by @TheSchachter

An introduction and explanation of the Simple Rating System can be found at the start of my GW17 article.

Lets start by going over GW19 predictions and results:

7/10 correct predictions. Great week for the model! As I thought, the model came in handy for a GW like this. Although the Arsenal – Utd game is a miss, the result is extremely close to the prediction, and the % spread shows how open the game was. Incredibly, the model was spot on in 5 matches – with BLA, CHA, CHE, LEE, BOL.
Lets look at the Blackburn – Liverpool match. A hard match to call I would say. The model predicted Blackburn as the 2nd highest scoring home team in this GW with 1.98, on the other hand we have Liverpool with an away score of 1.41 which is relatively high as well, margin is 0.6. This suggests to me a high scoring match with 1 margin Blackburn win most likely. Therefore, I had 2 BLA players on my web team (didn’t help as I captained Rebrov)
Another match worth a look at Upton Park. West ham with 0.18 at home while Bolton 0.8 (score that suggests at least 1 goal) away gives a margin of -0.6. As I mentioned before, any negative prediction already suggests quite a big gap in quality. Add the very low home score for WHU, and you get a nice and easy 0-2 away win.
2/3 correct recommendations. Spurs oh Spurs, the biggest troll for our SRS model, seems like the Sullivan deal has broken the dressing room. Spurs on a bad run, but at least they still give back attacking returns. None of the players I recommended returned here. But as we say, the games that the model are most sure about (highest predicted margins) also turn out to be his biggest mistakes sometimes.
Game of the season was spot on. Both teams to score, which is why I benched blanc (not that I got the CS elsewhere). Don’t sell RVN he will tick along regardless of the fixtures. Charlton also brought it home, although the game was way below my scoring expectations.

Now for GW20 ratings and predictions:
SRS Ratings:

Home & Away Ratings:

Bayes Rating:

GW20 Predictions:

For the first time since I started this, I’m ditching spurs in my recommendations.

This week we have a very rare 0 predicted margin in the Arsenal Middlesbrough game. This suggests a very open game and basically anything can happen. Both teams to score and 1-1 seems like a legitimate score. I personally am trying to avoid this game and will go only with Gordon involved.

Leeds are the obvious choice this week, and the model highlights with 2.2 margin and expected goals of 2.76 for Leeds. This is a lot. If you’re like me and you have Viduka but haven’t filled your 3rd Leeds spot yet, go for Keane. He is going to be heavily involved. I will have him and Viduka upfront.

Last recommendation, Chelsea are expected to have a new manager bounce, and what better opportunity than home to Everton for exploding. They have the 3rd highest predicted margin this week after Leeds and Liverpool and I can definitely see a 2-3 goals and a clean sheet for them in this match.