Whiskey: it's hard for me to say for sure, but this guy's WAY off what the other lists register. What he has done is gone back in time and not included early years. His games and wins totals all are seriously less in number for the historical teams. One can say that what he is doing is legitimate in that he has some definition in his head of football games which are "modern enough" to count, but we can see how that might be a tough criterion to settle upon. "Only after the invention of the forward pass?" "Only after getting rid of leather helmets?" "Only after the invention of the spread offense?"
I looked at the other statistical sets which go back to the "first game", and these sets calculate "winning percentage" by adding up actual wins plus half your ties. One could argue that if you tied, you didn't win, but no one seems to take that line. In those stats, Michigan is: 1253 games and 921 wins for a percentage of 73.50%. Notre Dame is 1207 games and 886 wins for a percentage of 73.41%. If Michigan loses its bowl game, and we win ours, we are still second in percentage by a narrow margin of about 00.02%. So we still have work to do next season.