Here's an excerpt from an earlier post describing what I call "Fair Moneyline Odds" (FMO):
Use week 3's DEN at OAK -125/+105 game as an example. I strictly use decimal odds* in all of my calculations, it's much easier to use in calculations. The decimal odds are DEN 1.80, OAK 2.05.
NFLSim determined the probability of DEN winning to be 77%. To determine the Fair Moneyline Odds, divide 1 by the probability (as a decimal). 1/0.77 = 1.30. Assuming the NFLSim has assigned the correct probability - and this only works if it does - then out of 100 games, DEN will win 77 of them. Starting with $100 and making 100 $1 bets, you will win 77 of bets. At a payout of 1.30, 77 winning bets gets you back to $100, you break even. Any payout larger than 1.30 will produce a profit after 77 wins.
In the case of the DEN game, DEN offers a 1.80 payout. At NFLSim's probability, 77%, you are expected to win 77 bets at 1.80. 77*1.80 = $138.6. There is an expected profit of $39. A payout of 1.80 is greater than the break even odds of 1.30 (the Fair Moneyline Odds), so it is a good bet.
On the other side of the matchup is OAK 2.05. The probability of OAK winning is 23%. The FMO would be 1/0.23 = 4.35. If you win 23 bets at a payout of 4.35, you break even at $100. Currently, Vegas pays 2.05 for an OAK win. If you win 23 bets at a payout of 2.05, you end up with $47.15 for an expected loss of $52.85. A payout of 2.05 is less then the FMO of 4.35, this is a bad bet.
In the spreadsheet, I list the break even moneyline odds for the specified team. If your sportsbook has odds greater than what is listed for that team, it is good to place that moneyline bet. However, if the odds are less than break even, it is a good idea to bet on the opposing team (even if they're picked to lose). With this system, it is irrelevant who you think will win the game. This is all about exploiting the inefficiency in the line.
*To convert American odds to decimal odds:
For the underdog, add 100 then divide by 100. OAK +105 => (105 + 100)/100 = 2.05
For the favorite, divide -100 by the odds, add 1. NYJ -125 => -100/-125 + 1 = 1.80
Showing posts with label odds conversion. Show all posts
Showing posts with label odds conversion. Show all posts
Sunday, September 27, 2009
Subscribe to:
Posts (Atom)