As best I can figure it redietz thinks that the use of mathematics implies the belief that mathematical models necessarily yield correct results, which is obviously untrue.

People make casual calculations all the time ("I think there's a 50/50 chance we win if we reject this settlement offer and go forward, and if we do win we'll win $X, therefore...") knowing that the premises of the model and its inputs are not necessarily correct.

If the judge is corrupt and therefore the lawyer's likelihood of winning is not what his historical record would suggest, does that mean his method of estimation was wrong? Should he have used some alternative "mathematics of non-random events"?