Two Final-9's. Which do you prefer?

[Dan26]

In the current system, IMO, the second week of the finals tends to have the least interest.

Rubbish, Chaz, and you know it.

The finals become more interesting, and more important every advancing week.

Both these matches are knockouts but they have less significance than the "non-knockout" qualifying finals.

Why would you write utter crap. Just for a response?

The winner of the Qualifying Final goes to a Prelim. The loser gets another chance.

The semi-final in week two has the same reward for the winner (that being a trip to the Prelim) but the loser is out.

So, how in the hell can the week-two semi-finals have less significance than the Qualifying finals, when the rewards for winning are the same, the but the penalty for losing is far greater in the semi-final?

Here's how much more important the semi-finals are than the Qualfying Finals:

Semi-Finals
- Winner advances to the Prelim where they have a 25% chance of winning the premiership
- Loser has a 0% chance of winning the premiership (because they're eliminated)
25% difference between winning and losing


Qualifying Finals
- Winner advances to the Prelim where they have a 25% chance of winning the premiership.
- Loser goes to semi-final where they have a 12.5% chance of winning the flag.
12.5% difference between winning and losing





Do you Chaz, REALLY think that the 12.5% difference between winning and losing is more important than the 25% difference? You obviously have done no reasearch or you don't know the finals system very well, if you think the Qualifying final is more important than the semi-final, because it quite clearly is not. And that's a fact. You can't argue that. It's simple maths

Kinda backfired on you there didn't it? Oh well, shit happens.

 

[Dan26]

Ok numnut
The final 6 was stupid
Your final 9's are even stupidier

Stupidier? Ok, you need to shut the hell up. You waltz in here with no idea what youre talking about and lecture me about something I clearly know far more about than you. I can tell you the probabilties for every finals system ever been used. I can tell you how to construct double chance systems, pure knockout system, with any number of teams odd or even. I used to write letters with ideas to the AFL about this as far back as 1991.

Not all of the systems are desirable, not all would be used.

I never said any of the three knockout finals-9s I constructed should be used. But I made the effort of constructing them? What did YOU do "numnuts." All three finals-9's are mathemtically fair.

You did S.F.A. You sit there and pass judgement from a distance aboout something you clearly don't know anywhere near as much as I, without coming up with one idea of your own.

At the moment any of the top four teams have to win 3 games in a row to win the premiership.

5-8 have to win 4 games in a row

Thankyou captan obvious. How would I have know this without you telling me?

Hey, because you posted that I can now sleep safely with the knowledge that 5,6,7,8 under the current final-8 all need to win 4 games. How would I have known without your unique observation of the current finals system?

Oh and by the may you have "developed" a finals system with two games less then the current one with an extra team with your knockout crap, i'm sure the AFL will love it.

Knockout crap? Don't you know what happens to the loser of the Grand Final and Preliminary Finals under the current system? They are knockout you fool. Do I really need to explain that to you?

Also, none of the 3 final-9 systems have two less games than the current final-8. Not one. Get it right, or don't post.

All of them are ideas. Ideas are what smart people come up with. Surprise me and come up with one of your own. Even just one idea.

My ideology for finals is that they are about performing on the day. That's what they should be about. i.e they should always be knockout.

I agree and I am fully aware that the current final-8 is mathemtically fair. I know exactly how it works you idiot, and I know that it is "fair."

Yes, the current final-8 is a fair system, but it's a fair system that goes against the ideology of what finals should be about. YOU yourself (along with 99.9% of the population) enjoy the drama and suspense of knockout finals. We all do. We all love a match where the season is on the line.

We should do it as the NFL does. A proper knockout system that still gives advantages to the higher seeded teams (via a week off.) We should do it in a similarly structured way.

I think we will eventually have a knockout final-10 when we have an 18-team comp, with 9 finals over 4 weeks. 7v10 and 8v9 in the first week, with the winners progressing to the second week.

It's funny how people seem to like (or meekly accept) the system that they have grown up with. People like the "safe" option of what they're used to, instead of something new and better.

I wonder if we had used pure knockout for 100 years if you or Chaz or anyone else would be campaigining for a new "double chance system." I highly doubt it. You'd probably love the knockout system that had been used for 100 years.

Its true isn't it? You woudn't want it changed in that situation, would you.

