634x258 Community Submissions
15 December 2015 40 comments
tm245 tm245
Share:

After reading the Hot Topics on Effective Ownership and thinking about how incredibly well the top scorers have been doing this season. This is especially the case up top with forwards dominating the Fantasy points most weeks, and in the case of the likes of Jamie Vardy, at a very affordable price. In total half of the 30 top Fantasy Premier League point scorers for the season cost 7.0 or less, distorting the concept of value. I wondered what it would be like to change things up and follow one tenet: play the game as simply as possible and opt for the top scorers, with no club bias, no past experience and no underlying statistics to guide me.

I composed a brand new team on the FPL site (almost instantly deleted, never played it, in case the sentries at FPL towers are observing) with the budget of 100m and populated it with the season’s top performers. I soon found out that the top fifteen scorers cost just 99.9m.

The Dream Team

Butland 4.9 (Gomes 4.7)

Smalling 6.9 Alderweireld 5.7 Kolarov 6.5 (Dann 5.3 Sagna 5.3)

Mahrez (c) 6.9 Ozil 9.7 Barkley 7.4 Wijnaldum 6.9 (Ayew 7.1)

Vardy 7.6 Lukaku 9.1 Ighalo 5.9

So I set up the team under this basic premise of selecting by FPL points, with the captain being the overall leader.

Here’s how it went, with x donating a player that did not start, in this case Kolarov and Smalling, and with Riyad Mahrez as captain.

Gameweek 16

11  (7)

X 1 X (6 1)

26 6 2 2 (2)

7 9 7

78 points, Gameweek rank around 73,000

I am going to maintain this team (by hand, FPL, by hand) by changing it whenever someone breaks into the top 15 overall (two GK, five def, five mid, three fwd), even for a hit if I have to. For example, this week Nacho Monreal has leapt into fifth so I will need to switch him in for Sagna. I will also keep track of transfer value and selling value (five risers so far, though none of them doubled up) as if it really existed.

Conclusions

What does this squad, which one might call The Full Casual, prove? Well so far, nothing, but it sheds a bit of light on a few of the more bizarre components of this confounding FPL season.

The premium attackers have underperformed at a shocking rate:

None of the top 10 overall scorers started at a price more than 8.5.

Of the top 30 overall scorers, only eight of them currently cost 8.5 or greater.

Of the 30 most expensive players, all of them attackers, only 16 of them have scored more than the 30th best overall scoring defender.

Chelsea are at the top of the list for failure:

Their assets, so expensive yet such insecure sources of points, cannot possibly justify their price tags at this time, and are really challenging the maxim that form is temporary while class is permanent.

Eden Hazard, Diego Costa, Cesc Fabregas, John Terry, and Branislav Ivanovic all were among the highest scorers in PPG last season would have been counted on for a robust 5-6 PPG.  Costa is currently the highest of the big five at 3.3 PPG, with even Ramires outpacing Fabregas, Terry, and Ivanovic.

It seems almost impossible to get a very cheap player to accompany your expensive stars:

A popular preseason strategy for FPL managers would have been the approach of loading up on attackers from the big six clubs while trying to piece together a fairly inexpensive backline.

Instead only seven of the top 20 attackers come from the four Champions League qualifying squads this season, plus Spurs and Liverpool. Meanwhile, eight of the top 11 defenders come from that group.

Even more interestingly, perhaps, only five of the top 30 defenders cost less than 5.0.

Ignoring almost all of the deeper thinking that makes FFScout such a powerful contributing force to playing the game well, such as fixture analysis, underlying stats, and the price change mechanics of the market itself, it will be interesting to see how this team fares over the season, especially when the double Gameweeks come around.

40 Comments Login to Post a Comment
  1. J0E
    • Fantasy Football Scout Member
    • Has Moderation Rights
    • 16 Years
    9 years, 11 months ago

    Thanks for this. Look forward to the updates.

  2. Fergietime
    • 10 Years
    9 years, 11 months ago

    When is double game week? Is it boxing day?

