Learning How to Weigh Different Variables in Sports BettingAdded - Nov. 3, 2014 Sports
The process of winning at sports betting is all about developing a system for ranking teams and their ability to score against each other. When you're developing this system, there are tons of variables that you'll have to take into consideration depending on what type of sport or event you're wagering on. Along these lines, it's not always clear to know how much weight you need to apply to each of the factors you're including in your evaluation. We want to give you some ideas on how you can figure out the relative weights for your own system.
First off, you should start by making a list of the different factors that come into your calculations used for establishing point totals and things of that nature. A lot of people use recursive methods that have to do with evaluating the strength of the offense and defense of different teams in games like American football and football/soccer. These recursions mean that a small error in your calculations will become magnified over time, so one place to start is to try to play with different values to see how much they will change your end results. The more a small change will upset your end results, the more delicate you have to be with how you weight it and the more conservative you have to be with it.
We recommend an approach that uses three different versions of your model at the same time. The differences between these models is that you'll have one that weighs defense higher, one that weighs offense higher and one that tries to be balanced. See how these models do over previous seasons by running game scores and results through your system in an effort to see which set of criteria performs the best. If your offensive-oriented or defensive-oriented scales work out much better than the rest, then you'll need to adjust things so that you have a new "balanced" version that more accurately represents your results.
From there, you can make two new offensive and defensive versions of your new baseline and see how things work out from there. In the long run, doing this type of refinement will help to give you a balanced set of variables that weigh all of the relevant factors as accurately as possible for the sake of your system.