Banter Who will be better in 2024? Carlton or Collingwood? Part 2

[Arr0w]

See, this is why we don't work with averages...

The more games one plays, the more difficult it is to maintain an unusually high average.

This time last week Walsh was ahead of Daicos in 'average coaches votes'....

Coaches votes, free kicks to forwards, etc...

But the basis of you argument regarding goatie was that he lead the averages for votes in certain awards last year

"You just tie yourself in knots"

Oh boy

 

[Arr0w]

You guys really do spend a lot of time 'imagining', don't you?

I guess that's all you've has for more than two decades...

Who will be better in 2024? Carlton or Collingwood?

Or perhaps we can discuss 6 decades if you like, or 8 decades

 

[WalshistheGOAT]

You guys really do spend a lot of time 'imagining', don't you?

I guess that's all you've has for more than two decades...

Aren't you the same guy who's been squealing from the rooftops that Nicky is the best player in the comp and if he didn't get injured he would have won every award there was? That sounds like some serious 'imfadgening' to me

 

[Fadge]

This might be the most bizarre maths-related statement I have ever heard.

You have also proven why averages SHOULD be used with your second statement. Because with an increasing sample size the outlier results get smoothed out and you get a truer sense of the real result (i.e., the average).

Indeed, if you have a significant sample size for all data sets.

But not when we have small data sets, and one set of data is 25% to 50% more than other sets of data, and we're talking about 17 games compared to 12 games, or 9 games, or 5 games....

 

[Arr0w]

2024 is looking bleak for pie supporters, sitting 12th, goatie not leading every award, no pointy end draft picks and have the oldest list in the competition

Boy oh boy wowee

 

[Fadge]

Aren't you the same guy who's been squealing from the rooftops that Nicky is the best player in the comp and if he didn't get injured he would have won every award there was? That sounds like some serious 'imfadgening' to me

Indeed. When a player is leading both awards, and is only overtaken in the final round for both awards after missing the final three games, that is what one would regard as a reasonable data set.

As opposed to 17 games v. 12 games v. 9 games v. 5 games...

 

[carlton_99]

Indeed, if you have a significant sample size for all data sets.

But not when we have small data sets, and one set of data is 25% to 50% more than other sets of data, and we're talking about 17 games compared to 12 games, or 9 games, or 5 games....

Walsh has played 13 games and Daicos 17. In the context of an AFL season that is a significant sample size for both players. Totals in no way take into account differences in number of games between players - that's what averages are used for.

 

[Fadge]

Walsh has played 13 games and Daicos 17. In the context of an AFL season that is a significant sample size for both players. Totals in no way take into account differences in number of games between players - that's what averages are used for.

Nope.

Otherwise why would Walsh go from 'leading GOATCOS based on average votes' to 'behind GOATCOS based on average votes', in a week when neither player received votes?

 

[WalshistheGOAT]

2024 is looking bleak for pie supporters, sitting 12th, goatie not leading every award, no pointy end draft picks and have the oldest list in the competition

Boy oh boy wowee

Don't worry... They've still got Lachie Schultz

 

[WalshistheGOAT]

Wowee... Just looked at the Pies goal scorers for this year and they only have 1 player above 20 goals for the season... Didn't Fudgey always used to say that the Piebaggers were the best team in the comp because they had 12+ players that could kick 25 goals in a season??? WTF happened?

Curnow and Mckay have kicked more goals than their entire top 5 combined

 

[carlton_99]

Nope.

Otherwise why would Walsh go from 'leading GOATCOS based on average votes' to 'behind GOATCOS based on average votes', in a week when neither player received votes?

They did receive votes - 0 votes. 0 is an integer - you don't just ignore it. As Walsh has played less games his average will fluctuate more week to week. Both averages drop with 0 votes but because Walsh has a smaller sample he drops by more.

 

[Arr0w]

Don't worry... They've still got Lachie Schultz

A 3rd rounder, "Elijah, D grader in a C grade side"

While the Pies are going B2B, Schultz a great acquisition for a 1st and 2nd

Oh boy

 

[jezzajay]

Oh no.

I've done some 'analysis' on this famous table that Carlton supporters are referencing to 'prove' they've had the hardest draw in 2024....

For all the teams who played in Opening Round, their first two matches didn't happen.

So Carlton's games against Brisbane (who couldn't beat anyone at the time) and likely wooden spooners Richmond didn't contribute to the calculation.

Likewise, Collingwood's games against GWS in Sydney, and likely minor premiers Sydney, didn't contribute to the calculation. Both teams were ranked #1 for Carlton when they played them a few weeks later.

In further proof of the amateur nature of the 'data', the calculations are actually WRONG....

The table appears to use a divisor of 16 for all teams, on the basis each team has 16 matches contributing to the calculation. However, the teams that played in Opening Round all have an additional bye, meaning the divisor for those teams should be 15.

Boy oh boy wowee.

 

[Arr0w]

Pies need to win 5 of their last 6 to guarantee finals, win 4 and percentage will be crucial

 

[WalshistheGOAT]

Pies need to win 5 of their last 6 to guarantee finals, win 4 and percentage will be crucial

Hows their % looking?