 

[Perry Pie]

Dan your final 9's all have a major flaw, that is what happens in other games can impact on on other teams progression. (in example 3 above if 9th beats first it changes what happens to the winners of 7th & 8th).
Under the current system all progression is determined by the results of the teams playing, destiny is in the hands of the teams playing, thats what makes finals different. What happens in other games does not effect what game the two teams playing any final play next.
The current system is fantastic. We dont need a 9 or 10 team final system. The current 8 will work better in an 18 or 20 team competition.

 

[jpdixon]

The finals become more interesting, and more important every advancing week.

The winner of the Qualifying Final goes to a Prelim. The loser gets another chance.

The semi-final in week two has the same reward for the winner (that being a trip to the Prelim) but the loser is out.

So, how in the hell can the week-two semi-finals have less significance than the Qualifying finals, when the rewards for winning are the same, the but the penalty for losing is far greater in the semi-final?

That's a big weird. It's more of a crunch game as there is more on the line to be lost, sure, but how is a week off and a home final the same reward as an away final against a team who has had a week off? You seem to switch between considering time off and home games as reasonable rewards and the crude 50-50 probability model at very strange times.

But that's really beside the point, because Chaz said less interest, not less signficance. I guess we could talk about different sorts of significance, but it's certainly not a ridiculous claim that the 2nd week tends to draw less interest. In September, most interest goes to the teams considered serious GF contenders, and it's not that uncommon for the two (rightly or wrongly) "obvious" contenders to be resting that week. I'm sure there's a decent number of people who paid more attention to G and StK's qualifying finals form, than who got knocked out of the race the next week.

 

[Cthublhu]

The only people who should be pushing for a Final 9 are Richmond. Even then, they'd likely finish 10th.

 

[Dan26]

That's a big weird. It's more of a crunch game as there is more on the line to be lost, sure, but how is a week off and a home final the same reward as an away final against a team who has had a week off?

The winners of both the semi-final and the Qualifying final go to the Prelim where they have a 25% chance of winning the premiership.

I take your point about the winner of the semi-final being the "away" team the next week, but that doesn't make it any less important. It's still a Prelim they are advancing to, and it's still knockout if you lose.

The winner of the Qualifying Final also advances to the Prelim (the 25% chance stage) but the loser is not eliminated. It's clearly not as important to win the Qualifying Final as Sydney (in 2005) and West Coast (in 2006), and Brisbane (in 2003) proved.

I'd say Adelaide in 2009 would testify that it IS kinda important to win the week-2 semi-final, don't you think?

So, no matter how you cruch the numbers the second week of the finals is more important, interesting and significant than the Qualifying Finals.

Every advancing week is more significant and important than the week before. That's a simple unarguable fact thats true of any play-off system anywhere in the world.

Where's Chaz anyway? lol.. typically M.I.A.

Can't say I blame the poor bugger.

 

[Dan26]

Dan your final 9's all have a major flaw, that is what happens in other games can impact on on other teams progression.

That's actually not a "flaw." That's just the way the systems work. A certain number of teams are guaranteed to progress straight to the Prelim finals under two of the finals 9's. If any of the top teams lose, then logically (and FAIRLY) the next highest winner takes their place.

So, under the second final-9 system, where three teams advance to the prelims after the first week, if, say 1st lose to 9th (and 2nd and 3rd win), then the winner of 5th and 6th will be the next highest placed winner. So they advance. And fair enough.

The 4th placed team starts their campaign in week two and would need to win one game to get to the Prelims, just like 1,2, and 3.

So, under that situation it is perfectly fair for the winner of 5th and 6th to advance to the Prelims IF one of the top 3 don't take their opportunity. That's not a "flaw." It rewards the "next highest winner" if the top teams fail.

You don't have to use a system like that. But it's not "unfair" to use it. And it works. It's just a different way of running a knockout finals series. The probabilites are all fair.

Think of it this way:

There are 4 matches in week one. Three winners go straight to the Prelims. If 1,2,3 do their job they will be those three teams. If one (or more) of them loses, the next highest winner get that reward and so on. There is no "fault" in that - it's just a different way to do it. It's perfectly fair.

There are all sorts of fair finals systems around (the current final-8 being one.) It's more a matter of finding the one that best suits the ideology of what finals should be about. And that's knockout.

The final-10 will be the one we will have IMO when we go to 18-team league. But all of my final 9's work, and are mathematically fair.

