Deemed Measure of Expectation and Performance: AFL Round 2

Darren McSweeney Roar Pro

By Darren McSweeney, Darren McSweeney is a Roar Pro

 , , ,

12 Have your say

    After two rounds, the big picture is starting to form for DMXP and we have some big sliders as well as some major climbers, including a nine place rise for one team.

    We will expect some a few more teams to jump around in the coming weeks while everyone settles into the natural order of things.

    First Greater Western Sydney 3.738 (no change since last week) (Ladder first)
    Another solid win and the Giants maintain the top spot. They get bonuses for travel and a six-day break.

    Second Gold Coast 3.603 (+2 since last week) (Ladder third)
    Who’d have thought the Suns would be at this end of the ladder, even if it is only after two rounds.

    Travel gives them a bonus, as does their score of 101, and a tiny bonus (0.005) from less than seven days break having played Saturday night in Round 1 then in the afternoon this week.

    Third Hawthorn 3.043 (-1) (Ladder fourth)
    The Hawks are doing well after two rounds. A bonus for their score over 100.

    Fourth Port Adelaide 2.464 (+3) (Ladder second)
    Port Adelaide climb three places with a win away against a higher ranked team. Bonus points for interstate travel boost them a bit further too.

    Fifth West Coast 1.367 (+9) (Ladder eighth)
    The big climbers, West Coast gain nine places from their big margin over the Bulldogs. They also get bonus for travel and scoring 100 points.

    Sixth Melbourne 1.215 (+6) (Ladder sixth)
    Melbourne’s win boost them by six places. They also get three bonuses for travel, a short break and scoring more than 100 points.

    Seventh Fremantle 1.134 (+3) (Ladder 14th)
    Freo’s win over a higher ranked team gives them a handy lift. Their only bonus is from a score over 100.

    Eighth North Melbourne 0.779 (+7) (Ladder fifth)
    A decent win over a higher ranked team gives the Roos a boost. They get a bonus from their six day break.

    Ninth Essendon 0.650 (-6) (Ladder 11th)
    Essendon drop six places by losing to a lower ranked team. Their travel bonus doesn’t affect their position however, even without it they would still be in the same place.

    Tenth Adelaide 0.403 (+6) (Ladder seventh)
    Adelaide climb six places, but from a low starting base, they’ll have to keep putting in the wins. Bonuses apply for a short break and for scoring more than 100 points.

    11th Geelong -0.045 (-2) (Ladder 10th)
    Geelong’s one point loss counter their three point win last week. They gain a bonus from their high score, but otherwise they are pretty much even. Unfortunately for them, even is actually 11th place.

    12th Sydney -0.175 (-7) (Ladder ninth)
    Not many would expect to see the Swans this far down the table. Losing to a lower team by a decent amount sees them lose seven places. They go without any bonuses this week.

    Dane Rampe Sydney Swans AFL 2017

    (Photo by Cameron Spencer/Getty Images)

    13th St Kilda -0.745 (-7) (Ladder 13th)
    The Saints losing to a lower team drops them, but their drop in rating is underpinned by the amount of their loss. They do gain a bonus from their six-day break.

    14th Richmond -0.772 (-6) (Ladder 12th)

    Their big loss sees the Tigers slide six places. They do gain a bonus from travel.

    15th Carlton -2.000 (-4) (Ladder 17th)
    Remarkably, the Blues’ loss to Gold Coast didn’t hurt them, as they were ranked considerably lower than the Suns. No bonuses mean that they are ranked exactly as they should have been after two losses.

    16th Brisbane Lions -2.182 (-3) (Ladder 15th)
    Brisbane’s two losses were expected, so they are ranked about where they should be. They have no bonuses this week.

    17th Collingwood -2.731 (no change) (Ladder 16th)
    The second loss for the Pies was expected, so their placement seems appropriate at the moment. They also have no bonuses.

    18th Western Bulldogs -3.601 (no change) (Ladder 18th)
    Expected to beat the Eagles, the Bulldogs are now finding themselves entrenched at the bottom. They will need some big improvements to turn things around, but this is where the Swans were last year and they finished atop the DMXP ratings in 2017.

    It wasn’t a good week for DMXP tipping, with only Adelaide, Fremantle and Hawthorn getting up. The total after two rounds in nine from 18.

    Next week, we tip Carlton 83-82 over Collingwood, Port Adelaide 86-76 over Brisbane, Melbourne 92-91 over North Melbourne, Fremantle 91-83 over Gold Coast, St Kilda 89-79 over Adelaide, GWS 96-84 over Sydney, Richmond 89-87 over Hawthorn, Essendon 91-87 over Western Bulldogs, and West Coast 92-72 over Geelong.

    Getting hassled by a parent or partner about spending too much time playing video games? Now, you can tell them the story of how some ordinary gamers scored $225k for just seven weeks of work.

    Oldest | Newest | Most Recent

    The Crowd Says (12)

    • April 5th 2018 @ 9:34am
      me too said | April 5th 2018 @ 9:34am | ! Report

      I’m not sure on what this ladder brings at all. it will be wildly inaccurate early and then simply mirror the real ladder. does it take quality of opponent into consideration? the 6 day break bonus for the saints? why when they played a team on a 6 day break coming back from a long road trip played in horrendous condiitons – did that make it in as a penalty?
      I’m not surprised it wasn’t a good week.

