Use this thread for general daily football discussion.
This thread can also be used to discuss Transfer rumours and to post Tier 4 sources.
As this may fill up please sort by new to try and avoid constantly repeating the same question.
Join our Discord for live discussion and don’t forget to follow us on twitter.
I have read the full article by the way… I think it’s a very good idea but you need to run more numbers imo, and integrate more variables.
Taking the total of goals scored and total of goals conceded and then generate the expected points on average doesnt seem to predict much. According to that model, Chelsea in 04/05 should have had 74 points. But they didnt, they reached 95 points, breaking the record at the time. Best-ever attack is ManCity (but it’s 106, not 108 like in the article), they were supposed to reach 89 points according to the model, but they reached 100.
I think what that demonstrates is that there is more than just total of goals to consider if you want to know if attack or defense is more important to win a title.
Imo the hypothesis that needs to be tested is two-fold (and we would have to define the metrics to test it, but it has to go further than goals scored vs goals conceded):
- The first condition is that good teams need to play possession football and a certain degree of attacking football, in order to compete at the top of the table.
- Once the first condition is fulfilled, defense (in a broader sense than just goals conceded: solidity off the ball, pressing, control, etc.) is the difference to win titles.