 

[Tarkyn_24]

Do you Chaz, REALLY think that the 12.5% difference between winning and losing is more important than the 25% difference? You obviously have done no reasearch or you don't know the finals system very well, if you think the Qualifying final is more important than the semi-final, because it quite clearly is not. And that's a fact. You can't argue that. It's simple maths

Kinda backfired on you there didn't it? Oh well, shit happens.

I think he's referring to the fact that QF finals tend to be better spectatcles as a viewer. I can think of numerous brilliant QFs that have stuck with me, but not so with the SFs

 

[mcgarnacle]

im also of the opinion that all post-season matches must be playoffs, but for it to work properly the top seeds must be rewarded with the first week bye and home advantage as an incentive for an easier progression throughout the series. im also of the opinion that both teams in a playoff must play-off for the same rewards, i.e., progression to the following week.

problem is, it works well only as a 6-team (3 week) series, or a 12-team (4 week) series, and neither is a realistic option in an 18 team comp. 6 teams is too few in a league where the regular season is an unfair and inequitable manipulated farce in which teams play different teams twice, yet are compared to each other in a single ladder. on the flipside, rewarding a 10th, 11th & 12th seed (and losing records) with a post-season berth is overkill in an 18-team league.

assuming higher seeds win..........
6 teams
Wk 1 - Semi Finals
a - 3 seed vs 6th seed - WB vs BRI
b - 4 seed vs 5th seed - COL vs ADE

Wk 2 - Preliminary Finals
c - 1 seed vs Lowest-seeded winner [a/b] - STK vs COL
d - 2 seed vs Highest-seeded winner [a/b] - GEE vs WB

Wk 3 - Grand Final
e - Winner c vs Winner d - STK vs GEE

12 teams
Wk 1 - Elimination Finals
a - 5 seed vs 12 seed - ADE vs SYD
b - 6 seed vs 11 seed - BRI vs WC
c - 7 seed vs 10 seed - CAR vs PA
d - 8 seed vs 9 seed - ESS vs HAW

Wk 2 - Quarter Finals
e - 1 seed vs Lowest-seeded winner [a/b/c/d] - STK vs ESS
f - 2 seed vs 2nd lowest-seeded winner [a/b/c/d] - GEE vs CAR
g - 3 seed vs 2nd highest-seeded winner [a/b/c/d] - WB v BRI
h - 4 seed vs Highest-seeded winner [a/b/c/d] - COL vs ADE

Wk 3 - Preliminary Finals
i - Highest-seeded winner [e/f/g/h] vs Lowest seeded winner [e/f/g/h] - STK vs COL
j - 2nd highest-seeded winner [e/f/g/h] vs 2nd lowest-seeded winner [e/f/g/h] - GEE vs WB

Wk 4 - Grand Final
k - Winner i vs Winner j - STK vs GEE

 

[Chaz Pendlebury]

I think he's referring to the fact that QF finals tend to be better spectatcles as a viewer. I can think of numerous brilliant QFs that have stuck with me, but not so with the SFs

Cheers Tarkyn_24, that's part of it. Unfortunately Rainman26 struggles to think deeper than the most basic statistics.

Qualifying finals are between two premiership contenders (at least they, and their supporters, both think they are before the match).

Semifinals are between "outsiders" (from the bottom half of the 8) and the "premiership contender" who just had at least some proportion of the wind knocked out of its sales the week before.

Occasionally a team from outside of the 4 might be a contender, (Adelaide 2009), but, if my memory is correct, only two none top four teams have won semis in the last 10 years, both beaten the following week (by not much admittedly)

Teams that have won their QFs have gone on to make the grand final 17 of 20 times (I'm pretty sure that is correct). Using "mathematical" chances, completely ignoring the effects of the week off, home final, confidence etc, is next to meaningless.

Using "empirics" your chance of making a GF if you win the QF goes to 85% while losing it drops to 15%. (Winning the premiership, A winning QF takes you to 35% while the loser drops to 15%). Not perfect but that is using 10 years of empirical reality as a sample, rather than probabilities with all the detail sucked out of them.

QFs generally have more interest, draw bigger crowds etc because its between two premiership contenders. People know you are a giant leap closer if you win and its a long haul back if you lose

I.e For most people, not rainman26, to whom non-knockout finals are like flying, flaring up his aspergers, the non-knockout nature of QFs does not reduce them below SFs or EFs

 

[Chaz Pendlebury]

Where's Chaz anyway? lol.. typically M.I.A.

