Michigan is a perennial 3 loss team, which is what made Mike Hart's comments on that subject so foolish.
Let's use your 20 year metric for UM (bold = 4+ loss seasons)
3+ UM losses in 90, 93, 94, 95, 96, 99, 00, 01, 02, 03, 04, 05, 07, 08, 09
0-2 UM losses in 91, 92 (3 ties), 97, 98, 06
TO = Tom Osborne, AO = After Osborne
3+ NU TO losses in 90, 92
0-2 NU TO losses in 91, 93, 94, 95, 96, 97
3+ NU AO losses in 98, 02, 04, 05, 06, 07, 08, 09
0-2 NU AO losses in 99, 00, 01, 03
UM lost 3 games 75% of the time, and did so consistently with every coach.
NU lost 3 games 50% of the time, and the vast majority of those were After Osborne, circa 98-09
Let's look at tom's entire career:
3+ losses in first six years: 74, 76, 77, 78
0-2 losses in first six years: 73, 75
3+ losses in remaining yrs: 81, 85
0-2 losses in remaining yrs: 79, 80, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97
Osborne lost 3 games 5 times in his first 6 years.
Osborne lost 3 games 2 times in his remaining 19 years.
83% win pct Osborne... lost 4 games ZERO times.
75% win pct Carr......... lost 4 games 3 times.
76% win pct Moeller..... lost 4 games 4 times.
79% win pct Bo........... lost 4 games 4 times.
You treat the 94-97 stretch as though it is an outlier, when those four years stack up favorably by themselves, when really NU was much more consistent in their wins & championships than UM.
You also treat the Callahan & Rodriguez years as outliers, but that only strengthens the case for Nebraska.