I've been trying to write a program that will pick me 80% winners (su). As I've said before anything above 67% has been impossible. I started to get close by combining some aspects of the different programs but as soon as I posted I got blasted that night.  I tweeked a few things and am back. Lets see where this gets me. Or gets us if this turns out to be able to produce profits.

My system gives every single favorite a numerical rating. The top half of those teams win straight up about 85% of the time over my two weeks of data. I should probably back test further than when I started keeping this information that would be a giant pain in my ass and take some of the fun out of this process.

Now the bottom half of the favorites win less than 50% of the time. This is where I'm going to try and exploit the books.

There were 12 teams in the bottom half of my ratings. I narrowed them down to my favorite 6 and put them in a round robin.

• Oklahoma City at Atlanta (Moneyline ) Oklahoma City
  3.30
• Towson at St Josephs (Moneyline ) Towson
  3.85
• Western Ky at Middle Tenn St (Moneyline ) Middle Tenn St
  2.15
• Canisius at Fairfield (Moneyline ) Canisius
  3.90
• NC Wilmington at Virginia NC Wilmington + 17.0
  1.91
• San Antonio at New Orleans San Antonio
  1.91