Fantasy Football

Fantasy Premier League: Updated Methodology and Gameweek 10 Rankings

Regular readers will know that I have developed (am continuing to develop) a methodology to identify the players with the most potential for points in the Fantasy Premier League (FPL) game. Before I jump into the rankings for Gameweek 10, I have done some work since my last update to strengthen the weaker areas of the process, and so the new methodology is as follows (for those interested; for those not you can skip ahead to the rankings):

  1. I correlated FPL points from 2014 to 2016 with key underlying metrics for each position, which through a simple linear regression analysis gave me equations which explain what should be happening each game from an FPL perspective. This method is a way of ranking players by their probability of performing through assessment of their on-pitch actions, or what I’m calling form, over the last four games.
  2. I then rank each club based on the forthcoming six fixtures. This is done by understanding the propensity of the opponents to concede big chances and the propensity of the midfielder or forward’s team to create big chances. For goalkeepers and defenders, the reverse is taken into account. Update: I now weight these rankings based on the strength of the previous four fixtures relative to the forthcoming six. For example, if the previous four opponents conceded an average of one big chance per game but the coming six have been conceding two per game, then the potential to create chances is doubled.
  3. For the goalkeepers and defenders, clean sheets play a significant role in assessing their form, which is a team action covered by the fixtures ranking. Therefore, in order to get a true view of a player’s form regardless of their team’s activity, I neutralise the clean sheets metric by giving all players a value of zero for this metric.
  4. The overall rank over the player is discovered by simply calculating the average between the form (individual) rank and the fixture (team) rank. Update: to reflect the importance of the team effort, I have up-weighted fixtures over form for goalkeepers (75%-25%) and defenders (60%-40%)
  5. Update: I have also replicated this methodology looking only at the next gameweek. This should give me a view on captaincy potential.

Here’s an example to illustrate my methodology. Let’s take Troy Deeney of Watford.

  • From gameweeks 6 to 9 his form suggested he should be scoring 0.0213 points for every minute he was on the pitch (not including appearances), which earns him a form rank of 15th. Whether he scored this amount is not of interest, what is important is that he should have done based on the equation for forwards.
  • Watford in that time created 2 ‘big chances’ according to Opta, or 0.5 per game. In that same period, Watford’s next six opponents from gameweek 10 onwards conceded an average of 2.0 big chances per game. The last four opponents’ conceded 1.438 chances per game, so Watford’s opponents are judged to be likely to concede 39% more chances per game in the next six that they did in the last four. Therefore, we calculate Watford’s attacking potential over the next six as 0.695 big chances per game (=0.5*139%), which earns him a fixture rank of 16th.
  • Taking the average of the form rank (15th) and fixture rank (16th) gives him an overall rank score of 15.5 (=(15+16)/2). 20 forwards earned a better rank score than this, so his overall rank amongst forwards is joint-21st (with Leicester’s Okazaki).
  • For the next gameweek (10), his form rank remains the same (15th), but Watford’s fixture rank (away to Swansea) is 10th. His overall rank score (=(16+10)/2) of 13 is bettered by 15 forwards, so he sits in 16th in the rankings for forwards for gameweek 10.

Please note that only players who have played more than the equivalent of more than one full game in the last four qualify for the rankings. Please also note that I have chosen Deeney for this example to highlight my annoyance at having a forward of low potential in my front line.

It should also be noted for anyone looking to use this analysis as the gospel that some players will over-perform against the model and others will under-perform. The research is designed to highlight probability, not provide a definitive answer. If you have any questions, I can be found on Twitter @artemidorus_1.

 

Goalkeepers

gks

Chelsea’s improved defensive performance (three successive clean sheets) has been helped by the concession of just 2 big chances in the last four, which has elevated Thibaut Courtois to the top of the rankings for the next six, although a trip to Southampton means he ranks only 10th for GW10.

For those looking to bring in a new ‘keeper for GW10 the numbers suggest that Heurelho Gomes of Watford is a strong candidate. Watford’s defensive potential over the next six is second only to Chelsea (a tricky tie at Liverpool the only major fly in the ointment), and for those seeking immediate rewards a GW10 home tie against a Hull side with no big chances in the last four matches pushes him to the top of the one-week ranking.

