This writeup looks at a successful product with a small number of features that was thereby distinguished from a field of unsuccessful products with a large number of features. Accounting for many products, considering both successes and failures (i.e. using a wide selection of data), it argues that the distinguishing factor of less features was related to the device’s popularity.

In the canonical example of survivor bias, the only bombers being examined (for their characteristics) in the original flawed analysis were the ones that made it back; the planes that were shot down (and their characteristics) were not being considered — an error.