ALFA DATA BLOG




A look at a full season of value betting


In this article I will look back at the first season spent on value betting on matches from the top 5 European Leagues. I will show the results and some numbers around them. I will show which were the most profitable bets and try to understand the factors behind the bets I won.

Let's look at the amount of money bet first. My initial bankroll was 200£ and I set the initial amount on each bet at 5£. This is 2.5% of the bankroll and I tried to keep this rato stable over the season.
I placed a total of 190 bets over these 9 months. 85 of these were placed on a Draw and 105 on a home or away win. Notice that 45% of the bets were placed on a Draw.

The total amount won placing these bets is 194.38 £. I won 64.02 £ betting on draws and 130.36 £ on an home/away win. It looks like bets on away/home wins were more profitable in the end.

Over these 9 months the total amount stacked on bets amounts to 1443 £. The total ROI calculated as the ratio between total amount won and amount stacked is around 13%. This is a very good ROI and it allowed me to almost double my initial bankroll. Bets on draws gave an ROI just below 10% while bets on wins gave and ROI above 16%. A big difference that confirms bets on wins to be more profitable overall.

Out of the 190 bets placed the most profitable (as amount won) was the bet on Leipzig-Hamburg, at 4.54 odds. This ended in a draw and gave me a profit of 31.86 £.

Of course not all the months were profitable, below I show the performance of this strategy over the month of October 2017 to May 2018.

Month Number of Bets Total Stake Won ROI
October 26 130 32.70 25%
November 26 182 -11.69 -6.4%
December 12 81 81.09 100%
January 32 267 76.32 28%
February 12 108 -59.76 -55%
March 4 32 -0.32 -1%
April 49 394 42.60 10%
May 26 234 5.94 2%
- - - - -
Total 187 1428 166.88 11.7%
Monthly Average 23.3 178 20.8 32.5%


Over the course of 8 months I had a few negative ones. In February I lost more than 55 £, for a record negative ROI of -55%. Months with negative ROI were only 2 over these 8 months. I had 5 profitable months and one month (March) with very small loss.

The average ROI has been 32% with a maximum value of 100% in December. There is a high variance in the ROI but the model definitively has some predictive power and it yields profits over a long enough betting period.

Bankroll management and being able to identify bets with value are very important factors when designing a betting strategy. The more time passes the more I see betting as an investment opportunity. It's risky but not more than investing in stocks. It's important to be persistent and not give up when things are going bad. But it's important to be active and adjust the betting strategy to your bankroll.

I have taken a break this Summer but I will start again once the big leagues are back. In 2017/2018 season the model focused exclusivey on the top 5 leagues in Europe. For the next season I plan to apply the model on more leagues, which will increase the number of bets per month at least a factor 2. If I can manage to keep the same ROI while at the same time increase the bet volume of a factor 2 this will double bet earnings.

According to my research done on past performance there should be a some minimal difference between profitability between the leagues. This can reflect in bets on some leagues bringing more profits than others. Also, the difference may have the effect on changing the error associated to the model predictions. Results on some of these bets will then be more reliable and in principle it could be better to focus most bets on these reliable/high profitability leagues. For all these reason, in the second part of this blog post I plan to present an analysis on the model performance league by league. Stay tuned.

If you want to get more stats from Alfa Data, receive picks straight in your inbox and more register here.