    1. J0E
      • Fantasy Football Scout Member
      • Has Moderation Rights
      • 16 Years
      9 years, 11 months ago

      None pencilled in yet. The first tend to follow a blank gameweek for the Capital One Cup. Snow may create more blanks to be filled in as dgwks as well.

  3. Cowboy John
    • 13 Years
    9 years, 11 months ago

    Interesting, thanks.

  4. muddlingalong
    • 13 Years
    9 years, 11 months ago

    I wouldn't hold out too much hope.

    Applying a variation of this theory my calculations suggest you would have finished last season on 1837 points, and you'd currently be on 802, and that's before any transfer costs are taken off.

    These figures assume you always play a 343 formation and playing the top scoring players for each position (In week 1 you would select last seasons top players. I've not included any players who didn't play at all each week, assuming you would know in advance not to pick them.

    On top of the modest score, you'd have needed to use 123 transfers last season (-340 points), and 73 (-228) already this season.

    The past 2 weeks would have given you 2 of the top 5 weekly scores produced by this method since the start of last season. So I'd suggest that level of return is likely unsustainable.

    1. muddlingalong
      • 13 Years
      9 years, 11 months ago

      Transfer costs would actually be a lot lower than that, as this assumes the squad is only 11 players. Still, I don't think the points return is enough to justify the level of transfers you'd need.

    2. muddlingalong
      • 13 Years
      9 years, 11 months ago

      Also just realised I haven't included captaincy points. So feel free to ignore me.

    3. RedLightning
      • Fantasy Football Scout Member
      • Has Moderation Rights
      • 15 Years
      9 years, 11 months ago

      Last season, Billy Gilmore's computer program found two theoretical optimal dead team squads that would each have beaten the overall champion by 2 points.

      Optimal dead team squad A for 2014-15 (3-4-3): (GW1 paper value 100.0, would have finished 1st with 2472 points, 2 points ahead of Simon March)
      (G) Pantilimon(SUN,4.5, 123 in 27 GWs), Heaton(B) (BUR,4.5, 50 in 11 GWs)
      (D) Ivanovic(CHE,7.0, 179 in 37), Terry(CHE,6.5, 177 in 37), Clyne(SOU,5.0, 142 in 35), Yoshida(B2) (SOU,4.5, 57 in 12), Wisdom(B3) (WBA,4.0, 21 in 3)
      (M) Hazard(CHE,10.0, 233 in 37), Sanchez(ARS,10.5, 207 in 34), Silva(MCI,9.0, 191 in 32), Sigurdsson(SWA,6.0, 154 in 32), Walters(B1) (STK,5.5, 91 in 17)
      (F) Aguero(C) (MCI,12.0, 216*2 in 33), Kane(V) (TOT,5.0, 191 in 33 + 48 in 5), Austin(QPR,6.0, 176 in 33)

      In most recent seasons the optimal dead team squad would have beaten the overall champion by a rather larger margin.

      1. tm245
        • 14 Years
        9 years, 11 months ago

        Forgot about that experiment. Thanks for reminding as it provides a good method for focusing on trade offs between price and points.

  5. tm245
    • 14 Years
    9 years, 11 months ago

    Thanks, Jonty, for posting the article.

    As Muddling suggests above, this should fail. If it doesn't then I think the season has become even more unpredictable than it already has been, though I guess the entire exercise is a challenge to many of our instinctive responses of "surely this can't continue."

    1. RedLightning
      • Fantasy Football Scout Member
      • Has Moderation Rights
      • 15 Years
      9 years, 11 months ago

      The system would probably fail during the early game-weeks, but after a time it should become easier to field something closer to the dream squad for the season so far without taking too many hits.

      The method would probably get off to a poor start, and the hits might also take their toll.

      Also, are you intending to retain players who have built up big scores in the early game-weeks before suffering long-term injuries?

