The Supercoach Scoring System

[Spartan]

At the start of the season there was a thread created that Supercoach was rigged and that certain players were often favoured by those sneaky bastards down at Champion Data. With the exception of a couple of posters most of us just had a bit of a laugh at this modern day conspiracy theory.

I guess i created this thread for a bit of a rant into the scoring system coupled with a discussion on the actual system. I must admit i'm still a little hazy on how it works and i'm too lazy to check. However at face value it seems that the system itself is seriously flawed. Players apparently lose points for not impacting the game and these points are then handed over to players impacting the game. How does SC choose who it robs points off?

As an example, I was watching the game Brisbane vs Richmond last week. Browndog (who i don't have in my team) was a monster. He won the hard ball, took contested marks, kicked goals, etc etc. However at the end of the game he was awarded a 'massive' 97 pts.

On the other hand, Butler (who i also don't have but i kept an eye on him cause my best mate has him and we're pretty competitive ) in my opinion played an absolute shocker for us last night. He was clangering it all over the place and was as ineffective as all hell. At the end of the game i checked the scores again.. 87 points!

I've been keeping a little eye on the players that seem to pull good scores even when playing shit and vice versa.. and when in doubt will be trading more in favour of these players as opposed to others.

Now i'm sure each of us have similar stories from various games each week. Which begs the question..

Who shot JFK?

 

[GUMBLETRON]

Given how scores are weighted to determine a player's impact on a specific game, you can only really compare players' score-to-performance ratios within that game. Example, Adcock pulled a monster 140+ earlier in the season with woeful efficiency, but because it was pissing down and everyone was sloppy that was effectively taken out of the equation. As far as 'robbing' players, every time one player touches the ball it equally robs every other player on the field. If Laycock nabbed the ball at the first bounce then both teams sat down and watched him fall over himself for two hours, he'd score 3300.

That said, the SuperFooty match reports say Butler had 70% efficiency last night, and J.B. only 58% last week.

 

[TheVelourFog]

One thing that I want to know is that last year some players who kicked a winning goal or point were awarded extra points and it resulted in some huge scores like Wonaeamirri vs Freo, and Minson kicking a match saving point against Richmond.

This year Zaharakis kicks the winning goal against Collingwood and his score is 59. Is this extra points system scrapped or was Zaha just on a score like 9 at 3/4 time.

 

[ocat1979]

I'd just like to see Champion Data tell us EXACTLY how the point scoring works. I have the general idea, with points weighted and all that but i still scratch my head at some scores players get. I think the 3300 points awarded to each game is a good idea because it takes the weather conditions and style of play of the result. Just like to know more about the weighting system

 

[GUMBLETRON]

This year Zaharakis kicks the winning goal against Collingwood and his score is 59. Is this extra points system scrapped or was Zaha just on a score like 9 at 3/4 time.

Ten soft touches at 60% efficiency, so without that goal his SC would've been well below his 44 DT. See Tredrea's 184 this week for a solid game + last-minute match-winner score.

I'd just like to see Champion Data tell us EXACTLY how the point scoring works.

Wouldn't be very good for business.

 

[Un_OC]

The only player i think should get a significant bonus is Fev.

Supercoach points are meant to be based around how important the payer was to teh game rather tahn what they did. i.e. if someone made a big impact they willg et more points than someone who always touches the ball but turns it over and never actualyl does anything.

Fev sacrificed his own (DT) game to help teh blues, and through his game he helped teh blues win by... he should score heaps, but i don't know how tehy would score this... but he did heaps by doing nothing.. if you follow me

 

[GUMBLETRON]

It's a flaw for sure, but like you said, how do you score a player who doesn't active do "something"? Same reason taggers and key defenders score poorly.

 

[Knowlan]

Could some explain to me how Boak and C.Cornes got the 84 today?

Boak: 60% Eff 25 Disposals 18K 7H 1G 0B 0CM 5UM 9T 6CL 5C 2FF 0FA 8CP 16UP 6I50 2A
Cornes: 75% Eff 16 Disposals 11K 5H 1G 0B 2CM 4UM 3T 2CL 1C 1FF 1FA 6CP 9UP 4I50 0A

Obviously Boak's 5 clangers are going to cost him, but 9 tackles, no FA, where as Cornes has 1FA 6 less tackles. I'm not exactly sure on the scoring of tackles but from memory, Sewell had 11 on Friday, huge score, Paddy Ryder had 13 Anzac day, huge score.

I know DT doesn't count for much here, but Boak scored 127 there and C Cornes only 77. Boak was on 71 at half time for SC, then only managed anther 13 points in a half, whilst doubling his DT score? Something seems whack to me.

 

[RWnBluey]

It's a flaw for sure, but like you said, how do you score a player who doesn't active do "something"? Same reason taggers and key defenders score poorly.

