How do we know Baylor's offense is good? Rethinking total offense and defense rankings

A discussion on this thread this week got me thinking about whether there might be a better way to rank teams when we talk about offense and defense.  Here's the posts that did it.

So, a hat tip to SJD183 for inspiring this post.

Here's the problem.  Just as the posters pointed out above, Michigan State's #1 total defense ranking hasn't exactly come against Baylor.  And Baylor's #1 total offense ranking wasn't achieved by playing Michigan State.  It's reasonable to think that Michigan State's and Baylor's #1 rankings are just as much a result of the caliber of offense and defenses they've faced as the actual caliber of their teams.

If only there were a way to go beyond just an yards per game average in a way that might account for the quality of offense or defense a team faced.

Advanced Stats to the Rescue

Dividing a team's total yards of offense and defense by the opponent's season average yards of defense or offense gives a percentage that shows how many more or fewer yards a team was able to generate (or keep and opponent from generating) in a way that indexes for an opponent's offense or defense quality.

For instance, if Team A generates 500 yards of offense vs Team B, and Team B's season average defense is 300 yards per game, 500/300 = 1.67.  In this case, Team A generated 167% of the expected offense.  If Team C generates 400 yards of offense vs Team D, and Team's D's season average defense is 200 yards per game, Team C generated 200% of the expected offense.

Team A's YPG is 500 and Team C's is 400.  Is Team A's offense better?

Team C generated 200% of expected offense.  Team A generated 167% of expected offense.  Is Team A still better?

By averaging out these percentages for a team's offense and defense, we get a clearer picture of which team's are making the most of the opportunities presented to them regardless of quality of opponent.  Generating the same amount of offense vs a good defensive team will result in a higher percentage than vs a bad defensive team.

The next table ranks the offenses and defenses, thru Week 8, in this way.  Obviously, a higher number is better on offense, and a lower number is better on defense.

I think we can effectively dismiss the criticism that Michigan's State defense doesn't deserve its ranking because it hasn't played anyone.  The teams that it has played it has held to an average of 60% of their average season yardage.  Likewise, Baylor has made the most of the opportunities presented to it, averaging 168% of it's opponents' defensive yards per game.