2015: Losing record; went to bowl game on a technicality; lost several 1-score games because the team didn't have the will to finish.
2016: 9-win season against the easiest schedule in memory. Or does everyone forget when we were 7-0, ranked in the top 10, and all the ESPN dudes laughing at us being undefeated (shortly before we took a big fall). Then getting stomped by Iowa to finish off the year (and losing to anyone with a pulse).
It's unfortunate that history only really focuses on the W-L record, i.e. it doesn't matter if we lose to Iowa or Wisconsin on a last-second FG 5 years in a row, the L still shows up on the record, just as much as if it were a 50-point blowout. And the "displacement" doesn't indicate the "acceleration", i.e. the W-L record doesn't give any clue as to if the team is getting better or gravitating into mediocrity.
Even counting the number of 1-score losses to compare MR to SF doesn't provide accurate data. Losing a 1-score game because you keep shooting yourself in the foot with penalties (but still giving good effort) is totally different from losing a 1-score game because you don't have the fortitude to finish. But those who use statistics to compare coaching success don't take those things into account.