Can't say I blame the poor bugger.

Hahaha, I just read this....only one poor bugger 'round here....

I can guarantee, if the "winner" of this thread is whoever posts last, than you my sad little Rainman26 will be victorious....

 

[McCrann]

Yes they are. They are deplorable.

Mate - Double chances for the best performing teams through the year are a feature of the VFL/AFL for over a Century.

Our Finals have been going on longer than the NFL's for instance - far longer. We've had double chances since at least 1897 - probably before.

Why do you want to junk 110+ years of history to give us a more Americanised system - when in this case at least - it is completely unwarranted!

Where is the evidence through gate takings, TV viewers etc. that people don't like to see the best teams rewarded after a year of slog?

Where is the evidence?

I certainly don't see any and no reason to just blindly follow an American sporting history rather than our own.

 

[mcgarnacle]

The finals become more interesting, and more important every advancing week.

this should be the case in theory. and it would under a playoff format throughout an entire post-season where the top seeds enter the fray after a 1st week bye.

So, how in the hell can the week-two semi-finals have less significance than the Qualifying finals, when the rewards for winning are the same, the but the penalty for losing is far greater in the semi-final?

the fact that two of the major (and based on the stats, probable) competitors for the premiership are absent from the week 2 matches under the AFL's final 8 lessens the appeal of the current system when measured by the terms of your first quote above. Week 2's divisional round playoffs in the NFL are regarded as the 'best weekend of football' in the USA, and the reason for this is because each game may feature the possible Superbowl combatants over 2 days of double-headers. Week 1's wildcard playoffs operate under a similar format, but are missing the top 2 seeds from each conference because they have been rewarded with the bye.

the 1st week in the AFL is comparable to the 2nd week in the NFL, but the 2nd week in the AFL loses momentum and is actually comparable with the 1st week of the NFL.

 

[Dan26]

Where is the evidence through gate takings, TV viewers etc. that people don't like to see the best teams rewarded after a year of slog?

I'm all for rewarding the best teams. Why in the hell do you think that the use of a double chance is the ONLY way to reward a team? Havn't you read anything in this thread at all? People like to see the best teams rewarded. So reward those teams! Just not with a double chance.

Mate - Double chances for the best performing teams through the year are a feature of the VFL/AFL for over a Century.

Our Finals have been going on longer than the NFL's for instance - far longer. We've had double chances since at least 1897 - probably before.

Why do you want to junk 110+ years of history to give us a more Americanised system - when in this case at least - it is completely unwarranted!

Where is the evidence through gate takings, TV viewers etc. that people don't like to see the best teams rewarded after a year of slog?

Where is the evidence?

I certainly don't see any and no reason to just blindly follow an American sporting history rather than our own.

You can't be serious?

You're argument basically is, "We've been doing it this way for X years so, why change?"

So, if we'd used a pure knockout system for 100 years would you want that to stay too? You would. I'm sure you would. You just don't want change (typical of lots of people unfortunately.)

Take a step back, imagine you are inventing a finals system and you have no knoweldge of any finals system that has ever been invented. Do you honestly think YOU would invent the current double chance final-8?

Do you honestly think that we in Australia (the only country that uses such a system) just happen to have got it right? Are you that arrogant to suggest that our way is the right way? And the rest of the world does it in a worse way?

Or maybe, just maybe you are used to the double chance system. You've never known any other way, and therefore you are afraid of any other better system. You feel comfortable with it because it's all you know.

Remember this is NOT an argument about fairness. And it's not an argument about what works. The current final-8 is fair. And it works. As do pure knockout systems. They all work.

So I don't want to hear crap like "The current system is fair and has served us well, and it works so why change?"

Well of course it's fair! And of course it works! DUH! It's not about that. Knockout systems are fair and work too.

It's about having a system that suits the ideology of what finals are about. Finals are about performing on the day - they are not about getting second chances. I don't know how that can even be argued as a basic finals ideology.

 

[Dan26]

this should be the case in theory. and it would under a playoff format throughout an entire post-season where the top seeds enter the fray after a 1st week bye.



the fact that two of the major (and based on the stats, probable) competitors for the premiership are absent from the week 2 matches under the AFL's final 8 lessens the appeal of the current system when measured by the terms of your first quote above. Week 2's divisional round playoffs in the NFL are regarded as the 'best weekend of football' in the USA, and the reason for this is because each game may feature the possible Superbowl combatants over 2 days of double-headers. Week 1's wildcard playoffs operate under a similar format, but are missing the top 2 seeds from each conference because they have been rewarded with the bye.

the 1st week in the AFL is comparable to the 2nd week in the NFL, but the 2nd week in the AFL loses momentum and is actually comparable with the 1st week of the NFL.