      • Roar Pro

        April 5th 2018 @ 3:15pm
        Darren McSweeney said | April 5th 2018 @ 3:15pm | ! Report

        The rating does certainly take the quality of the opposition into consideration. Of course what the rating considers quality is based on its own ratings, so it may not align with what the football community considers to be quality.

        All interstate travel and rest breaks of less than 7 days are taken into consideration regardless of your oppositions rest. It doesn’t take the return from travel in the following week into the calculation, although it was something I had thought about adding in, I couldn’t get it to work. Likewise there is no alteration for the weather. I can’t formulate that reliably.

        I’ve reverse engineered the results going back to 2010 and it doesn’t always reflect the ladder. Last year it rose Sydney to #1 position by the end, reflecting their end of season form. In some ways it is reflective of Ryan’s Pythagorean wins or even Gordon P Smith’s ELO-FF ratings. I created the rating mostly as a mathematic experiment and to play with Excel spreadsheets and see if I could tip mathematically.

        So take from it what you will. If it doesn’t bring anything for you….

        • April 6th 2018 @ 6:58pm
          me too said | April 6th 2018 @ 6:58pm | ! Report

          Fair enough Darren, the proof will be in the pudding. Gut intuition and years of observation is currently well ahead of you, but it’s a marathon not a sprint.

    • Roar Rookie

      April 5th 2018 @ 9:48am
      Lamby said | April 5th 2018 @ 9:48am | ! Report

      So how do you get St Kilda 89-79 over Adelaide when Adelaide is 0.403 and St Kilda is -0.745?

      • Roar Pro

        April 5th 2018 @ 2:59pm
        Darren McSweeney said | April 5th 2018 @ 2:59pm | ! Report

        I don’t have the spreadsheet with me to confirm the maths but it seems that St Kilda -0.745 in 13th is only just a little bit behind Adelaide on 0.403 in 10th.

        Adelaide 0.403 x (1.2 x (13-10) /24 ) + 90 = 90.06045
        St Kilda -0.745 x (1.2 x (10-13) /24 ) + 90 = 89.88825

        Adelaide then lose 2 goals because of travel, dropping them down to 78
        There’s obviously some difference in the rounding that I didn’t take into consideration either.

        • Roar Rookie

          April 5th 2018 @ 4:48pm
          Lamby said | April 5th 2018 @ 4:48pm | ! Report

          I think your maths/algorithm has a bug/issue.

          Assume Adelaide and St. Kilda keep the same value (0.403 and -0.745)
          But instead Adelaide are 1st and St. Kilda are 18th.
          Then your equations are:

          Adelaide 0.403 x (1.2 x (18-1) /24 ) + 90 = 0.403 x (0.85) + 90 = 90.34
          St Kilda -0.745 x (1.2 x (1-18) /24 ) + 90 = -0.745 x (-0.85) + 90 = 90.63

          So, in this scenario Adelaide are higher on the ladder and have a larger ‘value’ but end up with lower score. Your 2 negative numbers multiplied create a +ve number – which results in a larger value to be added to 90.

    • April 5th 2018 @ 2:11pm
      Pope Paul VII said | April 5th 2018 @ 2:11pm | ! Report

      Love it. Except North will donk Melbourne. Where is Buckers Buckland and his pythagoran wins?

    • April 5th 2018 @ 2:14pm
      Jorge said | April 5th 2018 @ 2:14pm | ! Report

      yeah, I fail to see the “expectation” in St Kilda & West Coast win. I am a Lions supporter, and if Brisvegas get within 10 pts – THAT IS A WIN !!!!!!!!

      • Roar Pro

        April 5th 2018 @ 3:01pm
        Darren McSweeney said | April 5th 2018 @ 3:01pm | ! Report

        It may be that the figure for my travel factor is too great and penalises the visiting team too harshly. I’m looking at revising that part using historic venue results instead, but that will probably have to wait til next year.

        • April 5th 2018 @ 3:21pm
          dontknowmuchaboutfootball said | April 5th 2018 @ 3:21pm | ! Report

          I’d look at making the travel penalty relative to the relative difference in DMXP. So low DMXP travelling to play high DMXP team = bigger travel penalty than high DMXP team travelling to play low DMXP team.

          Having said that, I don’t really follow the maths or all aspects of the thinking behind this system, so I don’t really know what I’m talking about (as per usual).

    • April 5th 2018 @ 4:17pm
      Pope Paul VII said | April 5th 2018 @ 4:17pm | ! Report

      One little humanoid factor there seems to be more tons (yay!) kicked. I reckon the predicted scores will be lower than the ones that transpire, barring rain.

    • April 5th 2018 @ 8:35pm
      Doctor Rotcod said | April 5th 2018 @ 8:35pm | ! Report

      There are seven teams already averaging over 100 points a game – drier part of the season ,I know and there’s 6 teams averaging over 100 points kicked against them, including Geelong and Richmond. Early days,but.
      Is there a way that you can decrease the negative effect of having a high score kicked against you if your opposition are regularly getting high scores anyway, or conversely, if the team you play is regularly being scored heavily against, then could the positive effect of your victory be equally reduced?

    Have Your Say

    If not logged in, please enter your name and email before submitting your comment. Please review our comments policy before posting on the Roar.

    , , ,