We need to change the way we calculate percentage.

[lenny&carl]

We "need to change" about 1% of things society suggests these days. The richest, fairest, winningest society of all time...apparently should focus on changing every bloody thing. Well, howabout no?

As for AFL, it's a corporate-therefore-authoritarian entity...and maybe there are a decent amount of things need changing....but the people there are roughly good... Howabout we *suggest* improvements rather than get angry about "needed" change.

It's a sports entertainment product at the end of the day.

 

[Duckimus Prime]

I can't believe I bothered to read all this. Killed 5 minutes of the half time break I guess. If you want higher scoring games, lets just make goals worth 10 points and behinds worth 2 points. Easy.

Thats sounds like its had the same amount of thought put into it as a normal AFL rule change from Stephen Hocking.

You've successfully put forward a suggestion for higher scoring games that actually makes kicking accurately for goal less valuable.

 

[Seeds]

Still waiting on real-life data.

Why? this is a maths discussion. You don’t need a historical example. Thé evidence is in the mathematical proof which supersedes empirical data as a method for acquiring knowledge.

if you want to go through seasons worth of data and recalculate the percentages for each team using both the current and my new approach to try to prove it wouldn’t have much impact on that season then please go ahead. I don’t have a spare 4 hours to do so. Especially since it’s irrelevant as thé evidence is in the maths. We have already gone through theoretically plausible examples where it could easily impact a teams season by knocking them a spot or 2 down the ladder come finals.

 

[BF Tiger]

Why? this is a maths discussion. You don’t need a historical example. Thé evidence is in the mathematical proof which supersedes empirical data as a method for acquiring knowledge.

if you want to go through seasons worth of data and recalculate the percentages for each team using both the current and my new approach to try to prove it wouldn’t have much impact on that season then please go ahead. I don’t have a spare 4 hours to do so. Especially since it’s irrelevant as thé evidence is in the maths. We have already gone through theoretically plausible examples where it could easily impact a teams season by knocking them a spot or 2 down the ladder come finals.

Your maths might be correct but you still have to justify it.

I, and others I’m sure, would like you to show us a real-life example of where a more deserving team who is promoted via your method. For that to happen you have to first convince me that the team is more deserving as well as have the maths work out.

You’re examples are not plausible because (AFAIK) you’ve only considered how two separate results impact percentage when both teams have scored 1000 points for and 1000 points against, giving a percentage of 100%, as calculated currently. You need to show how the percentage would be altered by the individual match scores when calculated using your method. I highly doubt 1000 points for and against would give an average percentage of 100 if nine games were won 109-95 (114.7%) and one game lost 19-145 (12.6%)

You’re the one making the claim so the onus is on you to justify it. Until then you’ve failed.

 

[CharlieMortdecai]

Why? this is a maths discussion. You don’t need a historical example. Thé evidence is in the mathematical proof which supersedes empirical data as a method for acquiring knowledge.

if you want to go through seasons worth of data and recalculate the percentages for each team using both the current and my new approach to try to prove it wouldn’t have much impact on that season then please go ahead. I don’t have a spare 4 hours to do so. Especially since it’s irrelevant as thé evidence is in the maths. We have already gone through theoretically plausible examples where it could easily impact a teams season by knocking them a spot or 2 down the ladder come finals.

Lol, could you please point me to your post with the "mathematical proof".

 

[Seeds]

Lol, could you please point me to your post with the "mathematical proof".

There’s like 7 of them. I’ve posted the Equation. that’s all the proof that Is possibly needed.

 

[CharlieMortdecai]

There’s like 7 of them. I’ve posted the Equation. that’s all the proof that Is possibly needed.

That's your equation for an alternative calculation for percentage. It is using the SANFL method, but instead of using a cumulative total (total for/[total for + total against]), you use a mean of individual percentages calculated on that basis.
That is not a mathematical proof that the current system is flawed or unfair to low-scoring games. Where is your proof that your revised version is better?
That's why people are asking you to apply it to real life examples. For example, recalculate the percentages for the current season and point out why your system is better.
Or show us the mathematical proof that demonstrates the inequity of our current system.

 

[ItsAllAboutMe]

I have thought for a while, an easy tool the AFL could use to increase scoring is to get rid of % and just use total points scored as the tool to seperate teams on the ladder on equal points.

 

[megadeth86]

Or... crazy thought here... we could leave the game alone

That hasn't been done for about forty years - great idea though.

 

[Arkangel]

A key feature of all sports is a collection of fairly arbitrary rules that define success or failure. In the AFL, some of those arbitrary rules relate to the method of ordering teams on the ladder: First by their record of wins/draws/losses, then by percentage, defined as the points a team scores divided by the points scored against them.

By changing the method for calculating percentage, you would be swapping out one fairly arbitrary definition for another. There are possibly some decent arguments in favour of the calculation you propose, but these arguments are not mathematical: They're value judgements about what constitutes good performance in a football game, and what ought to be rewarded. Essentially, you appear to me to be saying, "I value this aspect of performance, so we should change the rules to correspond to my preference." That's fine, but if you want others to agree with you, then I think the argument you need to be making is about why that aspect of performance should be valued over others, rather than one that involves claims to some sort of objective superiority of your new and improved percentage calculation.

It seems to me that one way you might make that argument is to show that your method of calculating percentage is better at predicting success (e.g., in finals) than the existing calculation method. If so, that would seem to indicate that it captures some sort of latent "team strength" better. You would then need to explain why "team strength" ought to be rewarded, but I think most people would be fairly receptive to such a claim.

 

[Vindicater]

Make everyone play each other twice and we fix this bloody issue.

 

[SuperVolley11]

Right now we have the Cats and Port neck and neck on the ladder.

Cats 64pts 1768 1375 128.6
Port 64pts 1818 1428 127.3

Geelong are +393 on differential, Port are +390.

If we use OP 60 to 20 and 130 to 80 examples.
Cats 1828 1395 131.05 (+433)
Port 1948 1508 129.18 (+440)

In conclusion, defensive teams ARE rewarded with better %.