We finally left 2020 behind and we’re now a month into 2021. Within this month we wanted to see just how well our AI had performed across the various football leagues.
We certainly had a mixed month across the different leagues but overall we did pretty well given the changing circumstances of each league on a weekly basis. This blog will cover the different leagues and offer potential explanations as to their performance over January.
As stated previously we had mixed results across the leagues. Over the month of January we were in fact profitable with our Over 2 predictions across ALL English leagues with an ROI of +4%. On the other hand, for every prediction made on these leagues we had an ROI of -5%. Interesting.
If we deep dive into the data we find that our most profitable Over 2 predictions could be found in League One with an ROI of +51%, whereas our least profitable Over 2 predictions could be found in League Two with an ROI of -46%. This shows just how different each league has been this season even within the same country. Interestingly, if we look at all the predictions our AI made for each league the best performing league was the Premier League with an ROI of 9%. Although the worst performing league overall was League One with an overall ROI of -16%.
What this shows us is there’s seemingly a trend that lower leagues have been more difficult to predict than higher leagues this season with many more underdogs winning. A potential reason for this may in fact be fitness levels between leagues. There is a possibility that lower league teams find their players have a lower level of fitness as a result of the COVID break, so the ability for better teams to consistently perform is much more difficult. Furthermore, the lower leagues were suspended early last season due to COVID, whereas the Premier League continued to play games and ultimately finished the season. This may have had an impact on both momentum and fitness for teams which we are now seeing reflected in the results.
Now we have reviewed the results of the English leagues it makes sense for us to now cover the 2 European leagues we provide predictions for which are the Dutch Eredivisie and the Spanish La Liga.
Starting with the Dutch Eredivisie, a league which smashed the predictions last season but have been lacklustre since. If we look at the Over 2 predictions, the ROI for these was -24%. Given how this league has performed so far this season this was no surprise to us. However, the data which did surprise us was the overall predictions made in the league. The ROI for January’s predictions was in fact +5%. This shows us that our AI has found it difficult to predict close games where both teams winning have odds of over 2, however overall it has managed to generally determine how good teams are and whether they will win, lose, or draw a match. This is very promising as it indicates that the AI is slowly learning the new nuances of the Dutch league and making the required adjustments.
Now we move onto the Spanish La Liga. This has been an interesting league to predict as it has been relatively inconsistent. This has perhaps been proven by the January results which shows that over every prediction made the ROI for these was -5%. However, the ROI for our Over 2 predictions in the league was in fact +8%. This shows that the AI in La Liga is opposite to the AI in the Dutch Eredivisie. This is shown by the AI being able to predict closely matched teams much better than the overall strength of a team throughout the season.
As you have just read the AI across the different leagues has been relatively inconsistent this season. This is not surprising given the way the game has changed, from VAR being introduced to COVID isolations.
Although, what can be taken from January’s results is that the AI is adapting to these changes, and as a result is slowly starting to perform better in the struggling leagues such as the Dutch Eredivisie.
We strongly believe that the AI will continue to adapt and get better as the season keeps progressing. Make sure to keep posted on our social medias and out blog post so you can see join us for this journey.
Akka = The name of our AI.
ROI (Return on Investment) = Profit as a percentage of total money invested.
Calculation: (Money Won – Total Money Staked)/ Total Money Staked
Over 2 = Any of our AI’s predictions where odds were greater than 2 and less than 3 (decimal odds).
Expected Accuracy = Converting bookmaker odds to an accuracy.
Calculation (based on decimal odds): 1 / Odds
Stats Zone = Part of our Football Package offering stats for 14+ European Leagues.
Backtest = Use the AI to predict past data showing the accuracy levels if used in the past.