Coach maximum Discrepant Anomalies In Coach OpinionS Index(D.A.I.C.Os index)

[Meteoric Rise]

I'd like to see it continue too, but CBF crunching the numbers.

I'd request the known minimum discrepancy, as I described above, and over how many games those votes were spread. I don't think there is any benefit using the maximum potential discrepancy, when it is unknown and can range from 0-4.

So using the 2024 top 2:

Daicos 13 votes over 13 games, neatly works out to 1 vote per game.
Cripps 6 over 13 games, something like 0.5 votes per game.

Do everybody for every game this year and see what it tells us Meteoric Rise

It is a bit laborious, but it is a long season, so even 1 player season per day will give us good amount of data to pick apart.
For a start I just want to focus on the max possible discrepancy figures. There will be plenty of time to go back and fill in the minimum figures later. Ideally, you want both which gives you the range of what is possible. At some stage I will collate what we have into a table as well.

To be honest I don't want to make it all about Daicos either. It is just he is the one getting attention(probably rightly so) so it is nice to see where he and others fit into the bigger picture. It is very early days, but the impression I get early is that this has been going on a bit longer and on a larger scale than we might have realised. But we will get more clarity on that.

 

[BF Tiger]

Perception is reality… and the widely held perception by oppositions fans and media here is that Fly has a strong man-crush on Nicky and it translates into AFLCA votes.

 

[Mr Meow]

It is a bit laborious, but it is a long season, so even 1 player season per day will give us good amount of data to pick apart.
For a start I just want to focus on the max possible discrepancy figures. There will be plenty of time to go back and fill in the minimum figures later. Ideally, you want both which gives you the range of what is possible. At some stage I will collte what we have into a table as well.

To be honest I don't want to make it all about Daicos either. It is just he is the one getting attention(probably rightly so) so it is nice to see where he and others fit into the bigger picture. It is very early days, but the impression I get early is that this has been going on a bit longer and on a larger scale than we might have realised. But we will get more clarity on that.

Was there a part of "I don't think there is any benefit using the maximum potential discrepancy, when it is unknown and can range from 0-4" that you didn't understand? It's a valid point. If you are using max discrepancy then your thread should simply be renamed "A guessing framework for assessing coaches votes integrity".

 

[Meteoric Rise]

Was there a part of "I don't think there is any benefit using the maximum potential discrepancy, when it is unknown and can range from 0-4" that you didn't understand? It's a valid point. If you are using max discrepancy then your thread should simply be renamed "A guessing framework for assessing coaches votes integrity".

If you think there is better method roll your sleeves up and get on with it.

 

[Mr Meow]

If you think there is better method roll your sleeves up and get on with it.

I'd request the known minimum discrepancy, as was described above, and over how many games those votes were spread.

Also there is no onus on me to provide an improved method to invalidate the work you put together. I can just as easily call out a rubbish scientific study without needing to complete an improved paper on the topic myself.

You've actually failed to counter any valid criticisms of your method (and more so the sweeping conclusions you then make from it) so far. "Go on - do better yourself then!" isn't a rebuttal.

 

[sabso]

Was there a part of "I don't think there is any benefit using the maximum potential discrepancy, when it is unknown and can range from 0-4" that you didn't understand? It's a valid point. If you are using max discrepancy then your thread should simply be renamed "A guessing framework for assessing coaches votes integrity".

Put the toy mouse down for a second and work out that 5-4-3-2-1 votes from two coaches can only be spread certain ways.

 

[Meteoric Rise]

I'd request the known minimum discrepancy, as was described above, and over how many games those votes were spread.

Also there is no onus on me to provide an improved method to invalidate the work you put together. I can just as easily call out a rubbish scientific study without needing to complete an improved paper on the topic myself.

You've actually failed to counter any valid criticisms of your method (and more so the sweeping conclusions you then make from it) so far. "Go on - do better yourself then!" isn't a rebuttal.

