The Oracle method is a mathematical ranking method to evaluate paired comparisons. The Oracle takes into consideration only basic data about each match: The two teams that played, the outcome, and the points each team scored.
About the method: The Oracle views the tournament as a network where the flow of information flows from one team to another. The Oracle automatically consider information about each team opponents, their opponents, and so on. Network methods are a natural extension of SOS (Strength of Schedule). The main innovation of the Oracle is the addition of a customizable node in the network that does a better job evaluating upsets. For complete details, the original method was published in the following article:
About me: My name is E. Cabral Balreira, and I am currently an Associate Professor of Mathematics at Trinity University. My main research are is in global dynamics and differential topology, but my passion for sports has led me to think a lot about Sports Analytics. For the past few years, I am teaching a Sports Seminar and consulting with different organizations about data analytics.
The Oracle method has also been used in mathematical biology. In 2016, we published the following article:
Jordan M. Bush, McKenzie M. Quinn, E. Cabral Balreira, and Michele A. Johnson, “How do lizards determine dominance? Applying ranking algorithms to animal social behaviour”, Animal Behaviour, Vol. 118, Pg 65-74, 2016.
If you would like to keep up with the Oracle Rankings and receive our weekly newsletter with updated Oracle rankings and predictions, enter your info below:
Here are some of our media appearances:
- KSAT (Local ABC) story about our Spurs Predictions
- Express News cover story about Spurs NBA title Prediction
- Can math be used to predict outcomes of sporting events?
- Math professors use program to predict NCAA tournament results