[M T]

MT, don't you mean 9 beating 8 gets 1 point?

You are correct.  (I fixed it)

However--isn't there always something? Doesn't the Big 10 get two points for two one-seed-difference upsets? Think they should have two points, not zero, unless I'm missing something.

You are correct.  (I fixed it)

This isn't meant to be perfect--more of a back-of-the-envelope estimate.  It definitely doesn't credit a one seed for beating a 16, which is a small devaluation for the one seed.  Nor does it credit an 8 seed beating a 9 (if I do my sums correctly!), which seems more unfair.    I like the idea of using the historical probabilities (perhaps after adding noise to smooth them out a bit), but that's harder to calculate.    But to say one conference is better because it has more teams in the tournament is like  ....[example removed because it references politics] ....

[M T]

MT, don't you mean 9 beating 8 gets 1 point?

You are correct.  (I fixed it)

However--isn't there always something? Doesn't the Big 10 get two points for two one-seed-difference upsets? Think they should have two points, not zero, unless I'm missing something.

You are correct.  (I fixed it)

This isn't meant to be perfect--more of a back-of-the-envelope estimate.  It definitely doesn't credit a one seed for beating a 16, which is a small devaluation for the one seed.  Nor does it credit an 8 seed beating a 9 (if I do my sums correctly!), which seems more unfair.    I like the idea of using the historical probabilities (perhaps after adding noise to smooth them out a bit), but that's harder to calculate.    But to say one conference is better because it has more teams in the tournament is like  ....[example removed because it references politics] ....