We were mediocre, that is what our record says.

Outliers are not meant to be casually discarded because they indicate a problem. The cause of the outlier should be considered and perhaps investigated if time, money, and experience agrees to it.

I have seen time and time again, the best technical gurus discard an "outlier" as merely being a one off, whereas it was really the tip of an avalanche, and millions of dollars and weeks of time, were consequently wasted.