634x258 Community Submissions
23 December 2014 39 comments
The3rdTurd The3rdTurd
Share:

It seems to be one of those generally accepted rules of thumb of the game that when it comes to picking your captain, you should go for a ‘heavy hitter’, a player that comes with a hefty price tag attached.

A quick look at the Fantasy Football Scout captain poll in recent weeks seems to back this up. Since Gameweek 12, only once has a player costing less than 9.0 picked up 10% or more of the votes in the captain’s poll (Charlie Austin in Gameweek 17, and even then he was a distant third behind Yaya Toure and Robin van Persie in the voting).

There is an implicit assumption that there is a strong correlation between a player’s price tag and their point-scoring potential, which may be leading us to overlook contenders for the captaincy at lower prices,  such as Stewart Downing and Austin.

In the FFS captain polls since Gameweek 12, Downing has averaged just 1% of the votes and Austin 6%. Neither of them were even listed as regular options in the captain polls until Gameweek 13.

Cheap v Heavy Hitter

With nearly half the season gone, I thought it would be instructive to see how strong the correlation actually is between a player’s price tag and their point scoring potential.

This is what we get when we plot player prices against Fantasy Premier League (FPL) points scored:

fplptcompare

 

And this is what we get when we plot player prices against points per game (ppg)

ppg comapre

 

Whilst there clearly is a correlation between points and player price (for those who care about such things, the r-squared is 0.3 for total points and 0.35 for PPG), it is not as strong as one might think, and there is a pool of players at the mid-price range who are outscoring several of the higher-priced players.

Downing, Austin, and even Christian Eriksen and Graziano Pelle have more than held their own against higher cost rivals such as Yaya Toure , Robin van Persie, Diego Costa and Wayne Rooney, and are not far short of the players at the top of the pile – Sergio Aguero, Alexis Sanchez and Eden Hazard.

What about a player’s explosiveness? Surely we look to captain the heavy hitters because they have more potential for racking up the explosive scores?

Double-Digit Scores

Let’s take a look at the number of times a player has scored 10 points or more this season:

doubledigit

Once again, Downing and Austin in the mid-price range compare well against their higher priced rivals. With three double-digit scores, each them match Van Persie, Aguero, Angel Di Maria and Cesc Fabregas and have done better than Costa and Rooney. They are also only one short of the players at the top of the pile who have four double-digit scores apiece (Toure, Sanchez, Hazard). The mid-priced duo of Leighton Baines and Christian Eriksen, it should be noted, are also at the top of the pile with four double-digit scores.

Blanks

What about reliability, then? Don’t the heavy hitters have a lower risk of blanking? Let’s take a look at the number of times players have blanked (I’ve defined this as scoring two points or fewer):

blanks

If avoiding blanks is a primary concern, it would seem we should avoid captaining van Persie, who has racked up 10 blanks in his 16 appearances this season. The rest of the heavy hitters do appear to suffer from fewer blanks than their lower priced rivals, but with five blanks this season, Downing can match Aguero, Sanchez and Hazard, while Austin is only one behind on six blanks, which is equal with Rooney and one better than Toure.

Conclusion

Ignore Austin and Downing as captain contenders at your peril!

Notes

–          Stats include all matches up to and including Gameweek 17

–          Screenshots taken from http://tinyurl.com/2014-15FPLstats-value

39 Comments Login to Post a Comment
  1. J0E
    • Fantasy Football Scout Member
    • Has Moderation Rights
    • 16 Years
    10 years, 12 months ago

    Fantastic article...backs up what many have been saying for some time that Costa as an armband option cannot compete with the likes of Austin.

  2. walkman666
    • Fantasy Football Scout Member
    • 11 Years
    10 years, 12 months ago

    Very cool stat analysis. Thank you. Insightful with clear implications for team composition and captain choices

  3. mambo
    • 11 Years
    10 years, 12 months ago

    Nice work, well done!

