Not a great week for the prediction model, however in a crazy week where there were 9 goals at the Etihad, and Palace and Watford won at home against Chelsea and Arsenal, it didn’t stand much of a chance!
Aside from the above 3 games, and Huddersfield winning at Swansea (which was a result I flagged as an outlier) of the remaining 6 games the model predicted 1 correct score, with 2 further games being just one goal out.
Last week it had 0 correct scores, and 4 games that were a single goal out, but in a more predictable set of scores. I feel improvement is being made!
The model thinks that Watford will win at Chelsea, which, when you think about it, isn’t that crazy. They’ve just beaten Arsenal and are in good form, and Chelsea are a different team without Kante. Although Morata will be back.
I can’t see Burnley scoring at City so I think that’s wrong, and Spurs being predicted to score 2 is based more on Liverpool’s defensive record on the road.
Fantasy wise, the model does not see many clean sheets this week, with only West Ham, Utd and Newcastle in that group. Palace are going to suffer from their early season form for a while here, I feel, and as I said above I think you can add City to that list.
City are backed again to score goals (probably not enough, mind), but this is not expected to be a high scoring week if you back the stats.