Lol, you sound like a very important person.

 

[Mr Meow]

Put the toy mouse down for a second and work out that 5-4-3-2-1 votes from two coaches can only be spread certain ways.

The only thing we know for sure is that a player receiving 10 votes was voted BOG by both coaches. That means two people agreed. It doesn't mean on that particular day the coaches had stronger integrity.

There are all sorts of combinations below that and there will be disagreements at times, as there would be between any two random people you might get to complete a 1-5 voting system. If there is stronger agreement between those people it does not mean they had higher integrity or vice versa.

 

[Mr Meow]

Lol, you sound like a very important person.

In other words, again you cannot offer any defence and instead go entirely ad hominem.

That's basically you conceding every point made has been valid and you have no real rebuttal. Cheers.

 

[Meteoric Rise]

The only thing we know for sure is that a player receiving 10 votes was voted BOG by both coaches. That means two people agreed. It doesn't mean on that particular day the coaches had stronger integrity.

There are all sorts of combinations below that and there will be disagreements at times, as there would be between any two random people you might get to complete a 1-5 voting system. If there is stronger agreement between those people it does not mean they had higher integrity or vice versa.

That statement is incorrect. I think you are out of your depth.

 

[Cyclops]

Decent banter thread, but that title surely precludes any sensible discussion.

OP wishes Dusty was as good as Daicos at the same age, I guess jealousy breeds febrile theories.

 

[Meteoric Rise]

Decent banter thread, but that title surely precludes any sensible discussion.

OP wishes Dusty was as good as Daicos at the same age, I guess jealousy breeds febrile theories.

They say never judge a book by its cover Cyclops. I always say it is fine to judge the cover by the cover though. Maybe I should try to jazz the thread title up a bit.

 

[Mr Meow]

That statement is incorrect. I think you are out of your depth.

9 Josh Dunkley (BL)
8 Zac Bailey (BL)
4 James Jordon (SYD)
4 Darcy Wilmot (BL)
3 Jarrod Berry (BL)
2 Dayne Zorko (BL)

Give me your 5-1's for Fagan and Cox from this game. You need to be certain. Saying "well Dunkley received a 5 and a 4, Bailey a 5 and a 3, Wilmot a 2 and a 2" is useless information. You need to know the actual discrepancies and which way they swung.

If Bailey receives a 5 from Fagan and 3 from opposition coach one week, but then receives a 3 from Fagan and 5 from opposition coach the next week - well it looks like a discrepancy is being built up. But it shows no favouritism at all. Your "study" only has any validity at all on the assumption that all coaches would overrate their own star player week in, week out - wherever there is a disagreement. It is a pretty wild assumption, given how subjective performance appraisal is.

Again, you could complete the same study on any two Richmond supporters MVP votes from a season. Their week to week discrepancies would not show a supposed favouritism or overrating of certain players. If the two posters happened to agree about the players and their order each week it would not mean they both had stronger integrity. The conclusions you make simply don't match the flawed data you have haphazardly put together.

 

[Meteoric Rise]

9 Josh Dunkley (BL)
8 Zac Bailey (BL)
4 James Jordon (SYD)
4 Darcy Wilmot (BL)
3 Jarrod Berry (BL)
2 Dayne Zorko (BL)

Give me your 5-1's for Fagan and Cox from this game. You need to be certain. Saying "well Dunkley received a 5 and a 4, Bailey a 5 and a 3, Wilmot a 2 and a 2" is useless information. You need to know the actual discrepancies and which way they swung.

If Bailey receives a 5 from Fagan and 3 from opposition coach one week, but then receives a 3 from Fagan and 5 from opposition coach the next week - well it looks like a discrepancy is being built up. But it shows no favouritism at all. Your "study" only has any validity at all on the assumption that all coaches would overrate their own star player week in, week out - wherever there is a disagreement. It is a pretty wild assumption, given how subjective performance appraisal is.

