Survivorship bias also crops up in media portrayals of successful people, in finance, sports, cinema and music. Mutual funds are another classic example, where losing funds are either closed and/or merged into other funds, therefore increasing overall performance. There's also the idealized view of startup companies: We typically only hear about entrepreneurs and startups that have achieved great success but fail to realize that about half of all startups don't make it past five years because we never see their stories. Companies that go bankrupt or no longer exist for whatever reason are excluded from performance ratings, therefore improving the overall indicators of the other companies in a given report. There are other major instances to consider. This episode clearly illustrates the problem of survivorship bias: The examples that don't “survive” are not able to explain their side of the story, so all our considerations will be biased toward the “survivors," the winners, the ones left to expose their point of view. It would make sense to add more armor to the areas where they weren't hit. The planes they were analyzing only made it through combat because their most crucial parts were spared damage. But as Wald wisely observed, the best option was actually the exact opposite. After looking at the data sets displaying the location of bullet holes, the first impulse would be to advise protecting those same damaged spots.
0 kommentar(er)
