I’ve analyzed the models last 68 soccer predictions with some interesting findings…
My inclination was that there must be a correlation between the predicted scorelines given and the actual games total goals
For the analysis I looked at repeated predicted scores given and the over/under 2.5 goals market in soccer
Here are the results:
As you can see from the table above the models accuracy on goal totals has been impressive
These accuracies increase if you were to take the over/under 1.5 goals market instead
It’s clear that games predicted 1-0 have a high probability of being low scoring, and games predicted 3-0, 3-1+ will be high scoring
This understanding should help your soccer selections moving forward when reading the tables
Next I’ll be doing the same analysis for NBA, in an effort to find the correlation between the models predicted scorelines given, and the likelihood of a over/under in a game
NoNonsense.
Super interesting man, and a huge reason I follow you...awesome analysis