Community

The xG stat – Which forwards are being slept on?

 The xG Approach

Whenever I read FPL articles discussing player selection, the two most commonly discussed criteria are Form and Fixtures. In this article I will be looking into the Expected Goals Per 90 Minutes (xG90) stat. Perhaps, we would then all be less likely to fall into the trap of “chasing last week’s points” when choosing our transfers. Players overperforming their xG stat would mean that they have been lucky and are unlikely to sustain the same scoring rate in the future. This is, of course, with the exception of players who are clinical finishers – who are more likely to convert their chances into goals and thus need a smaller xG value to return actual goals.

What do the numbers say?

The current top FPL scorer amongst forwards (and also 2nd overall) is Sergio Aguero. Though he is the 2nd most expensive FPL player, his xG numbers proves a strong case in his favour. He sits at first place with a staggering 1.10 xG90, To put that into perspective, the second place in xG90 only has 0.75 (I’ve filtered out players who have played less than 360 minutes this season). Aguero is also joint 10th in Expected Assists per 90 Minutes (xA90) at 0.33, one of the only three forwards to feature in the top 10 xA90. It does look good for Aguero, but xG90 does of course mean nothing if he does not play.

Tammy Abraham has been a major talking point this week. Personally for me, he’s one of this season’s best value players. Priced only at 7.7, the main concerns for those tinkering with their frontlines would be whether this young and inexperienced striker can sustain his incredible scoring rate. He’s currently sitting as high as third in xG90 at 0.72.

Here’s an interesting one – Jamie Vardy ranks at 48th with 0.27. This is the lowest he’s ever had across his 6 seasons at Leicester – down from 0.63 last season. Granted, Leicester did already play against 5 of last season’s top 7 sides, which could explain his low xG90. Even more impressive is when we consider that he’s racked up five goals this season already. For me, this is a rare case where Leicester’s tough start to the season coupled with Vardy’s world class finishing ability would throw that number of 0.27 right out the window – he could be a good differential pick with just 10% ownership.

A shocking blank for Pierre-Emerick Aubameyang last week did not deter 110,000 players from transferring him in. Similarly to Vardy, his xG90 is at its lowest since he joined Arsenal, coming in at 11th with 0.54. He has outperformed his xG, with chances this season being few and far between for him. Personally I would imagine that it’s hard for him to sustain his current goal ratio (7 in 8 games), especially when we factor in the imminent return of Alexandre Lacazette.

Roberto Firmino is someone I’m considering to have due to his price for a way into Liverpool’s attack. He’s not doing too badly either – his current xG90 of 0.49 ranks 15th, and that’s combined with his 0.41 xA90 which ranks 5th. Both numbers are his highest recorded since he came to Anfield.

The most transferred out this gameweek has been the disappointing Teemu Pukki. It’s hard to argue against it, too – he sits lowly at 20th with just 0.42.  Looking at the similar price bracket, Sebastien Haller owners would be tempted to move him on despite returns in each of his last two. I was surprised to see him ranked as high as 6th in xG90, with 0.61. Perhaps Haller has gone under the radar for us – he’s been getting many chances at an exciting West Ham attack. A differential captain pick for this weekend, Callum Wilson is 10th with 0.57. Teammate Josh King offers a cheaper alternative, but his new role on the left flank is not ideal. He comes in at 58th with just 0.18. I would steer clear of him when picking a cheap forward for what is currently very valuable forward attacking slots. Wesley is currently 7th with 0.59, which could suggest that he could continue to score at his current rate. With United’s turn in fixtures, Marcus Rashford could prove to be a differential pick as he’s not far behind at 8th with 0.57. 

11 Comments Post a Comment
  1. Rotation's Alter Ego
    • Fantasy Football Scout Member
    • Has Moderation Rights
    • 12 Years
    4 years, 5 months ago

    Thank you Some Witty Pun! Does make me worry about Auba, just not getting the service at all.

    Great look here at some interesting xG data - worth reminding people that if they want to look at this data everyweek, plus a few hundred more statistical goodies that Opta have to offer, they should consider picking up membership here on Scout!

  2. abaalan
    • 7 Years
    4 years, 5 months ago

    So what does it all mean?! Get in Aguero basically?

    1. SammoUK111
      • 11 Years
      4 years, 5 months ago

      Nope, get in Chris Wood

    2. Baps hunter
      • Fantasy Football Scout Member
      • 6 Years
      4 years, 5 months ago

      Avoid Josh King! The rest is for debate.

  3. Sharkytect
    • Fantasy Football Scout Member
    • 9 Years
    4 years, 5 months ago

    So you think vardy could be a good pick, (xG90 0.27, costs 9.0) but find it hard to argue against transferring out pukki (xG90 0.42, costs 7.1).

    Completely contradicts the opening statement about xG90.

    1. Luskerud
      • 5 Years
      4 years, 5 months ago

      Vardy doesn't care about xG and hasn't done so for alot of PL seasons. Pukki has played 8 games in the PL. I think thats the point..

    2. fusen
      • 12 Years
      4 years, 5 months ago

      Which bit of "Vardy’s world class finishing ability would throw that number of 0.27 right out the window" didn't you read though?

      He's saying the xG stat doesn't really apply here yet as Vardy is a statistical anomaly.

      1. Sharkytect
        • Fantasy Football Scout Member
        • 9 Years
        4 years, 5 months ago

        I read that. I also read "This is the lowest he’s ever had across his 6 seasons at Leicester – down from 0.63 last season".

        The article starts by arguing a good case for reviewing xG90 rather than form or fixtures...then we reach Vardy, and we decide to throw those stats out the window and start assessing the teams he has faced already to justify his low rating. I think a lot of us are eyeing up Leicester's fixtures and making arguments to get their players. In Vardy's case, that appears to mean disregarding the stats. I notice the same logic was not applied to aubameyang, whose xG90 is better than Vardy's.

    3. Annie
      • Fantasy Football Scout Member
      • 11 Years
      4 years, 5 months ago

      Agreed, even if Vardy is a good finisher Pukki at 0.42 and 20th isn't far off Firmino at 0.49 which is 15th and "not doing too badly".
      Any stat starting with x I'm likely to ignore.

  4. Woy of the Wovers
    • 13 Years
    4 years, 5 months ago

    Who was 2nd? Or 4th.

    Seems odd to ignore such obvious high performers just because they are not popular picks

  5. tambourineman
    • 9 Years
    4 years, 5 months ago

    Bernardo Silva is 2nd and Sterling is 4th according to https://understat.com/league/EPL, which appears to be where the data has been taken from.