Elsewhere, another budget option is Victor Valdes of Middlesbrough, who has conceded just two big chances in the last four and shut out freewheeling Arsenal in GW9. Three of the next six opponents a no great shakes going forward, although the team’s defensive resolve will surely be tested by Chelsea, Manchester City and Southampton.

 

Defenders (top 20)

def

The aforementioned defensive record of Chelsea has elevated Marcus Alonso and David Luiz to the top of the rankings, with Alonso particularly looking promising at left wing-back. Following these two are the left-back of Tottenham and the right-back of Southampton, which have been shared in recent weeks due to injury and rotation, but is a testament to the structure of both teams that the players are interchangeable and equally as effective. Interrupting them is the new FPL darling Jose Holebas of Watford, fresh from his 15 point haul in GW8. Watford’s decent fixtures, particularly in GW10 also elevate Younes Kaboul up the rankings.

Elsewhere in the top 20 there are plenty of usual suspects, some surprising names (Paddy McNair, Michael Keane, Juan Zuniga), and a few omissions which we would perhaps expect to be higher, notably Tottenham’s Kyle Walker and Toby Alderweireld, Arsenal’s Laurent Koscielney and Everton’s Seamus Coleman, although the latter is handicapped mostly by the fixtures (remember, fixtures play a more prominent role in these rankings than individual form).

For the record, bottom of the pile in 88th place is Swansea’s Jordi Amat, an early season favourite but one now set to clog up the bench of ~14% of FPL managers (including mine)

 

Midfielders (top 20)

mid

Southampton’s 2.5 big chances per game and a personal record of seven shots on target in the last four has propelled Dusan Tadic to the top of the rankings. A FPL asset that has repeatedly flattered to deceive in the face of impressive underlying statistics, the midfielder’s place at the top of the rankings comes with an health warning, but with a visit to Hull approaching after a GW10 match against Chelsea he is one to watch for the 97.4% of managers without him in their team.

Most of the rest of the list features a lot of expected names that have been written about extensively elsewhere, and a quite a few big names missing (such is the diversity of talent available in FPL this season). Of the other names, the Manchester United representatives Juan Mata and Jesse Lingard are both struggling to hold down starting positions and so should be viewed with caution, as should Leroy Sane at Manchester City.

A name worth pointing out is Gylfi Sigurdsson of Swansea. The Icelander has strong underlying statistics (six shots on target in the last four) and is part of an attack that is creating a league-high three big chances per game over the same period. The fabled ‘new manager bounce’ has yet to materialize in South Wales, but following two goals at Arsenal in GW9 and a backs-to-the-wall effort from Watford in GW10 it seems that the Sigurdsson is poised to capitalise in the coming weeks. That said, the returning Fernando Llorente may impact Sigurdsson’s position on the pitch, so that’s something to keep an eye on.

 

Forwards (top 20)

fwd

The game’s two most expensive assets, Manchester City’s Sergio Aguero and Manchester United’s Zlatan Ibrahimovic sit top of the rankings for both the next gameweek and the coming six, despite both underperforming in terms of goals and FPL points. Neither player can be matched for shots on target (eight in the last four) amongst the forwards, whilst both clubs are embarking on an attractive run of fixtures. It shouldn’t be a surprise to see either player explode into life in the coming weeks; the question becomes whether they can regain their confidence before FPL managers lose theirs.

Elsewhere, cameo appearances from budget forwards are perhaps giving a false positive about their FPL potential, although the returning Fernando Llorente is perhaps the most interesting of these options (for the reason’s already covered in the Gylfi Sigurdsson analysis earlier). Undoubtedly the best budget option at the moment is Southampton’s Charlie Austin. Although only a single place ahead of Crystal Palace’s Christian Benteke in the rankings, the ‘rank score’ difference between the two is large, putting Austin up there with Aguero and Ibrahimovic as the best regular starters around at the moment.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s