This weekend is lacking serious action in the major football leagues due to the upcoming World Cup qualifiers during the week (which are not exactly my cup of tea). Therefore I sat down and, after finalizing my tips for next weekend’s league games (which I needed to prepare in advance since I’m going on vacation, yey!), I decided to test some betting strategies. The idea was to see if a betting strategy that was successful in the previous season could be replicated with the same success this one. For example, a team was prone to overs the whole 2014/15 season and blindly backing o2.5 in every game with 1 piece of currency, you would have been up around 10 points in the end of the season, best odds taken. So if you repeat the betting strategy in season 2015/2016, what are your results expected to look like?
I started with teams prone to draws and didn’t find anything that did not seem too random. Remembering that rating systems produced best predictive models for home wins, I decided to next check a betting strategy backing home win for all teams that had positive returns in the previous season following the same betting strategy. I have made the test for all teams in the 6 major countries since odds have been made available in football-data.co.uk and I found the results intriguing. It looks like this:
|min last season profit||this season profit|
Strategy Profit per Country
The point is that in case the profit from the same betting strategy in last season has any influence over this one, one would expect as we pick only teams with higher and higher last season profit the results for this year should improve further and further. As you can see this was partly the case, especially for the lower last season profits. When the profits increase to 7-8 units per season or above results look a bit strange, but it should be taken into account that there in most countries the sample left is very small. This is also the reason why the volatility gets much higher compared to the lower profits. So I think those results might hint to something. I have subtracted the average profit for backing every team at home to discover the premium awarded for following this betting strategy for every country.
Strategy Premium per Country
For Italy, Spain and Germany the trend seems great, for England and Scotland not so much and France does not deliver at all. It was a bit strange to me that for virtually all countries it goes downhill for the really big last season profits, but I still believe it is due to the small sample as I don’t see any intuition behind that. For teams with last season profit >10 for most countries I got less than 10 instances across all years and tiers, so you get the picture.
I have not calculated the yield, but it is easy to derive it from the numbers above. There you have average profit per team per season. So, say, if the profit is 0.67 and a team in this league had 17 home games this season, all backed with level stakes, you get 0.67/17 = 3.94% yield.
I consider giving the betting strategy a try for Italy, Spain and Germany with a small bank this season. Will let you know how that goes.
If you have any experience or thoughts on that kind of strategies, feel free to share in the comments. And stay tuned for other interesting findings from the world of football stats and betting strategies by following the Church in Facebook or Twitter to get updated about the latest posts. See you around!