Premier League football somewhat clashes with the F.A.Cup semi-finals this weekend. It also has meant that some of our data is for games taking place later in the week or a couple of games haven’t got data collected for them.

Some things jump out from the data we do have – firstly that Chelsea are 13.7% likely to get an exact 2-0 win and 13.8% likely to get a 1-0 win.

They are overwhelming favourites against Brighton at 68% likely winners in our model and 60% with bookmakers.

For some reason our model hasn’t picked up the Fulham v Arsenal or Man Utd v Burnley on Sunday.

It has however collected the data for Leeds v Liverpool on Monday night.

Liverpool are 14.4% likely to get a 1-0 away win against Leeds as the most likely score in that match. Frankly there aren’t many matches where I’d expect Leeds in where there is only one goal. Bookmakers also favour Liverpool by 8% more than our model does. I wouldn’t write off Leeds.

Everton play Tottenham tonight and our model 1% between the two teams and just 0.02% on expected goals.

I think our model is underestimating West Ham against Newcastle in tomorrow lunchtime’s game. If there has been a surprise team this season, West Ham are definitely one of them. I can’t think of anyone who thought they’d be in the top four at the start of the season. Our model predicts a draw but bookmakers have West ham on 47.5% as likely winners, with Newcastle on only 25%. I think it’s probably somewhere between what we have and what bookmakers mean odds are showing.