  6. DingoDongo
    • 11 Years
    9 years, 11 months ago

    How will team value work, the Monreal vs Sagna transfer takes you over the theoretical $100M on a full squad

    1. RedLightning
      • Fantasy Football Scout Member
      • Has Moderation Rights
      • 15 Years
      9 years, 11 months ago

      If this team had started earlier in the season then it should have increased its TV significantly by now.

      But the dream squad in the early game-weeks may not always have been affordable.

      And it may become unaffordable again if Aguero gets into it.

      1. tm245
        • 14 Years
        9 years, 11 months ago

        Exactly right about the lack of TV. This premise is asking how much the information of the first fifteen weeks was worth compared to the 4-5M you could have raised by playing the market.

        If Aguero supplants one of the three forwards the team will be in real trouble as it will have missed out on the hauls he had to crack the top three as well as suffering through the blanks that the cheaper forwards would have likely had.

  7. tm245
    • 14 Years
    9 years, 11 months ago

    Thanks for the questions. I always presumed that I would treat each Gameweek as if I were trying to do the exact same thing, field the highest scoring fifteen overall that I could for the money I have.

    In this case, I would look to drop the fewest points possible so am thinking I would need to do
    Sagna and Ayew out for Monreal and Cabaye in, -4.

    I hadn't really thought of long term injuries but I would assume that I would simply keep the players until they were bounced from the top fifteen.

    Would welcome any thoughts here as the method is certainly not set in stone.

    1. RedLightning
      • Fantasy Football Scout Member
      • Has Moderation Rights
      • 15 Years
      9 years, 11 months ago

      When looking to drop the fewest points possible, do you take into account that you are more likely to need points from your bench players if players in your starting eleven are known to be injured or suspended?

      This might mean that you sometimes ought to drop the player who has accumulated the most points but is now out for the long term, rather than another player who has fewer points.

      1. tm245
        • 14 Years
        9 years, 11 months ago

        Makes sense, otherwise it would just become a dead team. Might want to follow the logic of the most casual, reasonable player possible.

  8. the cromulent one
    • 12 Years
    9 years, 11 months ago

    Hey TM245. Sorry that your season is not going quite as you expected. Kudos for running the experiment. It should be interesting, and I'll be checking the results eagerly.
    It is truly bizarre that the premium attackers have under-performed, while the highest scoring defenders are mostly of the somewhat-expensive variety. That the highest-scoring 15 could be in one team is mind-boggling.
    In any case, it is good that the experiment is starting midway through the season. There would be too much churn in the top 15 in the early part of the year. But it also means that your team will be chasing last week's points by design. What the experiment will show, to a degree, is to what extent past performance is predictive of future performance. If you were buying a mutual fund, there would be a warning that this is a bad strategy (but I suppose that is exactly what casual investors do).
    The experiment may be clouded by long-term injuries, however. Would the full-casual look at the whole season, or the past x weeks when deciding who the best players are?
    Best of luck. I'm cheering for the experiment to do well.

  9. John t penguin
    • 11 Years
    9 years, 11 months ago

    good job TM
    your conclusion is reflected in my ML leader. Normally this guy floats around the bottom of our league, even below guys who have given up. This year he has a 60 point lead and has been between 5,000 and 20,000 overall most weeks.
    At the moment he has
    Cech, Monreal, sagna, kolarov, smalling
    is only defence transfer have been for similar players.
    and buying Kane 4 weeks ago has been his most expensive player.
    I can pretty much guarantee he will never buy aguero, sanchez or hazard.

    I think are still holding onto the theory ' when aguero starts perfroming' or ' when sanchez returns'
    but I don't even think one of them blitzing the rest of the season will make a big difference. At very best I can only see them gaining 50-60 points on those who don't have them.

  10. tm245
    • 14 Years
    9 years, 11 months ago

    Cromulent and JMcTD, thanks for the posts. Your points are well taken as I think this whole idea was prompted by me wondering how long an approach many of us would feel disdain for will succeed in the face of conventional wisdom, general predictive strategy, and past history, not to mention the resources of FFS, underlying stats, price predictors, etc.

