Skip to main content
Betting with a Model

A model can improve your betting process

Steven avatar
Written by Steven
Updated over 3 years ago

An everyday issue that a bettor has to solve is that of finding profitable betting opportunities. Through the use of a model, you can make your entire betting process more objective and testable. This can give the bettor an understanding of his level of profitability and confidence that his strategy will be successful long term.

Model Basics

A model is a system that can calculate odds of a particular event occurring. In the context of sportsbetting, this system is likely calculating the odds of a team winning a particular map, or some prop bet on that map.

There are a variety of approaches you can take to build a model - for this article we will assume you have a model in place.

Since we have a model in place, this means that for a particular betting opportunity, we can systematically produce our model's probability of that outcome happening.

Finding Value Bets

If we have confidence in the probabilities that our model is producing, then we can use our model to place value bets. Recall our formula for expected value:

Expected Value = (Model Probability) * (Bet Payout) - Amount Wagered

Note that the Bet Payout can be written as Amount Wagered * Odds Offered (using Decimal odds). The we can rewrite our expected value formula:

Expected Value = (Model Probability * Odds Offered - 1) * (Amount Wagered)

Since the amount wagered is positive, we see that Expected Value is positive only if Model Probability x Odds Offered - 1 is positive. We say that Model Probability x Odds Offered - 1 is the Model Edge, and we have positive expected value when our model edge is positive.

Expected Value = Model Edge * Amount Wagered
Model Edge = Model Probability * Bookmaker Odds - 1

To find Value Bets (bets with positive expected value), we calculate model edge for every bet offered by our bookmaker. When model edge is positive, we've found a value bet - a bet that should make us profit in the long run.

Did this answer your question?