Scoring Anomalies – Round 21

Written by The Salamander on August 15 2019

High praise…

Well, what can I say? This column is now famous!

Fantasy Freako seems to have misinterpreted my breakdowns as suggesting that players directly get points for clearances; in light of this, I have re-worded that blindspot warning, as you will see below.

I’m not entirely sure whether “no points for score involvements” meant no direct points (which I was never suggesting), or if it also meant that there is no favourable scaling for them (which seems like a bit of an oversight, if the system wants to favour influence on games over raw stat accumulation, by I digress). In any case, I have temporarily disabled that particular blindspot warning, until this is clarified.

The only other change I have made this week is that I have added running bounces (2 points each) to the scoring formula. It was pointed out to me about a month ago that they are part of it; I thought I had already added them, but it turns out that I forgot. There are only a handful per game anyway, so it’s not like it has been blowing the scores out substantially, but I figured it was good to make the formula as complete as possible.

With that out of the way, Dane wanted to know how Gawn and Grundy’s SyntheticCoach scores lined up with their real ones.

CD: Gawn 99, Grundy 86
SyntheticCoach: Gawn 117 (129.5 pre-scaling), Grundy 96 (106.5)

Max Gawn (Melbourne)
• Effective kicks: 4 => 16
• Effective handballs: 4 => 6.0
• Clangers: 4 => -16
• • of which frees-against: 0
• Ground-ball gets: 2 => 9.0
• Bounces: 0 => 0
• Goals: 1 => 8
• Behinds: 0 => 0
• Marks: 1 => 2
• • of which contested: 0 => 0
• • of which intercepts: 0 => 0
• Tackles: 3 => 12
• Frees-for: 5 => 20
• Hitouts-to-advantage: 13 => 65
• Goal assists: 0 => 0.0
• Spoils: 3 => 6
• Other one-percenters: 1 => 1.5
Total: 129.5

Blindspots:
• SyntheticCoach knows nothing about when the game was on the line, let alone who did what when it was. This can have a massive effect on scaling.

Brodie Grundy (Collingwood)
• Effective kicks: 3 => 12
• Effective handballs: 3 => 4.5
• Clangers: 5 => -20
• • of which frees-against: 5
• Ground-ball gets: 6 => 27.0
• Bounces: 0 => 0
• Goals: 0 => 0
• Behinds: 0 => 0
• Marks: 2 => 4
• • of which contested: 0 => 0
• • of which intercepts: 1 => 2
• Tackles: 7 => 28
• Frees-for: 1 => 4
• Hitouts-to-advantage: 9 => 45
• Goal assists: 0 => 0.0
• Spoils: 0 => 0
• Other one-percenters: 0 => 0.0
Total: 106.5

Blindspots:
• Gathers from hitouts are worth 2 points each. Player had 7 clearances – perhaps some came from that source? (N.B. in this case, they probably didn’t – he probably took them directly out of the ruck contest).
• SyntheticCoach knows nothing about when the game was on the line, let alone who did what when it was. This can have a massive effect on scaling.

AllSaints also wanted to know why Adam Treloar and Clayton Oliver scored so differently.

CD: Oliver 96, Treloar 127
SyntheticCoach: Oliver 117 (129.5), Treloar 116 (128.5)

Clayton Oliver (Melbourne)
• Effective kicks: 6 => 24
• Effective handballs: 15 => 22.5
• Clangers: 7 => -28
• • of which frees-against: 2
• Ground-ball gets: 10 => 45.0
• Bounces: 0 => 0
• Goals: 0 => 0
• Behinds: 0 => 0
• Marks: 7 => 14
• • of which contested: 0 => 0
• • of which intercepts: 0 => 0
• Tackles: 11 => 44
• Frees-for: 2 => 8
• Hitouts-to-advantage: 0 => 0
• Goal assists: 0 => 0.0
• Spoils: 0 => 0
• Other one-percenters: 0 => 0.0
Total: 129.5

Blindspots:
• Gathers from hitouts are worth 2 points each. Player had 7 clearances – perhaps some came from that source?
• SyntheticCoach knows nothing about when the game was on the line, let alone who did what when it was. This can have a massive effect on scaling.

Adam Treloar (Collingwood)
• Effective kicks: 3 => 12
• Effective handballs: 17 => 25.5
• Clangers: 2 => -8
• • of which frees-against: 0
• Ground-ball gets: 13 => 58.5
• Bounces: 0 => 0
• Goals: 0 => 0
• Behinds: 0 => 0
• Marks: 4 => 8
• • of which contested: 0 => 0
• • of which intercepts: 0 => 0
• Tackles: 6 => 24
• Frees-for: 0 => 0
• Hitouts-to-advantage: 0 => 0
• Goal assists: 2 => 7.0
• Spoils: 0 => 0
• Other one-percenters: 1 => 1.5
Total: 128.5

Blindspots:
• Gathers from hitouts are worth 2 points each. Player had 6 clearances – perhaps some came from that source?
• Player had 12 uncontested possessions that did not come from a mark. Perhaps some came from handball-recieves (1.5 points each)?
• SyntheticCoach knows nothing about when the game was on the line, let alone who did what when it was. This can have a massive effect on scaling.