To reward teams mathematically in a knockout system, you need something other than 2,4,8 or 16 teams.

The 6 team knockout series you mentioned is what the NFL uses, as you stated.

That's why I think we will have a final-10 with the 18 team league because it allows the top 6 to get a week off, but the top TWO still have an advantage because they play the winners of 7v10 and 8v9, while 3v6 and 4v5 play in week two with all those 4 teams having the same advantage (i.e they all have a week off)

That's what will eventually happen, I reckon.

There is another way to conduct the NFL's knockout top 6 (top12) by the way which is not necessarily right or wrong its just another way of doing it, and it features all the teams in week one.

ALTERNATE NFL TOP 6 for NFC
Week 1
1v6, 2v5, 3v4
highest seeded winner to NFC championship game in week 3

Week 2
the two lowest winners from week one play off
winner to NFC championship game

Week 3
Highest winner from week one vs winner of week two match
winner to Superbowl

Probabilities:
1st - 25%
2nd - 18.75%
3rd - 15.625
4th - 15.625
5th - 12.5%
6th - 12.5%

- 1st is guaranteed to go through to the title game if they win
- 2nd needs one other result to happen (1st losing) to go straight to the title game
- the winner of 3rd and 4th need two other results to happen (1st and 2nd losing) to go straight to the title game
- 5th and 6th cannot go straight to the title game.

It's the same number of games as what they use now, but is technically mathematically fairer. The current system has 1st and 2nd with a 25% chance and 3,4,5,6 all with a 12.5% chance.

The top seed actually has a genuine advantage under the alternate system.

 

[Fadge]

Stupidier? Ok, you need to shut the hell up. You waltz in here with no idea what youre talking about and lecture me about something I clearly know far more about than you. I can tell you the probabilties for every finals system ever been used. I can tell you how to construct double chance systems, pure knockout system, with any number of teams odd or even. I used to write letters with ideas to the AFL about this as far back as 1991.

Not all of the systems are desirable, not all would be used.

I never said any of the three knockout finals-9s I constructed should be used. But I made the effort of constructing them? What did YOU do "numnuts." All three finals-9's are mathemtically fair.

You did S.F.A. You sit there and pass judgement from a distance aboout something you clearly don't know anywhere near as much as I, without coming up with one idea of your own.

Nah, for me DIDAICOS was right the first time - he knows more about finals systems than you, and he hasn't even needed to post much to prove that. So you've been writing letters to the AFL since 1991? I'm tipping you've received more support on this forum (which is very little) than you have from the AFL inner sanctum.

In fact, now that I re-read it, you say you 'used to write letters'. Why did you stop? Let me guess - you didn't get anywhere? But I guess that would just make them narrow-minded dimwits who simply do not understand your greater intellect. Just like 98% of us on this forum.

For the record, I'm another who has a greater level of interest in the Qualifying Finals than the Semi. I thought Collingwood were a much better chance for the Premiership in 2009 before the Qualifying-Final against St. Kilda, than I did after the Semi-Final against Adelaide... Though now you'll indicate that I'm wrong, because they were mathematically a 18.75% chance before the Qualifying Final and improved to a 25% chance after the Semi!

 

[Dan26]

I thought Collingwood were a much better chance for the Premiership in 2009 before the Qualifying-Final against St. Kilda, than I did after the Semi-Final against Adelaide

Well if that's the case, you're an idiot.

AFTER the semi-final, Collingwood was IN the Preliminary Final. They had to win 2 games to win the flag.

BEFORE the Qualifying final, Collingwood still needed to win 3 finals games.

So, you'd have preferred Collingwood to start their campaign in week one rather than be guaranteed to already be in the Preliminary Final (which is where they were after the semi-final.) You sir truly are a fool.

You'd take a Qualifying final over a guaranteed Preliminary Final.

It's clear you are sticking up for your fellow dimwitted supporter, through some sort of mutual, "he barracks for Collingwood too so I'm going to support him through any argument no matter how stupid" fellowship.

