Most people that follow sports to any significant degree are aware of gambling. Las Vegas (and all gambling establishments for that matter) generally work on the premise that the house gets a cut (called the vigorish (or 'vig' for short)) for taking the risk of taking the bet. Generally speaking it is known as the 11 for 10 rule meaning that one must $11 in hopes of winning $10. What many people don't know is what the purpose of this rule actually is: To equal the raw dollar amounts on each side of the 'line' so that the book always collects the vig and never loses money. To the bettor, what this means is that one must win roughly 52.4% of their bets to break even, so a coin flip for betting is a losing proposition; one will never break even (in fact one will lose the vig every time) if one wins exactly half their bets.
What must be remembered here is important due to the statistical, econometric and economic repercussions about the book seeking to divide the money equally on each side of the line: the book is not intended to predict the outcome of the event, but merely to balance the raw dollar figures on each side of the line. It stands to reason that given imperfect information (insider trading, if one deems) results in an inefficiency inherent in the line that makes it potentially profitable through statistical exploitation. Indeed, there are countless websites and betting hotlines that propose to be able to 'beat the house' and many have met with a degree of success if not a great deal. Therefore, any model which can explain more than 52.4% of the variation in game outcomes will be profitable on average. Should one feel the necessity, these concepts are explained in great detail by the works of Gandar and Zuber, et al and many others in numerous publications.
What most casual bettors do is play the 'underdog/favorite' line which is, basically, "Who will win and by how much? " This presents a bettor with two questions to answer: (1) Which team will win?; and (2) By how much will this team win? An easier proposition to handle is betting the over/under line which only requires the bettor to answer a single question: "Will the total number of points scored be more or less than the line?"
Many parlay cards offer a proposition similar to "Under 33 / Over 36." For the educated bettor, this is a losing proposition (remember, the house wouldn't be the house if it didn't win on average). As this end is unacceptable to the educated bettor, one must consider embracing the "3 for 3" or "4 for 4" game: a proposition in which the bettor only bets on the over/under line for exactly 3 or 4 games when winning all of them offers a significant return.
Some simple models can be offered and I have tested quite a few. What I can offer is this: if one takes into account a few number of significant variables, one can exploit the inefficiency created by the necessity of the line to balance the book.