 

[Arr0w]

Hows their % looking?

Sitting in the 8th to 11th range

 

[jezzajay]

How many free's has old mate Charles even gotten this season?

Mihocek and Cox average more frees than Charlie…
Hilarious..

 

[jezzajay]

I need to read through the thread before posting hard facts…I see some have already got it covered.
Facts are…the stats don’t lie…neither do the opposition salty melts.

 

[Arr0w]

 

[JustaBattler]

Fudgeymatics in full swing tonight -wot have I missed!!!

and where is his pussy mate to back his fudgeymatics up?

LOL comedy capers.

 

[Worm4]

Watch a certain poster put his personal spin on this graph , and tell us where it's all wrong.....

 

[Fadge]

Watch a certain poster put his personal spin on this graph , and tell us where it's all wrong.....

I love how you guys are very happy to repost 'analysis' from random twitter posters without first reviewing and performing due diligence on the data you're share.

I suspect this is because you don't have the intellectual capacity to perform the analysis yourselves, or assess the accuracy of the data you are resharing.

First there was the table calculating 'draw difficulty', that ignored more than 10% of the data (that has the potential to have a significant effect on the calculation - refer to my comments about Carlton and Collingwood's first two games), and then used the incorrect divisor for teams who played in the Opening Round.

So naturally there were going to be flaws in the table reshared by CArr0w, given it would have been shared without any thought.

The 'probabilities' provided for whether teams win or lose this week should not be a single figure... instead there should be a range.

Obviously, if Carlton win this week, their top 4 chances improve. But if Brisbane, Geelong, Fremantle, Essendon, Port Adelaide and GWS all win, their top 4 chances don't increase as much compared to if those 5 teams lose. And if some of the aforementioned teams win and others lose, the rate of increase falls somewhere in between...

Happy to school you all. Again.

 

[carlton_99]

I love how you guys are very happy to repost 'analysis' from random twitter posters without first reviewing and performing due diligence on the data you're share.

I suspect this is because you don't have the intellectual capacity to perform the analysis yourselves, or assess the accuracy of the data you are resharing.

First there was the table calculating 'draw difficulty', that ignored more than 10% of the data (that has the potential to have a significant effect on the calculation - refer to my comments about Carlton and Collingwood's first two games), and then used the incorrect divisor for teams who played in the Opening Round.

So naturally there were going to be flaws in the table reshared by CArr0w, given it would have been shared without any thought.

The 'probabilities' provided for whether teams win or lose this week should not be a single figure... instead there should be a range.

Obviously, if Carlton win this week, their top 4 chances improve. But if Brisbane, Geelong, Fremantle, Essendon, Port Adelaide and GWS all win, their top 4 chances don't increase as much compared to if those 5 teams lose. And if some of the aforementioned teams win and others lose, the rate of increase falls somewhere in between...

Happy to school you all. Again.

You do know what "~" means right?

 

[Robertio]

I love how you guys are very happy to repost 'analysis' from random twitter posters without first reviewing and performing due diligence on the data you're share.

I suspect this is because you don't have the intellectual capacity to perform the analysis yourselves, or assess the accuracy of the data you are resharing.

First there was the table calculating 'draw difficulty', that ignored more than 10% of the data (that has the potential to have a significant effect on the calculation - refer to my comments about Carlton and Collingwood's first two games), and then used the incorrect divisor for teams who played in the Opening Round.

So naturally there were going to be flaws in the table reshared by CArr0w, given it would have been shared without any thought.

The 'probabilities' provided for whether teams win or lose this week should not be a single figure... instead there should be a range.

Obviously, if Carlton win this week, their top 4 chances improve. But if Brisbane, Geelong, Fremantle, Essendon, Port Adelaide and GWS all win, their top 4 chances don't increase as much compared to if those 5 teams lose. And if some of the aforementioned teams win and others lose, the rate of increase falls somewhere in between...

Happy to school you all. Again.

Also if all the above teams win and Collingwood loses you can plan your September holidays.

 

[Arr0w]

I love how you guys are very happy to repost 'analysis' from random twitter posters without first reviewing and performing due diligence on the data you're share.

I suspect this is because you don't have the intellectual capacity to perform the analysis yourselves, or assess the accuracy of the data you are resharing.

First there was the table calculating 'draw difficulty', that ignored more than 10% of the data (that has the potential to have a significant effect on the calculation - refer to my comments about Carlton and Collingwood's first two games), and then used the incorrect divisor for teams who played in the Opening Round.

So naturally there were going to be flaws in the table reshared by CArr0w, given it would have been shared without any thought.

The 'probabilities' provided for whether teams win or lose this week should not be a single figure... instead there should be a range.

Obviously, if Carlton win this week, their top 4 chances improve. But if Brisbane, Geelong, Fremantle, Essendon, Port Adelaide and GWS all win, their top 4 chances don't increase as much compared to if those 5 teams lose. And if some of the aforementioned teams win and others lose, the rate of increase falls somewhere in between...

Happy to school you all. Again.

And again you miss some vital components in relation to this post

You ain't schooling anyone here