When does an AFL player peak? Peter Higginbotham

Filed in Game Strategy, Supercoach, Talking Strategy by on February 21, 2013 • views: 1562

Peter Higginbotham AFL Fantasy geniusPure mathematics is, in its way, the poetry of logical ideas. Einstein, A

As Supercoach and Dreamteam professionals we stand on the shoulders of giants. Archimedes, Isaac Newton, Gauss, Bernhard Riemann. The greatest mathematicians that the world have ever seen constructed the mathematical foundation for the game that we have all come to love.

And while their teachings have formed the foundation of my staggering rise to prominence in the Australian Fantasy Football industry the same can not be said of many up and coming fantasy coaches looking to form a name for themselves in this game. The time has come for me to pass on some mathematical truths that will change the way you select your Supercoach and/or AFL Fantasy team.

Free your mind of prejudice. Forget what you thought you knew. Free yourselves of Jock inspired “gut feel”, throw off the shackles that emotionally attach you to a certain way of thinking.

I want you to see this game as a majestic orchestra of kaleidoscopic numbers to which you hold the baton.

WHEN DOES A SUPERCOACH PLAYER “PEAK”?

Supercoach averages cross sample

Exhibit A: Supercoach average v player age

You will see in exhibit A above that I have collated the seasonal Supercoach average of 31 current AFL football gladiators with some surprising results. The Y axis represents a player’s Supercoach average. The X axis his age in years. A preliminary examination of these results reveal some truths that may surprise some;

A supercoach player is statistically likely to peak at 22 years of age

Yes our cherished community of fantasy football enthusiasts – sometimes it takes the elegant simplicity of numbers to challenge previously held misconceptions. Many of the un-initiated spruik the wares of prospects such as Patrick Dangerfield (22), Luke Shuey (22), Dayne Beams (23) and Steele Sidebottom (22) due to the fact that they “have not reached their peak”. I would like to propose that these young lions are statistically likely to have reached their peak already. 

A player is statistically most likely to enter a recession around the age of 26

Beware the mid career form slump. Another misconception is that a player is at the peak of their powers in their mid to late twenties. This is statistically untrue. In fact it seems that according to this sample that a player – after plateauing since the age of 22 – is likely to degenerate into recession around the age of 26-27 years of age. So it seems that – statistically speaking - Lance Franklin, Dane SwanBrandon Goddard – these gentlemen carry a reduced likelihood of an increased output in 2013.

In conclusion

There are infinite layers of knowledge hidden in the beautiful and predictable realm of mathematics my friends. To reach the zenith of Supercoach and/or AFL Fantasy performance you must apply sound mathematical logic when formulating a winning Supercoach team.

“Johnny come lately’s” such as The Mad Irishman may fumble upon elements of mathematical reason when arriving at a conclusion. I will let the community judge who holds the cards around this community with regard to mathematical prowess as I reveal more mathematical arsenal in the weeks to come.

“A man should look for what is, and not for what he thinks should be.”

Peter Higginbotham

Barton Tafe Applied Mathematics class of 1982

Click here for more from the desk of Peter Higginbotham

  • players used in sample: Adam Goodes, Jonathan Brown, Ted Richards, Chris Judd, Heath Scotland, Alan Didak, Corey Enright, Jimmy Bartel, Paul Chapman, Joel Selwood, Steve Johnson, Lance Franklin, Luke Hodge, Sam Mitchell, Brad Sewell, Brett Deledio, Chris Newman, Shane Tuck, Justin Koschitske, Lenny Hayes, Nick Riewoldt, Sam Fisher, Matthew Boyd, Robert Murphy, Beau Waters, Daniel Kerr, Dean Cox, Scott Selwood, Dustin Fletcher, Jobe Watson, Greame Johncock, Michael Rischitelli

Tags: , , , , ,

Comments (23)

Trackback URL | Comments RSS Feed

  1. Shimma says:

    What do the blue dots mean Higgo?

  2. Brick says:

    So Jack watts is a lock then …………………………………………………..?

  3. rodney says:

    Great post, however I believe this "mid career form slump" is more a mid career injury. Players at the age of 22 are far less likely (given no previous injurys) to substain an injury than older players. They are also playing harder and feel like they need to prove themselves more than the seasond veterans. So really, all it's saying is choose younger guns over older ones, the thing is however, the younger ones are harder to pick. It comes down to indivdual players on individual days. Judd will step up in a game with meaning near the end of the season, can you rely on someone younger to do that?

  4. Brick says:

    How about draft pick number correlation? I've got a "feeling" about Jack Watts…… What, still no takers?

  5. 31 players? Too small a sample size, this is very inconclusive.

    • Shimma says:

      there are only so many current players in that age group

      • Kev says:

        everyone currently in the AFL fits in that age scale shimma. I wonder what the data would look like if all players who played 16+ games in 2012 were plotted. Like Darkhawk1979 said, it's a bit inconclusive. Would like to see if a similar trend line occurs if this was done again in a few years.

