How do you measure "statistical significance" for a benchmark? Run your model 10 times and, getting the same result each time, you conclude that variance is 0 and the significance is infty sigma?
So to get reasonable statistics, you would need to split your test set into say 10 parts and then you can calculate mean and average, but that is only a reasonable thing to do as long as it is cheap to gather data for the test set (and running the test set is fast).
Bill Belichick: The left gunner, look Charissa I think that is important for you to get right for your viewers, the left gunner on a line drive punt, if the receiving team is trying to return he has to release outside, that means if he's blocked he has to work through the block and try to stay ahead of the kicking team's players on the inside, of course that changes if the returner isn't right handed, but it is the responsibility of the punter to remember how the returner catches the ball and well, we allowed a return instead of a fair catch on the second punt for 5 yards, that put us back when they punted.
Charissa (after halftime): Coach Belichick told me they just have to execute.