Best Conference? SEC? No. PAC10? No. Big East! Actually, a tie?

rwperu at the twoplustwo forums brought up an interesting point and asked if the on-the-field results ranking (OTFRR) methodology could shed some light on the best conference.

GREAT IDEA!

I thought it could and, judging by the results,
it definitely does. I've always had a problem
with traditional conference ratings. Talking
heads on tv are just that--talking heads. The
computer rankings of conferences all suffer
from the same flaws that the rpi suffers from
in NCAAB, a linear weighting of opponents.
By a linear weighting, it says that a
conference that has the number 1, number
28, and number 61 team in the land is the
same as the conference with number 29,
number 30, number 31. Both average #30.
However, which of those is truly a better
conference? It's the age old debate of
weighting the top heavy verse the depth
heavy and all the combinations in between.
To paint a fuller picture of conference
strength, I decided to use the OTFRR to
test each conference. I tested to see the
expected win and loss total for the #5, #15,
and #50 teams in the country playing the
full conference slate.  By looking at the
records, one obtains the win pct a team of
that quality would have playing against
every team in the conference. The results
are somewhat surprising. It seems most
talking heads have assumed this year to be
a two horse race between the SEC and Pac10.
The numbers make it seem a 3 horse race,
with the dark horse, the Big East, actually
being a nose ahead of the others. However,
it turns out the difference between the top 3,
Big East, SEC, and PAC10, are extremely
small and within therange of error. What we
actually have is strong competition
throughout the college landscape with some
subtle differencesthat get blown out of
proportion.

Here are the main results:
For a #5 team in the land playing every conference member:
Big East 6.042W 1.958L 0.755winp
SEC 9.129W 2.871L 0.761winp
PAC10 7.639W 2.361L 0.764winp
BIGXII 9.361W 2.639L 0.780winp
Big11Ten 9.374W 1.626L 0.852winp
ACC 10.602W 1.398L 0.884winp

For a #15 team
Big East 4.830W 3.170L 0.6038winp
SEC 7.249W 4.751L 0.6041winp
PAC10 6.052W 3.948L 0.605winp
BIGXII 7.607W 4.393L 0.634winp
Big11Ten 8.016W 2.984L 0.729winp
ACC 9.082W 2.918L 0.757winp

For a #50 team
SEC 4.543W 7.457L 0.379winp
PAC10 3.811W 6.190L 0.381winp
Big East 3.143W 4.858L 0.393winp
BigXII 5.086W 6.914L 0.424winp
Big11Ten 5.562W 5.438L 0.506winp
ACC 6.139W 5.861L 0.512winp

Many things are pretty telling from the analysis.
1) Looks like the talking heads are mainly right
--ACC isn't good (a #50 team would be above .500)
--Big11Ten isn't good (again, #50 > .500)
--BIGXII is somewhere in the middle
--SEC and PAC10 are good
2) Big East is severely underrated
3) The differences aren't that much on the whole
--hundredths and even thousandths of a proportion seperating
the top 3 in numerous categories.
4) The difference non-linearities create
--Sagarin has SEC a clear #1...this analysis, which uses his base,
shows otherwise.
5) The better teams have it tougher running through the
BigEast. The lower teams have a tougher road through the SEC.