You need to get over that "sticking up for each other" crap and focus on the argument. No need to agree with someone if what they say is crap just because he barracks for the same team as you.

The reality is that you've just made a fool of yourself. If you're going to use any sort of argument relating to finals systems make sure you double check it and do your research because I'll tear you to shreds again and again buddy.

 

[Fadge]

For the record, I'm another who has a greater level of interest in the Qualifying Finals than the Semi. I thought Collingwood were a much better chance for the Premiership in 2009 before the Qualifying-Final against St. Kilda, than I did after the Semi-Final against Adelaide... Though now you'll indicate that I'm wrong, because they were mathematically a 18.75% chance before the Qualifying Final and improved to a 25% chance after the Semi!

Well if that's the case, you're an idiot.

AFTER the semi-final, Collingwood was IN the Preliminary Final. They had to win 2 games to win the flag.

BEFORE the Qualifying final, Collingwood still needed to win 3 finals games.

So, you'd have preferred Collingwood to start their campaign in week one rather than be guaranteed to already be in the Preliminary Final (which is where they were after the semi-final.) You sir truly are a fool.

You'd take a Qualifying final over a guaranteed Preliminary Final.

It's clear you are sticking up for your fellow dimwitted supporter, through some sort of mutual, "he barracks for Collingwood too so I'm going to support him through any argument no matter how stupid" fellowship.

You need to get over that "sticking up for each other" crap and focus on the argument. No need to agree with someone if what they say is crap just because he barracks for the same team as you.

The reality is that you've just made a fool of yourself. If you're going to use any sort of argument relating to finals systems make sure you double check it and do your research because I'll tear you to shreds again and again buddy.

Yep, Dan26, you tore me to shreds. So much so that I actually predicted it.

I didn't expect you to understand my logic.

And for the record, a Qualifying Final win is a guarantee of a Preliminary Final berth. But I guess you knew that, right?

 

[skipper kelly]

Final 8.

 

[Falchoon]

I thought Collingwood were a much better chance for the Premiership in 2009 before the Qualifying-Final against St. Kilda, than I did after the Semi-Final against Adelaide...

if you can't rationalise it with percentages you're wrong.

as for resumes, when I was 12 years old I designed a 24 team home and away season for my footy cards, & I think 8/18 is just right

 

[mcgarnacle]

To reward teams mathematically in a knockout system, you need something other than 2,4,8 or 16 teams.

The 6 team knockout series you mentioned is what the NFL uses, as you stated.

i agree with this, otherwise it'd be just a straight tournament-style system all the way through which doesnt do well to reward top seeds (in addition to home advantage which itself can be nullified if seed 1 & seed 8 are vic clubs) after 5 months of regular season competition.

That's why I think we will have a final-10 with the 18 team league because it allows the top 6 to get a week off, but the top TWO still have an advantage because they play the winners of 7v10 and 8v9, while 3v6 and 4v5 play in week two with all those 4 teams having the same advantage (i.e they all have a week off)

That's what will eventually happen, I reckon.

i would hope this isnt the case. in terms of progression to the next phase, there is no difference between seed 1 and seed 6. rewarding with teams 5 and 6 with wk 1 bye is stretching it.

There is another way to conduct the NFL's knockout top 6 (top12) by the way which is not necessarily right or wrong its just another way of doing it, and it features all the teams in week one.

ALTERNATE NFL TOP 6 for NFC
Week 1
1v6, 2v5, 3v4
highest seeded winner to NFC championship game in week 3

Week 2
the two lowest winners from week one play off
winner to NFC championship game

Week 3
Highest winner from week one vs winner of week two match
winner to Superbowl

Probabilities:
1st - 25%
2nd - 18.75%
3rd - 15.625
4th - 15.625
5th - 12.5%
6th - 12.5%

- 1st is guaranteed to go through to the title game if they win
- 2nd needs one other result to happen (1st losing) to go straight to the title game
- the winner of 3rd and 4th need two other results to happen (1st and 2nd losing) to go straight to the title game
- 5th and 6th cannot go straight to the title game.

It's the same number of games as what they use now, but is technically mathematically fairer. The current system has 1st and 2nd with a 25% chance and 3,4,5,6 all with a 12.5% chance.

