I'm interested in how accurate (pinny) closing moneylines for MLB are. Basically I have a list with all the games of every team from 2009 with the closing lines and results, so each game is in the list twice with each listing having only one team's closing odds. Betting 1 on each event resulted in a -1.44% ROI but there's variance (lol samplesize) and I'm not sure if that number means anything anyway. Using 1.44% as vig you can get the probabilities for the games but all games obv don't have the same vig. Sum of all (probability)*.9856 for wins and (probability*.9856)-1 for losses was 2.58 or .05%/game (this should be 0 right? with proper sample/vig)
I'm a bit confused myself at this point too. Basically I have access to odds that aren't updated up to game time and I'm wondering if, and how much, betting there using closing lines as true probabilites is +ev, and how I could come to that conclusion myself using these closing line/result lists :)
