually offer a misleading conclusion. In order to avoid this, we should include all the variables in the analysis.
When it comes to survivorship bias there are tons of examples to try and describe it with. For instance, when someone argues that soccer players earn lots of money and THEREFORE it would be a good decision to become one. However, there is a major issue associated with such analysis. Here, we are ignoring all the individuals who have pursued becoming a soccer player and have either stopped trying or have lost the battle somewhere in the path. So, by omitting the ones who have been ‘unsuccessful’ we are unconsciously increasing the success rate for becoming a soccer star.
Generally, people tend to overestimate their chances of winning in every situation and survivorship bias is one of the fallacies that lead to such false conclusions. For example, if a certain economic school has introduced several notable economists, one might find it easy to conclude that there is a link between attending that certain school and becoming a Nobel Prize winner in economics. Although such reasoning may be actually true to some extent, we have to include the failed students in the analysis as well. This way the outcome will be closer to reality.