The top seed actually has a genuine advantage under the alternate system.

i personally dont agree with the differences of rewards on offer in each game (eg seed 1 vs seed 6) and fortunes determined by results in other games. it's a bit like the mcintyre 8 which i loathe for this reason. furthermore, a series such as this - as is demonstrated by the old NRL/current AFL final 8, and the McIntyre 8 - loses momentum in the 2nd week when top teams have a bye at this stage.

im definately a supporter of the playoff system right through a post-season, but i prefer the balance of introducing the top seeds at the 2nd phase, and ensuring only the top proportion of the post-season qualifiers are rewarded with this. that's why i reckon the NFL system is perfect.

 

[FootyJunkie]

Here's a Final 9 System I've come up with. It's not perfect, but it doesn't have dead rubbers, runs 4 weeks, and there are 9 finals, what the AFL would look for.

FINAL 9

Week 1

1QF - 2nd v 5th
2QF - 3rd v 4th
1EF - 6th v 9th
2EF - 7th v 8th

Week 2

1SF - 1st v Winner 2EF
2SF - Higher QF Loser (2nd, 3rd, or 4th) v Winner 1EF

Week 3

If 1st wins 1SF...

1PF - 1st v Winner 2SF
2PF - Winner 1QF v Winner 2QF

Otherwise...

1PF - Higher QF Winner v Winner 2SF
2PF - Lower QF Winner v Winner 1SF

This is to prevent a repeat match-up from Week 1 in case the Higher QF Loser makes it to the Preliminary Finals.

Week 4

GF - 1PF Winner v 2PF Winner @ MCG

Notes:

This system is similar to the Final 8 current system except that 1st replaces the lower of the 2 Qualifying Final losers, while that team is eliminated. If higher seeded teams win, the team that 1st replaces would be 5th.

1st gets Week 1 off, then gets handed a relatively easy opponent in Week 2, 7th or 8th. This is a tradeoff for 1st losing the double chance in this system which is only applicable to one of 2nd, 3rd or 4th... the higher of the 2 Qualifying Final losers.

5th still have a cutthroat situation in Week 1... they have a tougher opponent in 2nd and are the "away" team, but the upside is that if they win, they get Week 2 off.

I suppose one beef with this system would be that 3rd could be gone after Week 1 while 2 teams below them are guaranteed to play on in Week 2.

It's the best I could do.

Using the 2009 Ladder...

Week 1

Geelong v Adelaide - MCG
Western Bulldogs vs Collingwood - MCG
Brisbane v Hawthorn - GABBA
Carlton v Essendon - MCG

Week 2

St.Kilda v Essendon - MCG
Collingwood v Hawthorn - MCG

Week 3

St.Kilda v Hawthorn - MCG
Geelong v Western Bulldogs - MCG

Week 4

Geelong v Hawthorn - MCG

 

[dymmot]

With the 18-team league on the way, which system do you prefer? Both are total knockout as finals need to be. Because both are knockout they are both ridiculously simple.

FINAL NINE NUMBER ONE (Brackets indicate match-ups if higher teams win. Bold are the winners)

WEEK 1
Elimination Final: 4th vs 9th
Elimination Final: 5th vs 8th
Elimination Final: 6th vs 7th


WEEK 2
1st Semi Final: 1st vs lowest seeded Elimination Final winner (1v6)
2nd Semi Final: 2nd vs second-lowest seeded Elimination Final winner (2v5)
3rd Semi Final: 3rd vs highest seeded Elimination Final winner (3v4)

(Highest seeded winner straight to Grand Final)

WEEK 3
Preliminary Final: The two lowest placed Semi-Final winners play-off. (2v3)

WEEK 4
Grand Final: Highest seeded winner from week 2 vs winner of Preliminary Final (1v2)

Probabilities of winning the premiership (based on all matches being 50-50) under that final-9 system

1st - 25.00%
2nd - 18.75%
3rd - 15.625%
4th - 7.8125%
5th - 7.03125
6th - 6.640625%
7th - 6.640625%
8th - 6.25%
9th - 6.25%

Total knockout, as finals should be. Yet, it still gives a good advantage to the minor premier, and progressively to the other teams. The total elimination of double chances should be a priority.

