43,
I don't know if it was Prime Time, or you who decided to throw out the Chief's game but you can't throw that score out. That number, be it 21 or 14 is statistically valid. Outliers should have an assignable cause to justify their discard. Measurement error? No. Un-normal factors at work? No. You need to take the full data set to get the most accurate picture possible.
I have seen a lot of dishonest, or poor quality stats work, especially in the engineering and scientific community. If the engineers or scientists have a vested interest in getting a number they want, they find a way to throw out data they don't like, those that take them away from their desired target. There are so many ways to cheat.
Edited 2 time(s). Last edit at 12/04/2018 01:49PM by Coy Bacon.