  4. Cars and Mané
    • Fantasy Football Scout Member
    • 14 Years
    10 years, 12 months ago

    Love stuff like this, great work. Could you give the r squareds for all the graphs please?

    1. Cars and Mané
      • Fantasy Football Scout Member
      • 14 Years
      10 years, 12 months ago

      Also, is that tableausoftware link yours? Amazing work if so.

      1. The3rdTurd
        • 14 Years
        10 years, 12 months ago

        Yes it is, thanks!

        R-squared values:

        For total points - 0.31
        PPG - 0.36
        Double digit scores - 0.33
        Blanks - 0.05

        1. Cars and Mané
          • Fantasy Football Scout Member
          • 14 Years
          10 years, 12 months ago

          Cheers. Much lower than I expected. Though I suppose if they were higher then the people at fpl towers would be some kind of weird psychics.

  5. Kalix
    • 14 Years
    10 years, 12 months ago

    Good analysis!

  6. Kalix
    • 14 Years
    10 years, 12 months ago

    Is Costa missing from the charts?

  7. A-VB
    • 11 Years
    10 years, 12 months ago

    Nice work.

    I can't seem to find to cost in any of the charts, which suggests that your analysis is definitely spot on. 🙂

  8. Diggers
    • 11 Years
    10 years, 12 months ago

    This is really top stuff. Thanks for sharing.

  9. The3rdTurd
    • 14 Years
    10 years, 12 months ago

    Dammit! I inadvertently had a filter on the graphs for FPL ownership, which means Costa is missing from all the screenshots in the article.

    I've just uploaded some new screenshots with Costa now included:

    http://imgur.com/NVzAaV6,BERGjSn,TNg4t2X,hN4LFST

    or have a look at the Tableau viz where you can apply your own filters (last link in the article).

    I don't think Costa's omission changes much - Downing and Austin compare more than favourably at the moment.

    1. SC not pearls before swine
      • 15 Years
      10 years, 12 months ago

      Haha! Thanks I was squinting all over the place looking for Costa. Great article. Is there any way to look at previous seasons' data? Because I do wonder if unbalanced fixture lists can skew the data a bit (Austin has had easy home games so far this season, for example)

      1. The3rdTurd
        • 14 Years
        10 years, 12 months ago

        Yes there is, I've got the same charts for the 2013/14 season:

        http://tinyurl.com/2013-14FPLstats-Value

        I also have the raw data for the 2 seasons prior to that, but I've not crunched any numbers on them.

        1. SC not pearls before swine
          • 15 Years
          10 years, 12 months ago

          Thanks!

  10. Ludo
    • 12 Years
    10 years, 12 months ago

    This is very good work, thanks!

  11. bibi
    • 11 Years
    10 years, 12 months ago

    This is the best article posted not by Jonty. 🙂
    really really nice work mate

  12. Pie-o-My
    • 12 Years
    10 years, 12 months ago

    Great article, really shines the light on Costa. After this GW he's gone from my team in favour of a more reliable Haz Cesc partnership.

  13. funny money
    • 16 Years
    10 years, 12 months ago

    this is superb, as is the link at the bottom. Love seeing 'gut feels' substantiated by analysis, but less keen on being proven wrong 🙂

  14. Robben Mee Blind
    • Fantasy Football Scout Member
    • 11 Years
    10 years, 12 months ago

    Great article mate, cheers. Very tempted to follow up the Austin captaincy by going for Bony instead of Sanchez on Boxing Day!

    1. Mr Byrnes
      • 11 Years
      10 years, 12 months ago

      I'm in the exact same boat, I would say Bony is a safer bet to start and has been on the brink of an explosive score for a few gw's now

  15. Bad Kompany
    • 13 Years
    10 years, 12 months ago

    Excellent article, makes me more comfortableabout choosing a lesser priced captain...cheers

  16. Mr Byrnes
    • 11 Years
    10 years, 12 months ago

    Great article, there are a lot of factors to consider when captaining a player but i believe price should not be one of them. In any given GW you should look for the player with the greatest potential for a double figure haul.

