[CTcard]

I think this idea of "beating good teams" is too cute by half. Alabama gets every benefit of the doubt and outside of a couple good SEC teams regularly plays no one. FSU played absolutely no one last year and got into the playoff. OSU played a good MSU squad and no one else and still got in with one loss. Playing lots of good teams does not matter. In fact, the recipe seems to be "play 1-3 good conference teams and no one else and lose at most one of those games."

I think that is a little too simple description of what happened last year, but your bottom line mostly holds.

Last year the #1 criterion for making the playoffs was the number of losses (well after being in a Power 5 conference).

There was one undefeated team, FSU. They got in.

There were 5 one loss teams, they got all the real consideration for the other spots. At that point, strength of schedule comes into play.

Alabama (Sagarin SOS=2) and Oregon (SOS=20) got in fairly easily.

The question came down to Ohio State, TCU, Baylor for the last spot.
The Sagarin SOS ratings: OSU #29, TCU #51, Baylor #59. (and Sagarin rather liked the Big 12)
Bowl opponents:
OSU = Navy, Va Tech, Cincinnati, Maryland, Rutgers, Penn State, Illinois, Michigan State, Minnesota, Wisconsin
TCU = Minnesota, Oklahoma, Baylor, Ok State, W. Va, Kansas State, Texas
Baylor = Texas, TCU, W. Va, Oklahoma, Ok State, Kansas State

And of course tOSU absolutely destroyed an okay Wisconsin team the final week of the season.

------------
The lesson is that not losing matters the most, no matter what schedule you play. I guess that is another way of saying "playing lots of good teams doesn't matter".  After that, a combination of SOS and impressive late victories separates ties in the loss column.

[CTcard]

I think this idea of "beating good teams" is too cute by half. Alabama gets every benefit of the doubt and outside of a couple good SEC teams regularly plays no one. FSU played absolutely no one last year and got into the playoff. OSU played a good MSU squad and no one else and still got in with one loss. Playing lots of good teams does not matter. In fact, the recipe seems to be "play 1-3 good conference teams and no one else and lose at most one of those games."

I think that is a little too simple description of what happened last year, but your bottom line mostly holds.

Last year the #1 criterion for making the playoffs was the number of losses (well after being in a Power 5 conference).

There was one undefeated team, FSU. They got in.

There were 5 one loss teams, they got all the real consideration for the other spots. At that point, strength of schedule comes into play.

Alabama (Sagarin SOS=2) and Oregon (SOS=20) got in fairly easily.

The question came down to Ohio State, TCU, Baylor for the last spot.
The Sagarin SOS ratings: OSU #29, TCU #51, Baylor #59. (and Sagarin rather liked the Big 12)
Bowl opponents:
OSU = Navy, Va Tech, Cincinnati, Maryland, Rutgers, Penn State, Illinois, Michigan State, Minnesota, Wisconsin
TCU = Minnesota, Oklahoma, Baylor, Ok State, W. Va, Kansas State, Texas
Baylor = Texas, TCU, W. Va, Oklahoma, Ok State, Kansas State

And of course tOSU absolutely destroyed an okay Wisconsin team the final week of the season.

------------
The lesson is that not losing matters the most, no matter what schedule you play. I guess that is another way of saying "playing lots of good teams doesn't matter".  After that, a combination of SOS and impressive late victories separates ties in the loss column.