Comprehend football?

Curiosity to find game’s hidden laws - our main motive to incept the beta5ai project, publish first prediction in early 2021 and improve every day.

Unlike existing algorithms that are mostly Poisson-based, we believe there is a significant amount of games that are governed by different laws. Our research focuses on discovering such games and so improving the outcome predictions.

We:

  • aggregate human experience and combine it with AI to predict outcome of football games

  • validate our results in direct competition with the betting market

  • calculate optimal wager for the bet, integrating known and our own mathematical achievements

  • publish new bets every day

  • show bet history, not hiding our losses

  • expanded over numerous football tournaments and several betting markets

Currently:

  • our main betting strategy multiplied initial fortune many times

  • top-20 tipster on world’s renowned tipster site is powered by our algorithms

  • we provide odds and betting analytics for 25,000 games, and more than 1,000,000 bets a year

Coverage
UEFA zone:

204 national and international tournaments overall

Pre-match:

Bets are made every morning

94 events in 7 betting markets:

1x2, Double Chance, Draw No Bet, Asian Handicap, Asian Total, Both Teams to Score and Correct Score

19 international bookmakers:
5 betting platforms:
Data

We collect and integrate football data from different data sources: game details, betting odds, tournaments situation and geography of the game.

Our database contains over 300,000 matches played since year 2011.

Features

The features contain aggregated market opinion about games, taken from betting odds.

Team power ratings are being constructed:

  • Glicko rating adapts well to the short term changes

  • Elo rating proves consistency in long term.

Team results are accounted using techniques of sliding window and exponential moving average.

Team positions and motivation are received from tournament situation.

These and other results are used as data features in our AI models.

Machine learning

The machine learning is performed via tuned XGBoost algorithm, maintaining few models for classification of score sum, score difference, both teams to score and correct score.

Our models are validated via bootstrapped walk-forward routine where test data is the last 365 days, while train data is all earlier.

Our models are trained every day, using results of previous game day - focusing rather on proper features than on exact weights and biases.

Calculated classification probabilities are adjusted then via Dirichlet-multinomial game score distribution, and for case of Asian half- and quarter- odds respective quasi-probabilities are calculated.

Market

While generally producing balanced opinion on game results, betting market is affected by money flow and preventing arbitrage. These and other factors, combined with stochastics of the whole process, give our algorithms sweet spots where probability of the event calculated by us is greater than what bookmakers state, even including their margin.

Comparing calculated and market probabilities, we receive optimal event to bet on.

Our historical results show that:

Predictions

Having probabilities calculated and odds collected for the game, we select optimal prediction - a tip where the bet could be placed at.

For optimality criterion we elaborated the simple yet efficient approach, minimizing inevitable temporary losses.

This algorithm is used when we publish predictions at our website or make a tip at 3rd party portals.

Risk management

To calculate an optimal wager for a bet, a mathematical technique aiming profit maximization is combined with another technique aiming risk restriction.

For profit maximization we use results of Kelly.

For risk restriction we use own results coming from research of Random Walk process.

Our benchmark tests show that even in perfect case where all probabilities are calculated without error, the pure Kelly method ruins in 20% of cases. Moreover risk grows significantly, as inevitable mistakes take part in probability calculations.

Applying our risk restriction technique just slightly reduces resulting profits, while increases robustness a lot - even in case of slightly optimistically calculated probabilities this approach leads only to 1% of ruin, and negative effect of increased mistakes is less comparing to pure Kelly case.

Bets

Having proper risk management, it is possible to select optimal wager for arbitrary betting odds received from any bookmaker, or your neighbor in the bar.

To make betting even safer, we add artificial skepticism to our bets via discounting calculated probabilities a tiny bit.

One more important practical thing about our bets is that we play on original fortune, without necessity to add new money to it.

Having plenty of betting odds, we build optimal betting strategies for bookmakers automatically, while also apply on manual betting via different platforms, thus providing wide range of opportunities for our users.

Assistant

Betting opportunities are not restricted with supported bookmakers and platforms - quite contrary, it is possible to get and optimal wager, just knowing betting balance and odds of the event.

Consider not betting, if we say 0.

Consider also number of bets: rough calculations recommend making not more than 10 bets a day, to stay relatively safe.

Keep in mind, there is no way to eliminate the risk of losing; at least mathematics says so - but our mathematics is to reduce it to acceptable margin.

Transparency

We are absolutely transparent regarding history of our predictions and bets - giving you a true sight on where we win and where we lose.

Gambling is evil. Betting is risky. Remember that.

We do our best to help you to become richer, or at least to have fun as a minimal requirement.

So, before using our results, please learn carefully what we do, assess if it matches your routine, do some paper betting and take any other measures protecting your good mood now and then.

We are happy to help, so reach out for feedback.

And stay with us on a bumpy ride of football comprehension!