    If Podolski or Lampard were guaranteed game time for whatever reason i would consider them to have explosive scoring potential, but because of the low price they may not be deemed as 'heavy hitters' which means people may be more uncomfortable trusting them with the armband.

    I don't believe in taking punts on a captain to gain an advantage, however i fully agree with the article and being blinded to a great option because of price is misconception best avoided.

  17. d1sable
    • Has Moderation Rights
    10 years, 12 months ago

    Impressive work, kudos!

  18. Ginkapo FPL
    • 14 Years
    10 years, 12 months ago

    Can you knock it up to 3 points or fewer please? That CS point is knocking a lot of Midfielders out of the blank stakes.

    Great analysis.

    1. Ginkapo FPL
      • 14 Years
      10 years, 12 months ago

      Also what about, game to double digit haul ratio as a PPG version of that stat?

      1. The3rdTurd
        • 14 Years
        10 years, 12 months ago

        Not sure I completely understand but do you mean number of double digit scores as a percentage of total appearances, like this:

        http://i.imgur.com/LDRfJhf.jpg

        1. Ludo
          • 12 Years
          10 years, 12 months ago

          Now THAT is a very good table!

          1. John t penguin
            • 11 Years
            10 years, 12 months ago

            What is the bottom line?

            1. Ludo
              • 12 Years
              10 years, 12 months ago

              The literal bottom line of the table are players who haven't hit double figures this season.

              The more substantive bottom line of the table is that Austin/Siggy/Downing hit double figures as often as the likes of Aguero and RVP, it's just all about the timing. That's the point of the article, I suppose.

              Of course, Aguero and RVP play for top teams who can score several goals in a match, and so they've more likely to get a massive 15+ score at some point, while it's a bit of a limitation that Downing's 10s/11s get as much recognition as Aguero's 16s/18s.

              I also think it's interesting to see the likes of Baines, Eriksen and Yaya up there. Many see them as too unreliable as captaincy choices, as it's hard to predict when they'll score well, but clearly it can pay off, as they're as explosive as anyone else. That criticism has been levelled at Hazard in the past, though his home record this season has been consistent enough for him to dodge the label somewhat.

    2. The3rdTurd
      • 14 Years
      10 years, 12 months ago

      Just for you, my hirsute simian friend:

      http://i.imgur.com/7ddemuP.jpg

      It's also added to the Tableau viz.

      1. Ginkapo FPL
        • 14 Years
        10 years, 12 months ago

        Cheers for both.

        These are really useful. Its going to take me a while to intepret it all. My only instant thought is, God we need Di Maria back!

  19. Skokym
    • 12 Years
    10 years, 12 months ago

    Very nice work, thank you sir!

  20. GreenWindmill
    • Fantasy Football Scout Member
    • 14 Years
    10 years, 12 months ago

    Stunning.

  21. Live Lad
    • 10 Years
    10 years, 12 months ago

    Please suggest on my GW 18 team:

    Jones
    Apil, Baines, Bertrand
    Siggy, Eriksen, Hazard, Silva
    Bony(c), Costa (vc), Austin

    Subs: Green, downing, Taylor, Wisdom

    Transfers in mind for GW 19:
    Sterling, Benteke (in) Two successive favourable home games
    Eriksen, Bony (out)

    1. Dino
      • Fantasy Football Scout Member
      • 16 Years
      10 years, 11 months ago

      Excellent post, great contribution to the discussion

      1. beric
        • 13 Years
        10 years, 11 months ago

        Lol

  22. Kerz
    • 11 Years
    10 years, 12 months ago

    Great work. Fascinating read. Cheers.

  23. Footinmud
    • 12 Years
    10 years, 12 months ago

    Dude, this is really really good.