Based on that, I’d say handball receives probably had a bit to do with it, although scaling may also have been a factor.

He also wanted to know how this compared to Toby Greene’s high-clanger output.

CD: 122
SyntheticCoach: 129 (138)

Toby Greene (GWS Giants)
• Effective kicks: 13 => 52
• Effective handballs: 9 => 13.5
• Clangers: 10 => -40
• • of which frees-against: 2
• Ground-ball gets: 13 => 58.5
• Bounces: 0 => 0
• Goals: 0 => 0
• Behinds: 0 => 0
• Marks: 8 => 16
• • of which contested: 0 => 0
• • of which intercepts: 2 => 4
• Tackles: 5 => 20
• Frees-for: 3 => 12
• Hitouts-to-advantage: 0 => 0
• Goal assists: 0 => 0.0
• Spoils: 1 => 2
• Other one-percenters: 0 => 0.0
Total: 138

Blindspots:
• Gathers from hitouts are worth 2 points each. Player had 8 clearances – perhaps some came from that source?
• SyntheticCoach knows nothing about when the game was on the line, let alone who did what when it was. This can have a massive effect on scaling.

Given that he was just about the only GWS player to actually bother showing up on Friday night, his score seems a little low. Granted, GWS was terrible, and you generally expect the scaling to favour the winning team, but it wasn’t a particularly high-point-allocation game, so I’m not sure that explains much in this case. I don’t have Greene myself, but if I did, I would probably be feeling a little cheated at this point.

Finally, the greatest anomaly this week has probably come not from Champion Data, but from the Carlton board. Titus O’Reily perhaps said it best:

David Teague has been appointed senior coach of Carlton in an alarming display of competence by the board.

All I can really add to this is to borrow a line from then-Senator Palpatine: “A surprise, to be sure, but a welcome one.”



Were there any scores that seemed off to you on the weekend? Let me know in the comments below, and I’ll let you know what SynetheticCoach has to say about it!

P.S. If anyone reading wants to build their own SuperCoach machine, or conduct some other weird science experiment, and would like the raw data to facilitate that, drop me a line in the comments, and I’ll send it across to you.

11
0


Leave a comment / Scroll to bottom

17 thoughts on “Scoring Anomalies – Round 21”

  1. Brilliant stuff mate. Jack CRISP went on a bounceathon last week … and then nailed it through the big ones. Thank god for us owners. Could’ve been a lot worse!! 😉

    4

    0
    1. He wasn’t good early lots of clangers, think he may have given a 50m penalty away too which goes down as a clanger but i think is even more negative in SC but i’m not sure.

      1

      0
    1. According to the formula laid out in the official SuperCoach terms and conditions, the aforementioned stats are worth +5, +4, and -4 respectively.

      As an aside, I’m on the record as saying that I think 4 points is excessive for a free-kick-for, although I think the reason that they do it is so that the number of points allocated for free kicks for and against sums to zero, and thus (in the aggregate) don’t take up any of the precious 3300 points.

      2

      0
        1. Given that there is a delay between stuff happening and scores showing up, it can sometimes be hard to tell exactly what points are being awarded for when they show up, but I’ll try to have a closer look at it on the weekend for you 🙂

          4

          0
          1. Interesting. I’ve heard before that free kicks were considered contested, but I haven’t been able to find a source to back it up. I must say that it seems a bit strange – a free kick is surely the least contested kind of possession that you can have?

            1

            0
            1. Funny actually makes a lot of sense to me, I guess you could say that perpetrator of the FK has lost the contest by giving away a FK.

              1

              0
  2. I think the SC system for scoring is fantastic, where the my issues come is with CD’s recording of events, like what is and isn’t tackles can be extremely different not just game to game but minute to minute, this was back in 2018 https://www.youtube.com/watch?v=tGFQ0ZhaRQs & they’ve got more consistent but it’s really iffy at times.

    Couple other things, a contested clearance that’s marked by the opposition (turnover) is a clanger worth the same as an uncontested disposal turnover should just be ineffective IMO

    One thing i see a lot of is say a Oliver type shooting out a quick HB that a teammate then fumbles & that’s goes down as an ineffective disposal which i’m fine with but it should never go down as a clanger, which it does if it’s deemed uncontested, if the teammates had a good chance of taking possession.

    Anyway couple gripes that really mean nothing because i’ll be back next season anyway haha.

    6

    0
    1. Thanks for that link, Dann. I agree that two or three of those are bit iffy.

      Do you happen to have a source for CD recording turnovers from contested clearances as clangers? It wouldn’t necessarily shock me if they recorded it that way, but it’s the first that I am hearing of it.

      1

      0
      1. The only source is eye test sorry mate, but its a pretty eas one to see happen in real time then immediately go on the stats sheet as a clanger.

        2

        0

Leave a Reply to The Salamander Cancel reply

Your email address will not be published. Required fields are marked *