Having trawled the web extensively for better information to select matches for my football trading, I came across accastats.com and their free Bet Finder tool. I should say up front that I have no connection with Accastats, or anyone competing with them, I just decided to assess their selections and thought I would share my thoughts.

I tracked the outcome of their selections for one calendar month. The results are on this spreadsheet:

http://s000.tinyupload.com/index.php...35543201178347

I concentrated on 2 markets; Over 2.5 match goals was easy enough to get from the tool, as it lets you select that specifically, but I also wanted to look at games most likely to have a goal in the first half. This isn't on the list of options, but it does let you look for games with a likelyhood of one team leading at half time, which would require at least one goal to be scored, so I used that. Mostly I wanted to see how much of an indicator their percentage ratings were, so I stuck to games rated over 60%, I would have preferred higher but there were very few games in these markets rated over 70%, so 60% gave me a decent size dataset.

Things that stood out to me:
  • Overall it gives more successful results that unsuccessful, so you would be right more times than wrong if you followed these selections. But I also noticed that the percentage ratings for both successful and unsuccessful results was about the same across the board, so it's hard to say whether their selection criteria makes a real difference or not.
  • By breaking it down by league it's possible to use some common sense to cut down the number of unsuccessful selections, for example a number of selections are from lower divisions which are traditionally low-scoring, and both of the markets I was looking at require high-scoring games.
  • The tool examines 26 leagues, and it returned selections from 22 leagues over the course of a month. I thought something was wrong at first, but some leagues are in their off-season during the period covered.


I have added a profit and loss calculator to the spreadsheet and found that to at least break even, you would need to back 0.5 goals in the first half at 1.33 or higher, and back over 2.5 goals at 1.65 or higher. This doesn't take into account that you would not get the opportunity to do this in every game, also you could of course trade out before making a total loss, so it is really a projection if using the selections for outright betting rather than trading.

In summary, I wouldn't use it as an outright selection tool for football trading. I think it has some potential for highlighting matches that you can then look into further yourself and decide if they are worth trading or not. I wouldn't work off the percentage rating as it appears to be more 'broken clock accuracy' than giving a real indication, but the selections do appear to throw up some games with teams that have a high likelyhood of scoring.

The raw data is in the spreadsheet if anyone wants to have a play with it, would love to hear your thoughts.