We used to intentionally add wrong data to our datasets so we could circle it afterwards, declare it “outlying data” and gain the extra points for spotting and documenting it.

Nice trick!