Again, you could complete the same study on any two Richmond supporters MVP votes from a season. Their week to week discrepancies would not show a supposed favouritism or overrating of certain players. If the two posters happened to agree about the players and their order each week it would not mean they both had stronger integrity. The conclusions you make simply don't match the flawed data you have haphazardly put together.

I don't need to be able to divine the origin of all or even any of those votes to prove your statement incorrect.

And your statement was not only factually incorrect, it is also missing the whole point of the thread, which is to take what we do know, and see what we can make of it.

I make no wild assumptions, you have invented that. We haven't drawn any conclusions yet, I will leave that sort of nonsense to you. Sure we will take a bit of licence and have some fun along the way, but my mind is open at this point. It seems yours is not.

The reason I like the maximum discrepancy method is because within what we find we will certainly have captured every single example of bias(either way.) From there we just need to have faith in our ability to refine the parameters, make reasonable adjustements etc.

If you want to actually be useful, go through all the players I have done so far and work out the minimum discrepancies, so that we have ranges for everybody.

 

[CharlieMortdecai]

For a start I just want to focus on the max possible discrepancy figures.

You're doing yourself a disservice. It is just noise, it adds nothing to draw any meaningful conclusions from. Take the Carlton v GWS game in round 17 last year:


What does it meaningfully tell us to say that any one of those players could have got 4-0? If it was randomised, there's a 40% chance of 4-0, 40% chance of 3-1, and 20% chance of 2-2. Cripps polled in 3 games where the discrepancy was uncertain, and 13 where it was known. Those 3 games have an incredibly disproportionate weight, adding 8 votes, against the 6 votes in the 13 known games. And every one of them could have in fact been 0 discrepancy.

 

[sabso]

You're doing yourself a disservice. It is just noise, it adds nothing to draw any meaningful conclusions from. Take the Carlton v GWS game in round 17 last year:

View attachment 2262006
What does it meaningfully tell us to say that any one of those players could have got 4-0? If it was randomised, there's a 40% chance of 4-0, 40% chance of 3-1, and 20% chance of 2-2. Cripps polled in 3 games where the discrepancy was uncertain, and 13 where it was known. Those 3 games have an incredibly disproportionate weight, adding 8 votes, against the 6 votes in the 13 known games. And every one of them could have in fact been 0 discrepancy.

Two players definitely got 4-0 in that game. Max discrepancy is 4 for all the 4 players and min discrepancy is 0. Don’t reckon that game is especially useful for this analysis, there will be a significant number that aren’t but I guess trying to do something with the games that are more interesting would be somewhat useful.

Or the afl could tell us which coach voted for whom.

 

[Cyclops]

They say never judge a book by its cover Cyclops. I always say it is fine to judge the cover by the cover though. Maybe I should try to jazz the thread title up a bit.

Or you could delete it and move it to the Bay, Meteoric Rise.

The title is the best thing about this thread. The banter potential is high, but the slim premise is logically played out.

 

[Meteoric Rise]

You're doing yourself a disservice. It is just noise, it adds nothing to draw any meaningful conclusions from. Take the Carlton v GWS game in round 17 last year:

View attachment 2262006
What does it meaningfully tell us to say that any one of those players could have got 4-0? If it was randomised, there's a 40% chance of 4-0, 40% chance of 3-1, and 20% chance of 2-2. Cripps polled in 3 games where the discrepancy was uncertain, and 13 where it was known. Those 3 games have an incredibly disproportionate weight, adding 8 votes, against the 6 votes in the 13 known games. And every one of them could have in fact been 0 discrepancy.

You seem to think that once we have found the maximum discrepancies that we are snookered in terms of progressing from there toward our conclusions. This is definitely not the case.

However, I can see your point. The trouble is, as you know, to go through and find probabilities and proceed that way is just too much work to start out with. The other trouble is that it is potentially concealing biased voting and proportioning some of that bias to where it is unlikely to belong, just the same as the system I am using.

