Community

Points Per Game Per Million – In Search of Value

Following up on my very quick ‘points per game per million’ research I posted recently, here is an updated and improved list for anyone who is interested. I am not a statistician so all of this must be taken with a pinch of salt, but I found it very interesting to put together and I have enough faith in the research for it to inform my decision to go for five premium defenders at the back, with an even balance to the rest of the squad.

Rationale

When we play Fantasy Premier League we are essentially trying to spend 100m as efficiently as possible, so the calculation in the backs of our minds is always point per game per million (ppg/m) even if we don’t think about it when we are setting up our teams.

The idea of this table is to work out where the pounds are best spent around the squad by identifying the best value players and their positions based on their average points per game divided by their price at the end of the season.

The players are categorised by price range, broadly budget, mid-price, or premium within each position, with further gradations between these categories.

Results

Analysis

It’s still a crude study but there are still interesting things to note:

  • The list is dominated by cheap-mid price keepers such as Tom Heaton, then premium defenders and mid-price defenders, then cheap/mid-price midfielders, then premium midfielders, mixing fairly equally with premium attackers towards the bottom.
  • There are no super-cheap defenders. There are very few budget picks in midfield or attack either. This is surprising and suggests that the best value is to be found in the middle reaches of each positional price range.
  • The first forward on the list is in position 20.  This is, perhaps surprisingly, Swansea’s Fernando Llorente
  • The first premium forward (10m+) is Harry Kane in position 32.
  • Even the biggest hitters from last season are fairly low down the list. Alexis Sanchez and Eden Hazard are below a raft of other midfield picks.
  • For every million spent, for comparative purposes, the best-performing cheap-mid keepers get you 0.87 points per game. By contrast, the premium forwards (including Vardy here) only get you 0.53 points per game.
  • This tells me that packing a team full of premium attackers just isn’t worth it, unless (see below) you are going to captain them regularly. Even then, if you have more than one premium in a team then the others are largely going to waste in a given gameweek.

Conclusion

Subject to the below caveats, I would conclude that the best use of my pounds would be spent to upgrade players from the base level. The investment is most pronounced with defenders, particularly the attacking wingback, which really came to the fore last season. Then, decent mid-high price midfield is the next priority.

The sweet spot for midfielders seems to be the 8-9.5m mark (last season’s Sadio Mané, Dele Alli, and Christian Eriksen). Past the 10m mark, you end up getting worse and worse value from players in both the midfield and forward bracket.

Forwards are poor value in general and I will not be spending much money on my forwards except with one enormous caveat – captaincy. The super-premium forwards and midfielders are still the highest points scorers and at least one of these should be picked for captaincy.

Additional Notes

  • All data, including the pricing, is now based on last year’s finishing prices. This gives us a halfway house between expectation and reality as prices rose and fell slightly. Encouragingly, the rules have not changed since last season, and there have been no game-changing price tweaks. I recognise that some players fell in value towards the end of the season due to injury or falling out of favour. These players’ stats are going to be slightly boosted as a result, so bear this in mind.
  • I have included 20 players from midfield and scaled down from there to pick the best 10% in each position, based on points per game. Numbers are based on this year’s stats – squads haven’t changed that much in sum) There are 200 midfielders in the game, so it’s the top 10%. Likewise, there are 66 keepers so I have included 7 etc.
  • Players like Josh King are still classed as midfielders for this research
  • My categorisations as to budget-mid-premium are largely subjective, but I think they look about right
  • All players should have  played 2000 minutes or more. Sorry if the odd outlier slipped through.
  • I do not have statistics for points per game per million, so I had to choose best points per game and get the millions from that selection. There may be players who have slipped through the net who have a higher PPG/m, but I don’t think this has much of a bearing on the results.

Caveat #1 – Captaincy. You need a good captain every week, and you may well want to rotate two or even three based on form and fixtures. Kane’s 7.5ppg is incredible and, doubled to 15ppg, makes him worth the outlay.

Caveat #2 – this is based on points per game, not minutes, so cameo players are disadvantaged by the numbers. Unfortunately last season’s points per minute are not easily available.

Caveat #3 Playing devil’s advocate against myself, I recognise that the returns on premium strikers are more predicable than premium defenders. For example, who would have guessed that Chelsea’s Marcos Alonso would be so strong, or Kyle Walker in his sixth full season? Whereas we knew Harry Kane would perform well, and it’s worth paying a little more for this surety. So it’s all very well to say that Alonso is a better value pick, but is he the same pick as last year? Maybe not and, if not, how do we know where this year’s Alonso is going to come from?

Cavet #4 This just occurred to me as I was writing up. It’s all very well saving money by going for the best value, which clearly looks to be mid-price keepers and mid-premium defenders, but if we are saving money by doing this, and the top-level players are poor value picks, then what are we saving the money for? Based on my data I would advise making sure you should spend the money making sure you have invested moderately in each midfield and non-captain position, and free up the rest to get the best captain pick or two that you can.

96 Comments Post a Comment
  1. mearley14
    • Fantasy Football Scout Member
    • 10 Years
    6 years, 7 months ago

    Hi Red Arrows

    great piece.

    I was just wondering what you price brackets you used?

  2. hkcharlie
    • 10 Years
    6 years, 7 months ago

    Points per minute should be the stat, what have you lot been doing all summer? Get to work quick!

  3. trikster102
    • 11 Years
    6 years, 7 months ago

    Good read, and it gives me some confidence in my team which currently looks like this

    Foster (Elliot)
    Bertrand Alonso Kolasinac Milner (Hunemeier)
    Zaha KDB Willian Alli (Carroll)
    Firmino Lukaku (Quaner)

    I think I'l stick with the 442 formation because I'd rather have a good defender instead of an average mid / forward. I tried making a team with Lukaku and Kane but I just couldn't come up with anything I liked.