GW17 Predicted Scores [by SuperGrover]

Once again I have the pleasure of welcoming weekly contributor SuperGrover to the site for hsis weekly evaluation of last week's Premier League games and forecast of the games ahead.  @shots_on_target

We’ve introduced a new feature this week, the best and worst attacking performances of the week.  Below you will see the best and worst three attacks of the week when compared to the overall average of the league on the season.  For instance, Wigan’s performance against QPR generated 58.9% more offensive potential than the league’s average in 2012.   

Please note that these are measures of raw offensive output and are not impacted by opposition strength, game location, or team form.  I hope these will help shed some light on the underlying team ratings as well as provoke some discussion and hopefully constructive criticisms.

Performances of the Week (GW16)

Best Attacking Performance

Manchester City +73.0%

Wigan +58.9%

Fulham +54.3%

Worst Attacking Performance

Stoke  -100.0%

Reading (vs. Sot) -72.0%

West Brom -62.6%

That’s right; Stoke had the worst possible performance against Villa this week.  They generated no shots on target, no big chances, and only a single shot in the box.  Quite impressive in their futility, really.

Elsewhere, yours truly finally had an above average game week thanks to succumbing to the allure of Michu.  No one can deny his beauty as a fantasy asset this season, at least until the next big thing.

Here are the rankings for this week, followed by the predicted socres and some notes and commentary. 

 Gameweek 17 Predicted Scores

Changes and Notes this week:

Biggest Attack Gains

Manchester United + 4.10

Manchester City + 3.90

Sunderland + 3.30

That Manchester Derby was pretty fun, eh?  City dominated statistically but United’s counter attacks were clinical.  My model thought both were strong offensively but gave City the overwhelming advantage due to shot volume.  The question is should that be the case?  It’s possible the model is undervaluing counter attack strategies.  More work is needed here I think.

The Black Cats finally did something on the attack and have at least looked like a premier league calibre offense recently.  Will be interesting to see what happens this week against Man Utd.

Biggest Defense Gains

Aston Villa + 3.30

Tottenham + 2.30

West Ham United + 1.90

Villa have quietly been a very good defensive team this season save for a few blowout losses.  They absolutely dominated a pathetic Stoke attack that is starting to look like early season Sunderland.  At some point, you guys have to score to win, right?  Elsewhere, Spurs were solid at Everton and West Ham deserved a better fate against Liverpool who scored more than they should have for once.

Biggest Overall Gains

Sunderland + 3.30

Wigan + 3.30

West Ham United + 3.10

Sunderland are still struggling, but looked like they are going to put up a fight against relegation.  Again, this week’s game will be interesting to see if they can continue to attack without losing shape on defense.  Color me skeptical.  Wigan were certainly the better side against QPR (speaking of relegation…).  The Hammers outplayed Liverpool with Suarez suspended and should have at least drawn.

Biggest Attack Losses

Stoke City - 4.30

Everton - 3.10

Reading - 2.60

Stoke City generated two shots, one in the box and none on target.  Against Aston Villa.  That is not good.  Everton were pretty strong but more was expected from this elite unit.  It’s hard to contemplate how the Blues managed just four shots on target from thirteen in the box and twenty total.  Reading are pretty clearly the worst team in the league.

Biggest Defense Losses

Reading - 5.20

QPR - 5.00

Swansea City - 4.90

Did I mention that Reading aren't any good?  Yeah, they aren’t.  Neither are QPR who need to start turning shots into goals or will find themselves in the Championship.  Swansea has now alternated three very good weeks with two very poor weeks, making any investment in their backline risky although Davies represents a great value

Biggest Overall Losses

Reading - 7.80

Swansea City - 4.70

Stoke City - 4.00

Hey look, it’s Reading!  As mentioned, Swansea weren’t very good this week.  Consistency is needed there.  Stoke remains an elite defensive club with no attacking prowess whatsoever.  Maybe the return of Crouch might help?

Other Notes

• Slowly but surely, Norwich and Southampton continue to climb the rankings.  These teams are playing like mid-level clubs, not relegation candidates.  Norwich especially look deep and talented across the entire pitch.
• I am repeating the common refrain, but Carlos Tevez was phenomenal in just 40 minutes on Sunday.  How can Mancini continue to keep him on the bench?  He was clearly better than both Balotelli and Aguero, the latter of whom looked off all day.
• I feel Everton are on the verge of a run of clean sheets.  They have been strong defensively recently and have seen misfortune and a few errors lead to conversions the other way.  I predict they get four clean sheets in the next nine games.
• The death of Fulham’s attack has been greatly exaggerated.  They dominated Newcastle and deserved the three points.  It remains to be seen how valuable Berbatov will be without Ruiz, but that team can still score.

See you next week with some discussion of correlation and predictive modeling!  Fun