The season prediction thread got me to dust off an old stats program and run some scenarios.
If you had to list a win % for each game, what would it look like?
Mine looks like this:
UNI 80% IOWA 50% atToledo 75% KU 55% atTech 40% TCU 5% at Baylor 2% Texas 30% atOU 2% OSU 25% atKSU 35% atWVU 40%
UNI - 65%
Iowa - 50%
at Toledo - 45%
Kansas - 70%
at Texas Tech - 40%
TCU - 1%
at Baylor - 0.5%
Texas - 35%
at Oklahoma - 2%
Oklahoma State - 7%
at Kansas State - 35%
at West Virginia - 40%
I had a genius stats prof when I got my MBA. He said that when you have to predict probabilities, in the absence of actual data of prior outcomes, then there are only 4 useful %s to consider.
10% = very unlikely, but not impossible
40% = less likely than 50/50 but close
60% = more likely than 50/50 but close
90% = very likely but not metaphysical certitude
I have found this to be very practical to use in everyday life and work, because you usually don't have any prior outcomes data.
UNI: 90%
Iowa: 60%
Toledo: 40%
Kansas: 90%
Texas Tech: 40%
TCU: 10%
Baylor: 0%
Texas: 60%
Oklahoma: <0%
Oklahoma State: 40%
Kansas State: 40%
West Virginia: 40%
Yes it looks like I broke my own rule on BU and OU, but I consider those metaphysical certitude losses. I do think we have a punchers chance at home vs TCU.
This all leads me to think 5-7. I think the O-line will be better and the team will be competent. We'll know after UNI.