Survivorship bias
Survivorship bias occurs when through excluding samples from analysis by (unintentional) selection, focussing on only the selected samples leads to wrong conclusions.
The best known example for this cognitive bias is Abraham Wald’s work on aircraft returning from WWII flying missions: counterintuitively the bullet holes statistically observed are a bad indication on where they planes should be enforced. On the contrary: these positions are the ones strong enough to take a hit and still let the plane return. The positions with no bullet holes should be enforced instead as they are the ones that lead to crash.
This specific type of samplimg bias has been found in diverse domains from economics to cats falling from high buildings. Interestingly, even science itself might systematically skew results through the tendency to publish positive results.