Thats why things like spoils should be included as they can almost have just as big an impact on a match as kicking a goal can

 

[Spartan]

Agree with most of your comments fodzilla.

However i think a hybrid of SC/DT would be the perfect solution.

For example, what about a player who dominates the first half and sets his team up for the win and then goes quiet in the second half?

Also the definition of effectiveness as per Champion Data / SC is dodgy at best. Often a player will put the ball in front of a team mate to allow him to run onto it as opposed to hitting him on the chest which would slow down the momentum. This however is designated as an ineffective disposal.

Then we have those players in and under winning hard balls but the ball goes straight to an opponent within the confines of a telephone box. Probably labelled as a clanger.

You're right when you say it should be a player's impact on a game. Few players impacted the game more than Browny last week vs Richmond.

Theres too much hocus pocus when determining the final scores.

I want the truth.

And yes, I can handle the truth!

 

[anubis]

Adam Goodes was on 137 at 3/4 time last night, he had 5 possessions in the last quarter 2 contested and 1 tackle had no clangers in the last either and finished on 126 i thought that was odd lol.

 

[Rimmer]

A bit of an eye opener this thread,

I have never taken too much notice of the scoring system for SC, I was warned it was different to DT (which is pretty much stats driven) but I never realized how subjective the scoring system for SC was. Scores can be decided on the whims and personal interpretations of the 'back room boys' rather than what we see when watching the matches. (How do you decide someone who gets X possessions is 58% or 63% effective with them?)

My 'clever' selections are pretty irrelevant it seems and my success or otherwise is more akin to picking the winner of a horse race than astute judgment.

... one thing though ... I will be watching for the return of that Laycock bloke ... 3300 points for falling over for 2 hours ... sounds like a Super Coach gun to me ... better keep an extra trade handy

 

[Catsace]

Interesting reading the prospectus from 2007, the 2nd edition, which generally waxed lyrical about Sydney and West Coast's game style.
It said quote Pp 353 2006 Prospectus:-

'in 2005 the Rankings introduced weightings depending on proximity to goal and the state of play in the game. It is clear from the data that turning over possession within range of goal is more hurtful than losing it on the wing. It was also decided to reward players who were putting their hands up when the game was there to be won and the scores were close, over those who racked up meaningless possessions when the result was already in the bag.

Mthematically this state-of-play weighting is based on the derivative of match-winning probability with respect to a score change. We also normalised the total number of ranking points in a match to 3300- previously players in teams with a slow, methodical game plan were effectively being penalised because they couldn't rack up the same number of possessions as the Western Bulldogs.

Personally i reckon that the Swans were actually an exception with their game style to what generally is required to play winning footy. I am wondering if their decision based on mid 2000's era footy is now misleading since Geelong has changed the game with their huge possession tallies. I'd go so far as to say Geelong's scores are penalised to an extent as they might get say a 130 to an Ablett or a Bartel for a 40-possie game whereas a Lion like Black or say a Burgoyne will score a huge 170+ for a game with several less possessions. It is certainly different horses for courses, as the scoring will vary from one game to another depending on the tempo, score closeness etc.

Also though, if Geelong ice a game by the 2nd quarter, do they miss out on the freebie points that other sides may gain simply because they aren't good enough to take a game away from their opponents. Melbourne may have an advantage as they can stay competitive all game, keeping the scores close, thus adding impact to last quarter scores by players that Geelong players will rarely ever get. Just interesting how it goes now, and am wondering if the 3300 point standard was based on a style of footy atypical for today's times.

 

[Rimmer]

Adam Goodes was on 137 at 3/4 time last night, he had 5 possessions in the last quarter 2 contested and 1 tackle had no clangers in the last either and finished on 126 i thought that was odd lol.

I am glad FanFooty does't have running tallies of Super Coach and that we only get them after the game. This sort of thing would really give me the shits

 

[jakee]

like that time bartel got like 41 possesions and just 104 points or something like that. couldnt believe it..

i have no idea how it works.. i dont think anyone fully understands it..? haha.

 

[Spartan]

I am glad FanFooty does't have running tallies of Super Coach and that we only get them after the game. This sort of thing would really give me the shits

I have Goodes.

 

[Nasma]

'in 2005 the Rankings introduced weightings depending on proximity to goal and the state of play in the game. It is clear from the data that turning over possession within range of goal is more hurtful than losing it on the wing. It was also decided to reward players who were putting their hands up when the game was there to be won and the scores were close, over those who racked up meaningless possessions when the result was already in the bag.

Mthematically this state-of-play weighting is based on the derivative of match-winning probability with respect to a score change. We also normalised the total number of ranking points in a match to 3300- previously players in teams with a slow, methodical game plan were effectively being penalised because they couldn't rack up the same number of possessions as the Western Bulldogs.