        Top stuff Higgo, I was really surprised. I just don't think it's the gospel, sometimes that gut and years in the supercoach caper make you sit up and take notice.

        • @PeterHiggo says:

          Dear all,
          In fairness to Shimma and to help explain the small sample size, the data or players selected had to;

          A. Be a player who has been in the system for 7+ years
          B. Be a player with a high SC Average, ie. On people's radar.

          Of course there are more players who hit these two selection criteria but hey, when champion data employs me I will dedicate more of my time to it.

          Regardless, the data speaks and it has your brain thinking = mission accomplished.

          With thanks,
          Peter Higginbotham

        • Shimma says:

          yeah like Higgo said, I meant that the players listed had all been playing for a few seasons (7+)

    • OneTreeHill says:

      I agree, the small sample size means nothing, you can plot ~400 players who are in various stages of their careers along that chart quite effectively. There's also data around for those retired, so you can use that to plot the other end of the chart.

      Limiting yourself to high average players is very subjective and you're already affecting the end result of any stats. Caveat Emptor indeed

  6. jockreynolds says:

    Higgo – as I have said many times on the podcast and in person… this community is bloody lucky to have a main owning a brain like yours looking after the numbers mate.

    And Brick old mate.. players like your Jack Wattses and company look a bit more attractive with all this in mind

    • The K Bull! says:

      You guys keep mentioning Watts! I think he is a huge break out contender this year, and DPP, liking him more and more!
      Great work Higgo!

      • Brick says:

        I'm on board for Watts, differential and DPP. Worth a punt early and if you look at his stats – he's SC average is on the up since his debut season. He might be averaging 100 this season. But I'm hoping he can pump out the odd 120+ game – massive hope I admit. But while everyman and his dog has broughton you can either go the less travelled path or have the swing set. Winning??? We'll see.

  7. Spifflicator says:

    Fantastic work Higgo, one of your best ever!

    As a another measure I would like to see this peak analysis done based on the number seasons played.

    Most draftees will be in their 4th or 5th season by the time they turn 22. Players like M Priddis however, who turned 22 in his debut season, or J Giles whose debut season was at 24, are not going be represented under this sort of analysis, but are still very much Supercoach relevant.

    Stephenson, Terlich, Magner, Stevenson, Van Unen are all mature age recruits with varying degrees of senior experience at AFL level, I would be very interested to see what the statistics say about players in their first five seasons.

    What do you think Higgo?

  8. Terry says:

    Hi Guys,
    Thinking of running this midfield:
    Ablett, Pendles, Danger, Cotch, Priddis, Barlow, Viney, O'Meara, Crouch, Mitchell

    too heavy?

    • Shimma says:

      i have the same first 4, but not sure on Priddis and Barlow. You could try some cheaper players like Fyfe, Mundy, Jack Steven, or even Moloney and use the extra money on other lines.

  9. jimbotraralgon says:

    Hey Higgo.

    Love your work but just not that much of a fan on this one mate.

    As a man of mathematics, you must appreciate 31 players is a very small sample size (700 odd players on the list this year alone).

    It is an interesting set of stats but wont be my guiding light.

    I stand by the fact that footy is not played on a graph or in a spreadsheet.

    Also doesn't factor in those mature age recruits mentioned by Spiff.

    Just my opinion.

    • Brick says:

      A wise man once said "Statistics are like swimsuits, what they reveal is interesting,but what they conceal is vital".

      Still, I'm looking at top draft picks from 3-4 years ago for breakouts.

  10. TheBIGShot says:

    Higgo there is so much work put into this blog on age I cant be full of anything but praise. Great work , though it is begging to be done on a huge sample over a 10 yr period when someone has the odd PHD that needs completing.

  11. JohnnoB says:

    Hmmm… Strange comment, "averages mean nothing"… Just what they base SC prices on you clown.
    Thanks Higgo. It seems your blog is being trolled by other jealous, wanna be fantasy bloggers. Ignore or delete the tripe!

  12. Guest says:

    Not only is the sample size too small, it is hard to attribute relevance to this without knowing how the 31 players were chosen. I could randomly select another 31 players, and it just might show the peak occuring more at 24. Looking at your graph, I can identify 3 players: J Brown, D Fletcher, and C Judd as being amongst your 31 players. I could probably narrow down a few others, if I put a bit more time into it. My concern is, are your 31 players a representative mix of positions, development stages, seasons played etc. The 3 I have identified were certainly earmarked very very early in their careers as likely candidates to be good players. If the other 28 were as well, it is no wonder your graph peaks at 22.

  13. Hello, There’s no doubt that yohr blog could possibly be having web browser compatibility issues.
    Whenever I look at your web site in Safari,
    iit loks fine however, when opening in IE, it hass some overlapping issues.
    I just wanted to provide you with a quick heads up!
    Otger than that, fantastic site!