League predictability feature
It is a set of metrics that measures the quality and predictive power of the model per league.
- hit ratio: number of times the model predicts the right outcome (1, X, 2) over the last 100 matches of the league. The closer to 1,00, the better. The random prediction has a hit ratio of 0.33 (33%).
- log loss: average log loss over the last 100 matches of the league. This measures the probability quality—the closer to 0, the better. We consider the quality good when it is below -1.02 and high when it is below -0.98. The random prediction has a log loss of -1.09. So, we consider any log loss below -1.07 as poor predictability.
- predictability: not everybody is comfortable with numbers. The predictability tells you in a word if the league has a poor, medium, good or high predictability.
- predictive power: this tells you if the log loss has increased in the last 50 matches. It can be up, down or unchanged. When the predictive power is up, the league becomes more predictable.
We understand the importance of trust when making informed betting decisions. With our transparent approach, you can be confident in our predictions’ accuracy and easily make informed decisions. Want to know how to measure the quality of football predictions? Check out our Medium Article.
Speaking about making informed decisions, our Value Bet API is exactly what you want.
Value bet API
The Value Bet model processes thousands of historical odds data and market trends to find the best value opportunities. In other words: it compares bookmakers’ odds with each other and then gives you the best value bookmaker.
“Our value bet models use bookmakers’ odds to find the best betting option to beat the bookies.”
Using our value bet model, you can access valuable insights into the best bets, which can help you make better decisions. We continuously improve and test our model by analysing past odds and value bets to ensure it works correctly.
How does it work?
The detection algorithm runs every 10 minutes up to the beginning of the match once the odds are available. Each value detected by the model comes with a set of two features:
The stake feature
The stake helps manage the risk the model would take in the bet. The risk is measured with the volatility of the profit and loss of the value bet strategy. The stake is calculated to have an average risk of one unit (one euro, one dollar, etc.).
Example: When you bet on a 20% chance event with fair odds of 5 and a stake of 3 units, you’ll profit 12 units 20% of the time but also experience a loss of -3 units 20% of the time. This results in volatility or risk of 6 units.
The Value Bet API provides information on each value found in the stake, ensuring consistent volatility of 1 unit when the match is played multiple times. This means that every value detected in the API carries the same level of risk. The maximum stake allowed is 5 units; maintaining the same scale for consistency is essential.
Are you interested in additional info? Make sure to check the Medium Article: How to Compute Football Implied Probabilities From Bookmakers Odds
The fair odd feature
Our algorithm helps us determine the fair odds for a value. The fair odds are helpful when betting against bookmakers not listed in Sportmonks. Any odds higher than the fair odds can be seen as a value. In the previous example, any odds offered above 5 should be taken.
We only use machine learning for the Predictions and Value Bet API to ensure the most accurate and reliable predictions and value bets.
As mentioned before, we value transparency. That’s why checking out the Medium article about How To Backtest a Value Bet Strategy is a must-read!