That's a really interesting quote - where do you get the prospectus from?

Personally i reckon that the Swans were actually an exception with their game style to what generally is required to play winning footy. I am wondering if their decision based on mid 2000's era footy is now misleading since Geelong has changed the game with their huge possession tallies. I'd go so far as to say Geelong's scores are penalised to an extent as they might get say a 130 to an Ablett or a Bartel for a 40-possie game whereas a Lion like Black or say a Burgoyne will score a huge 170+ for a game with several less possessions. It is certainly different horses for courses, as the scoring will vary from one game to another depending on the tempo, score closeness etc.

The normalisation of 3300 points a game accounts for this. If we take a 30 possession, 85% efficiency Burgoyne game in the wet against the Swans, where the next best player has 20 possessions at 70%, he will rack up a monster score. If he has that same game against Geelong at the dome where there are 5 other players with similar stats, his score will be much lower as the 3300 points will have to be shared with some other high-stat performances.

Also though, if Geelong ice a game by the 2nd quarter, do they miss out on the freebie points that other sides may gain simply because they aren't good enough to take a game away from their opponents. Melbourne may have an advantage as they can stay competitive all game, keeping the scores close, thus adding impact to last quarter scores by players that Geelong players will rarely ever get. Just interesting how it goes now, and am wondering if the 3300 point standard was based on a style of footy atypical for today's times.

Geelong don't miss out on points. The decreased weighting when the game is won applies to both teams. What it means is that performances by both teams will get more weight towards the beginning of the game. As Geelong ended up dominating, these will mainly be Geelong players. Later on, when the margin is large, it's harder for players from both teams to change their scores. If we take Saturday's match as an example, players who did well at the start (from both teams) get more credit than those who did well at the end.

Of course, the above is for a match where one team kicks away at the beginning.

 

[sblack]

One thing that I want to know is that last year some players who kicked a winning goal or point were awarded extra points and it resulted in some huge scores like Wonaeamirri vs Freo, and Minson kicking a match saving point against Richmond.

This year Zaharakis kicks the winning goal against Collingwood and his score is 59. Is this extra points system scrapped or was Zaha just on a score like 9 at 3/4 time.

Pretty sure they have removed this feature if I am not mistaken. They must have felt that awarding 60+ for 1 possession to be too much of a bonus.

As explained in the 'SC is rigged' thread by a CD rep...the average score for a player is 75 (3300/44). Therefore, a score of 150 is the equivalent of 2 average players contribution.

So, with taking this into account..they might have felt that 1 possession cannot really be equivalent of 1 average player over the course of a whole game.

This I am not sure exactly.

Ten soft touches at 60% efficiency, so without that goal his SC would've been well below his 44 DT. See Tredrea's 184 this week for a solid game + last-minute match-winner score.

Wouldn't be very good for business.

Tredrea's score had nothing to do with game winning possessions. He was on something around 130-140 at 3/4 time. He took about 4 contested marks. Kicked 1-2 goals, had 1 goal assist with mostly contested ball. The 40-50 he scored was all standard points.

Could some explain to me how Boak and C.Cornes got the 84 today?

Boak: 60% Eff 25 Disposals 18K 7H 1G 0B 0CM 5UM 9T 6CL 5C 2FF 0FA 8CP 16UP 6I50 2A
Cornes: 75% Eff 16 Disposals 11K 5H 1G 0B 2CM 4UM 3T 2CL 1C 1FF 1FA 6CP 9UP 4I50 0A

Obviously Boak's 5 clangers are going to cost him, but 9 tackles, no FA, where as Cornes has 1FA 6 less tackles. I'm not exactly sure on the scoring of tackles but from memory, Sewell had 11 on Friday, huge score, Paddy Ryder had 13 Anzac day, huge score.

I know DT doesn't count for much here, but Boak scored 127 there and C Cornes only 77. Boak was on 71 at half time for SC, then only managed anther 13 points in a half, whilst doubling his DT score? Something seems whack to me.

Can't just look at that and figure out a general score. Uncontested and contested marks from opposition kicks are more valuable than keeping possession. Tackles are also weighted differently. I used to notice different scoring towards tackles that wrapped up the player and didn't allow any possession compared to tackles that resulted in a handball worth more. Also have to look at goal assists to behind assist.

Thats why things like spoils should be included as they can almost have just as big an impact on a match as kicking a goal can

Spoils do have a point value...not published, but they do score. Either 1 or 2. When GDL displayed scores, I noticed on many occasions that players like Merrett or Presti would increase after a spoil.

Kick ins, opposite to popular thought, have a value. I think it was 1 for a short kickin and 2-3 for a long kickin (usually with the added R50).

Agree with most of your comments fodzilla.

