I agree the result (win/loss) is the most important thing. I disagree on impact of margin. Margin of victory/defeat is an additional datapoint in comparing teams — within some reason.
Maybe we’re talking “bump in selection” differently. In this case, Michigan losing 2 games to a Top5 ranked Purdue by a total of 5 points…tends to support Michigan’s ranking somewhere in the 20s. However, had Michigan lost this game in similar fashion to the game at Nebraska, maybe the first meeting (last possession game) looks like a fluke and maybe Michigan isn’t really that close to top competition. Same win/loss total, different evaluation of the results.
(In my opinion not much difference between say a 16-17 point win and a 25-30 point win — comfortable margin either way; no need to run up the score beyond a range where you’re coasting to victory)