What I am doing is looking for all potential bias in votes given to certain nominated players. If we locate no abnormal patterns then we can refine our search a little further.

 

[CharlieMortdecai]

9 Josh Dunkley (BL)
8 Zac Bailey (BL)
4 James Jordon (SYD)
4 Darcy Wilmot (BL)
3 Jarrod Berry (BL)
2 Dayne Zorko (BL)

Give me your 5-1's for Fagan and Cox from this game. You need to be certain. Saying "well Dunkley received a 5 and a 4, Bailey a 5 and a 3, Wilmot a 2 and a 2" is useless information. You need to know the actual discrepancies and which way they swung.

If Bailey receives a 5 from Fagan and 3 from opposition coach one week, but then receives a 3 from Fagan and 5 from opposition coach the next week - well it looks like a discrepancy is being built up. But it shows no favouritism at all. Your "study" only has any validity at all on the assumption that all coaches would overrate their own star player week in, week out - wherever there is a disagreement. It is a pretty wild assumption, given how subjective performance appraisal is.

Again, you could complete the same study on any two Richmond supporters MVP votes from a season. Their week to week discrepancies would not show a supposed favouritism or overrating of certain players. If the two posters happened to agree about the players and their order each week it would not mean they both had stronger integrity. The conclusions you make simply don't match the flawed data you have haphazardly put together.

You're absolutely correct. But we can draw some conclusions about players that coaches more regularly agree about.

Josh Dunkley 5-4
Zac Bailey 5-3
James Jordon 4-0/2-2
Darcy Wilmot 4-0/2-2
Jarrod Berry 3-0
Dayne Zorko 1-1

Jarrod Berry's game was polarising, as was one of Jordon or Wilmot, Bailey to a lesser degree.

If one player is an outlier in terms of such anomalies, maybe it tells us something about their coach, as they are the consistent voter. But the Daicos case shows just how much a single polls-apart viewing can be. Getting 5-0 in the Carlton game last year completely skewed his discrepancy. I've never watched the game apart from the last 5 or 10 mins, but reading the match reports (AFL and ABC), and considering the Brownlow votes, if it was Voss that gave him zero it was a unique interpretation of his game.

Take that out and he has a discrepancy of 8 votes from 12 games, not worlds apart from Cripps's 6 from 13.

 

[CharlieMortdecai]

You seem to think that once we have found the maximum discrepancies that we are snookered in terms of progressing from there toward our conclusions. This is definitely not the case.

However, I can see your point. The trouble is, as you know, to go through and find probabilities and proceed that way is just too much work to start out with. The other trouble is that it is potentially concealing biased voting and proportioning some of that bias to where it is unlikely to belong, just the same as the system I am using.

What I am doing is looking for all potential bias in votes given to certain nominated players. If we locate no abnormal patterns then we can refine our search a little further.

OMG, it is like talking to a brick wall, can you not see that the data are unknown you should exclude them, it is not even like they are numerically significant?

 

[Mr Meow]

You're absolutely correct. But we can draw some conclusions about players that coaches more regularly agree about.

Josh Dunkley 5-4
Zac Bailey 5-3
James Jordon 4-0/2-2
Darcy Wilmot 4-0/2-2
Jarrod Berry 3-0
Dayne Zorko 1-1

Jarrod Berry's game was polarising, as was one of Jordon or Wilmot, Bailey to a lesser degree.

If one player is an outlier in terms of such anomalies, maybe it tells us something about their coach, as they are the consistent voter. But the Daicos case shows just how much a single polls-apart viewing can be. Getting 5-0 in the Carlton game last year completely skewed his discrepancy. I've never watched the game apart from the last 5 or 10 mins, but reading the match reports (AFL and ABC), and considering the Brownlow votes, if it was Voss that gave him zero it was a unique interpretation of his game.