However i think a hybrid of SC/DT would be the perfect solution.

For example, what about a player who dominates the first half and sets his team up for the win and then goes quiet in the second half?

Also the definition of effectiveness as per Champion Data / SC is dodgy at best. Often a player will put the ball in front of a team mate to allow him to run onto it as opposed to hitting him on the chest which would slow down the momentum. This however is designated as an ineffective disposal.

Not really...if the player 'gathers' the ball that was kicked towards him...the kicker is awarded with a efficient disposal. Although, efficient disposals are usually classified by the result of the kick. The definition of a short and long kick explains this. Say a kick that goes 25m to a 1-1 and results in a HBG (doesn't matter which team), is usually classified as ineffective. But a 20m kick to space where a player by himself runs onto it is usually effective

Then we have those players in and under winning hard balls but the ball goes straight to an opponent within the confines of a telephone box. Probably labelled as a clanger.

This is annoying sometimes...but is usually decreased effect as a HBG reduces the negative score of a clanger

You're right when you say it should be a player's impact on a game. Few players impacted the game more than Browny last week vs Richmond.

He does butcher it sometimes
Theres too much hocus pocus when determining the final scores.

I want the truth.

Adam Goodes was on 137 at 3/4 time last night, he had 5 possessions in the last quarter 2 contested and 1 tackle had no clangers in the last either and finished on 126 i thought that was odd lol.

There was probably a lot of contested ball, a few goals kicked and a lot of lead changes.. If he wasn't influencing the game nearing the end...his points were allocated elsewhere. Although that game resulted in the harshest scaling you will see. There probably won't be another game this year where multiple players lose around 10 points due to scaling

 

[HibSam]

Kick ins, opposite to popular thought, have a value. I think it was 1 for a short kickin and 2-3 for a long kickin (usually with the added R50).

This could be why Drummond scores so well- takes the kick-outs and more often then not kicks long to a target.

Wait, so let me get this straight. In a game the max points that the 2 teams get combined can't go over 3300? If true this would make sense, because players like Goodwin and Drummond always get good sc numbers as there is nobody else in their team to take points off them. Whereas Bartel dominates a game, but has to share his points with the other Geelong guns...

I didn't know this 3300 rule at the start, but for next year i will defs take it into account when picking my team.

 

[KUNG FU]

I do think the system is sketchy by virtue of the fact they have tried to overcook the stats with 'match-winning' points and all that.

Having said that, the idea of a 'conspiracy theory' is a stretch. I would have thought they'd have tried to undermine teams called "Sheedy'sSCteam is crap" and "DT> SC" from being in the top 50.

There is a pattern for it but I wish they were more up front about it.

 

[Spartan]

Picking multiple players from one team appears to be ill advised unless that team is dominant (as is the case with Geelong and St. Kilda this year).

Stupid Supercoach.

 

[Parasite]

Spartan = DT Troll.

 

[Spartan]

Spartan = DT Troll.

 

[sblack]

This could be why Drummond scores so well- takes the kick-outs and more often then not kicks long to a target.

Wait, so let me get this straight. In a game the max points that the 2 teams get combined can't go over 3300? If true this would make sense, because players like Goodwin and Drummond always get good sc numbers as there is nobody else in their team to take points off them. Whereas Bartel dominates a game, but has to share his points with the other Geelong guns...

I didn't know this 3300 rule at the start, but for next year i will defs take it into account when picking my team.

Drummond is also very very efficient, kicks Long rather than short. Gets plently of loose ball gets...and plenty of R50 and I50's. All makes him a gun

You do have to factor in the 3300. I made the choice this week of Deledio over SJ, just because of this fact.

 

[Jack4814]

Could some explain to me how Boak and C.Cornes got the 84 today?

Boak: 60% Eff 25 Disposals 18K 7H 1G 0B 0CM 5UM 9T 6CL 5C 2FF 0FA 8CP 16UP 6I50 2A
Cornes: 75% Eff 16 Disposals 11K 5H 1G 0B 2CM 4UM 3T 2CL 1C 1FF 1FA 6CP 9UP 4I50 0A

Obviously Boak's 5 clangers are going to cost him, but 9 tackles, no FA, where as Cornes has 1FA 6 less tackles. I'm not exactly sure on the scoring of tackles but from memory, Sewell had 11 on Friday, huge score, Paddy Ryder had 13 Anzac day, huge score.

I know DT doesn't count for much here, but Boak scored 127 there and C Cornes only 77. Boak was on 71 at half time for SC, then only managed anther 13 points in a half, whilst doubling his DT score? Something seems whack to me.

I love the sc scoring system, especially compared to the dream team system which.. isnt even worth mentioning. But Boak losing 20 points in the last quarter (didnt he kick a goal in the last?) is absolutely ridiculous