You are viewing a single comment's thread from:
RE: Predicting The Outcome Of Soccer Matches With Data
For about 15 years me and my friends are playing a soccer prediction game. In one of these years I try to create a model to predict the score of the game. As predicting the score is more difficult than predicting the result, my model resulted so bad than I thought.
While I was creating the model I include these factors:
- League positions of the teams and, goals for and goals against(gf/ga) according to home/away
- Last 5 matches between two teams and gf/ga
- Last 5 matches of the teams in the league and gf/ga
Actually the model was not bad when predicting the result, as it gives correct results between 40% to 60% but when it comes to score prediction it is not what i expect.
Very interesting, @udgu. I think to consistently predict the correct score is impossible because there are so many possibilities.
Asian Handicap might be more accurate the odds are so low it is often not worth it unless the bookies get it wrong :)
As you said there are so many possibilities but I just want to try it for our game. I used that for two weeks actually and quit it because my own predictions are more accurate :)