    Red Lightning made some good points above as well, as I had forgotten about the Dead Team work they had done.

  11. Piggs Boson
    • 14 Years
    9 years, 11 months ago

    I've always been an advocate of the cheap backline, including the titans(Aguero, Sanchez, Hazard etc), and filling the rest with the mid-priced bandwagons (who NEVER fail to appear every season).

    However this season seems to be the complete opposite. There's so much value in attack, and those investing in the likes of Smalling, Toby and Kolarov must be laughing.

    It's been a difficult season for those playing the way I do, but I still remain stubborn that the likes of Aguero and Sanchez will hit big again soon and the likes of Vardy and Barkley will tail off. They just have so much more quality, and should get many more big chances provided to them by their teammates. I think Vardy and Mahrez will still remain bargain mid-priced players, but I just expect players like Aguero to vastly outperform them in the near future.

    1. Monkey Hanger
      • 12 Years
      9 years, 11 months ago

      Live by the sword..................

    2. tm245
      • 14 Years
      9 years, 11 months ago

      I agree with much of what you said and have made decisions accordingly that have not paid off.

      Aguero and Sanchez are exceptional players, and might not fit into this statement, but I fear we are chasing last year's points a bit when we move hell and earth for the Walcotts, Silvas, Hazards who are either not fir or in form. Hopefully this experiment will show us just how permanent class might be.

      1. Piggs Boson
        • 14 Years
        9 years, 11 months ago

        Yeah, we may be. But you can see they still have quality when you watch them. And the teams they play for are still quality, and will provide chances for them. It's hard to see how they won't get pts! 🙂

  12. tm245
    • 14 Years
    9 years, 11 months ago

    For posterity, not sure anyone is still following this thread:

    Butland (Gomes)
    Smalling Alderweireld Dann (Kolarov Sagna)
    Mahrez (c) Ozil Barkley Wijnaldum (Ayew)
    Vardy Lukaku Ighalo

    89 points, GW Rank ~150,000

    GW 16 - 78 points
    GW 17 - 89 points

  13. tm245
    • 14 Years
    9 years, 11 months ago

    FPL Dream Team Project Update

    (Figure this will be a good place to archive the updates)

    Good news is that it is showing signs of slowing down after a blazing start.

    GW16 - 78 points, GW rank around 73,000
    (no FT needed)
    GW17 - 89 points, GW rank around 150,000
    (1 FT Sagna to Nyom)
    GW18 - 43 points, GW rank ~1.8M
    (2 FTs Kolarov Wijnaldum Ayew to Wollscheid KDB Arnautovic, -4)
    GW19 - 63(-4) points, GW rank ?
    Gw16-19 total: 263 points (~66/week)

    Here is the current squad for this week before any gw20 transfers happen:
    Butland (Gomes)
    Alderweireld Smalling Dann (Nyom Wollscheid)
    Mahrez (c) Ozil Barkley De Bruyne (Arnautovic)
    Vardy Lukaku Ighalo

    A few takeaways:

    -- the Dream Team is slowly changing, though the core of Mahrez/Ozil/Barkley and Vardy/Lukaku/Ighalo have done fairly well over the last four weeks and seem quite secure. I would add that this should not be a complaint about the template but rather an opportunity for making up ground, since Vardy's slowdown and the Everton fixture turn could be a nice time to pounce on differentials. Why anyone would touch Ozil, Mahrez, or Ighalo is beyond me, though. The first two are dominating an underperforming position while the last one is the value player in the game right now.
    -- Defenders are the area of biggest turnover, not surprisingly since so many are so close to each other in the top 30. Maybe a West Brom defensive asset is a nice way to gain some ground, or a Chelsea?
    -- the Dream Team got lucky by having Arnautovic available as the next best player at a cheap price to make the KDB transfer work, as it shed the underperforming Ayew and Wij while picking up a midfielder on fire as a salary makeweight. He is now a starter on the season long Dream Team as the fourth highest midfielder.
    -- That is as much an indictment of that midfielder position as it is a celebration of Arnie, however. I think a key to shooting up the rankings could be leaving your high performing forwards alone for the most part and making peace with not being able to have them all while playing the hot midfielders on a run of form in the 3/4/5 spot after Mahrez and Ozil.