Take that out and he has a discrepancy of 8 votes from 12 games, not worlds apart from Cripps's 6 from 13.

Yep I agree with that.

Finding outliers with known exact discrepancies would at least be a better method for correctly identifying interesting cases. We would still have to accept that in one off examples we don't know who voted which way, but yes a pattern may be identified across many games.

Trying to turn minimum discrepancy (with unknown exact figures) into a numerical assessment of coaches votes integrity, like MR has attempted, is simply without merit.

 

[sabso]

You're absolutely correct. But we can draw some conclusions about players that coaches more regularly agree about.

Josh Dunkley 5-4
Zac Bailey 5-3
James Jordon 4-0/2-2
Darcy Wilmot 4-0/2-2
Jarrod Berry 3-0
Dayne Zorko 1-1

Jarrod Berry's game was polarising, as was one of Jordon or Wilmot, Bailey to a lesser degree.

If one player is an outlier in terms of such anomalies, maybe it tells us something about their coach, as they are the consistent voter. But the Daicos case shows just how much a single polls-apart viewing can be. Getting 5-0 in the Carlton game last year completely skewed his discrepancy. I've never watched the game apart from the last 5 or 10 mins, but reading the match reports (AFL and ABC), and considering the Brownlow votes, if it was Voss that gave him zero it was a unique interpretation of his game.

Take that out and he has a discrepancy of 8 votes from 12 games, not worlds apart from Cripps's 6 from 13.

Possible breakdown (coach A-coach B)
Dunkley 5-4
Bailey 3-5
Jordon 4-0 (or 1-3)
Wilmot 1-3 (or 4-0)
Berry 2-1
Zorko 0-2

Jordon/Wilmot could have 0, 2 or 4 from this game factoring in the possibility either player got 2 from each.

 

[CharlieMortdecai]

Yep I agree with that.

Finding outliers with known exact discrepancies would at least be a better method for correctly identifying interesting cases. We would still have to accept that in one off examples we don't know who voted which way, but yes a pattern may be identified across many games.

Trying to turn minimum discrepancy (with unknown exact figures) into a numerical assessment of coaches votes integrity, like MR has attempted, is simply without merit.

That outlier I keep banging on about - Daicos's 5-0 in the carl v coll match from round 21 last year raises huge questions about Michael Voss's integrity.

What round do the stop publishing the coach's votes? Presumably it was Daicos, Heeney, Cripps as the top 3 at that stage.

So in a crucial game both Cripps and Daicos step up to the plate. General consensus seems to be Cripps BOG, Daicos next best.

McRae votes 5 Daicos, 4 Cripps. You know what? I forgive him that. He's allowed to come up with a perfectly valid, perhaps a little forced, justification to see it that way. Collingwood won after all.

But Voss goes 5 Cripps, 0 Daicos. A crucial late round game with Cripps and Daicos neck and neck. And Voss votes like that.

At the end of round 20 it was Cripps 86, Daicos 94. At the end of round 21 it was Cripps 95, Daicos 99.

Of course all the Blues are off in hiding at the moment understandably, but WalshistheGOAT is a top contributor to this thread. Do you remember the game? Can you justify (presumably) Voss giving Naicos nothing?

 

[CharlieMortdecai]

Possible breakdown (coach A-coach B)
Dunkley 5-4
Bailey 3-5
Jordon 4-0 (or 1-3)
Wilmot 1-3 (or 4-0)
Berry 2-1
Zorko 0-2

Jordon/Wilmot could have 0, 2 or 4 from this game factoring in the possibility either player got 2 from each.

You are quite right, I did it quickly and for some reason assumed Zorko must get both the 1s

 

[sabso]

You are quite right, I did it quickly and for some reason assumed Zorko must get both the 1s

Both score breakdowns are possible - there might be more but I cba working them out.

These are like those newspaper puzzles where you fill a square with numbers and each row and column has to add up to a number provided at the end of the line