I don't have time to build models and I failed 11th grade math 3 times, so I try to keep it simple. Has anyone considered using the posted games on ESPN's Streak For The Cash as a tool for handicapping?
One would think the "Streakmaster", as they call him, has incentive to post plays that will bury the public because the first to hit a predetermined streak (27 wins this month) gets paid.
I have no clue how this would fit into a larger model but I try picking my spots to fade the public, generally on TV games, and it seems to work pretty well. I have Dayton +3 and Arksansas -1.5 so far today.
I always look at what matches they have on there and like if they have a straight up match on there and the books have one of the teams -5 or so, I stay away ... I figure ESPN feels like it could be a trap ... I dont pay attention to the percent of players picking what game though