    I will post this on the original thread as well, but am always happy to hear feedback.

  14. tm245
    • 14 Years
    9 years, 11 months ago

    The tide might be turning...

    Gw20
    Butland (Gomes)
    Dann Alderweireld Smalling (Nyom Wollscheid)
    Mahrez (c) Ozil Barkley Arnautovic (De Bruyne)
    Lukaku (vc) Ighalo Vardy

    26 points, GW rank 3.379M

  15. tm245
    • 14 Years
    9 years, 10 months ago

    Gw21
    Butland (Gomes)
    Dann Alderweireld Sagna (Nyom Smalling)
    Mahrez (c) Ozil Barkley Arnautovic (Yaya)
    Lukaku (vc) Ighalo Vardy
    2
    2 9 2
    6 2 3 5
    2 2 2
    37 points, GW rank 1.344M

  16. tm245
    • 14 Years
    9 years, 10 months ago

    Gw22
    Butland (Gomes)
    Dann Alderweireld Sagna (Smalling Nyom)
    Mahrez (c) Ozil Barkley Wijnaldum (Arnautovic)
    Lukaku (vc) Ighalo Vardy

    7
    0 2 -- 6
    2 -- 2 13 3
    2 2 4
    43 points, GW rank 2.123M

  17. Ayew Serious
    • 10 Years
    9 years, 10 months ago

    Mahrez assist
    Ighalo goal
    Vardy goal

    Good week to start

  18. Eden Azzzar
    • 10 Years
    9 years, 10 months ago

    Not as bad as I thought at first
    Mahrez assist
    Kane goal
    Alli goal

  19. Eden Azzzar
    • 10 Years
    9 years, 10 months ago

    Alli of the bar

  20. tm245
    • 14 Years
    9 years, 10 months ago

    Gw23
    Butland (Gomes)
    Smalling Alderweireld Sagna (Dann Nyom)
    Mahrez (c) Ozil Barkley Wijnaldum (Arnautovic)
    Lukaku (vc) Ighalo Vardy

    45 pts, GW rank 1.621M

    (1 FT for gw24, Arnautovic to Ayew)

    1 FT remaining, 2.8m itb

  21. tm245
    • 14 Years
    9 years, 10 months ago

    1 FT, Gomes to Cech

    Gw25
    Cech (Butland)
    Alderweireld Sagna Smalling (Dann Nyom)
    Mahrez (c) Ozil Barkley Wijnaldum (Ayew)
    Lukaku (vc) Ighalo Vardy

    10
    8 x 1 (9)
    28 11 3 3
    9 2 2
    86 pts, GW rank 51k

    1.7m itb, 2 FT

  22. tm245
    • 14 Years
    9 years, 10 months ago

    Wrong order. Gw24 here.

    Gw24
    Butland (Gomes)
    Smalling Alderweireld Sagna (Dann Nyom)
    Mahrez (c) Ozil Barkley Wijnaldum (Ayew)
    Lukaku (vc) Ighalo Vardy

    1
    6 5 5
    12 3 16 2
    13 2 1
    66 pts, GW rank ~1M

  23. tm245
    • 14 Years
    9 years, 9 months ago

    1 FT, Nyom to Koscielny

    Gw26
    Cech (Butland)
    Alderweireld Dann Koscielny (Sagna Smalling)
    Mahrez (c) Ozil Barkley Wijnaldum (Ayew)
    Vardy (vc) Lukaku Ighalo

    1. tm245
      • 14 Years
      9 years, 9 months ago

      0.1m itb, 2 FTs

      Next week will be the test as there are four blankers in the team right now.