This low probability of an accident actually happening suggests that the condition probabilities presented in the article have huge standard errors. This may m

This low probability of an accident actually happening suggests that the condition probabilities presented in the article have huge standard errors. This may make the differences between sections of the plane insignificant. Then again, there aren't standard errors per se because we should have the population of all accidents rather than a sample. Interesting....