.
.
.
.
.
FINAL NINE NUMBER TWO (Brackets indicate match-ups if higher teams win. Bold are the winners)

WEEK 1
Elimination Final: 1st vs 9th
Elimination Final: 2nd vs 8th
Elimination Final: 3rd vs 7th
Elimination Final: 5th vs 6th
(4th has a week off)
The three highest placed winners advance to week 3


WEEK 2
Semi Final: 4th vs lowest seeded Elimination Final winner (4v5)
Winner to week 3


WEEK 3
Preliminary Final: Winner of Semi-Final vs highest-placed week one winner (1v4)
Preliminary Final: The second and third highest winners from week one play-off. (2v3)

WEEK 4
Grand Final: Winner of the two Preliminary Finals (1v2)

Probabilities of winning the premiership (based on all matches being 50-50) under that final-9 system

1st - 12.5%
2nd - 12.5%
3rd - 12.5%
4th - 12.5%
5th - 11.72%
6th - 11.72%
7th - 10.94%
8th - 9.38%
9th - 6.25%

*1st, 2nd, 3rd and 4th all have to win 3 knockout matches over the 4 weeks, with 1,2,3 having their week off before the Prelimnary Finals and 4th having their week off in the first week.

*5th and 6th will need to win 4 knockout games IF 1,2,3 all win in the first week. If anyone of 1,2,3 lose in the first week, the winner of 5v6 goes straight to the Prelim and therefore 5th or 6th would need to win 3 games.

*7th also need to win 4 straight knockout games, but they have a chance of going straight to the Prelim (and therefore needing only to win 3 games) but the chances of going straight to the Prelim are a bit less than 5th and 6th.

*8th also need to win 4 straight knockout games, but they have a chance of going straigth to the Prelim (and therefore needing only to win 3 games) but the chances of going straigth to the Prelim are a bit less than 7th. In fact, the only way they can advance straight to the Prelim is if 1st lose to 9th

*9th needs to win 4 straight knockout games and cannot go straight to the Prelim

Do not like this system at all. One system Really disadvantages the 4th placed team and creates an anomoly like was in the old McIntyre Final 6 system.

Top 4 should be guaranteed a double chance for their efforts during the season. Nor should the first placed team have the opportunity to be knocked out in week one. Teams under the second system will deliberately aim for 4th despite the strength of their lists. This is because they can get to see where they stand in terms of the comp by hoping team 1 and or 2 is knocked out in week one in a major upset. Then they get a fairly easy opponent in week 2 after being fresh with a week's rest. Straight to the preliminary they'll go.

First and second really should be given a higher and more favourable chance of winning than the rest of the teams, even though this only currently happens when the 3rd or 4th team is non Victorian or vice versa. It is unfair to teams like St Kilda and Geelong this season who lost less games combined than the third placed team yet get nothing for their efforts.

While your systems might be increadibly simple, they are also incredibly flawed. Thus I come to the conclusion that we have a final 8 system still (using the current system) or we can have a final 9 with the only change being an 8th vs 9th playoff in week one and teams 1-7 getting a bye before the current system starts.

In doing sdo, the split round will be axed due to time and it would eliminate any scheduling concerns the AFL will have in Round 22 as planned with this system as the game can be put on a Saturday night giving teams ample time to recover.

 

[jpdixon]

The winners of both the semi-final and the Qualifying final go to the Prelim where they have a 25% chance of winning the premiership.

I take your point about the winner of the semi-final being the "away" team the next week, but that doesn't make it any less important. It's still a Prelim they are advancing to, and it's still knockout if you lose.

Notice that I didn't say it makes it any less important - I just said that your claim that the results for the winners of the QF and the SF are the same is a bit weird. It's actually more than a bit weird. Both the week off and the home final are potentially significant, as you acknowledge when presenting your suggested final nines.

So, no matter how you cruch the numbers the second week of the finals is more important, interesting and significant than the Qualifying Finals.

Crunch which numbers? The numbers you're spouting don't go half the way to dealing with how interesting the games are, which was all that Chaz referred to. As I said in the part of my post that you've ignored, interest depends on more than just what is systematically riding on the game, even when you include the home advantage and a week off.

 

[jpdixon]

That's actually not a "flaw." That's just the way the systems work. A certain number of teams are guaranteed to progress straight to the Prelim finals under two of the finals 9's. If any of the top teams lose, then logically (and FAIRLY) the next highest winner takes their place.
...
You don't have to use a system like that. But it's not "unfair" to use it.

And Perry Pie didn't say that it was unfair. He said that it had a flaw. The word "flaw" is not such a limited word that in this context it applies only to your probabilities assuming level-pegged games, or even to a broader understanding of "fairness". Perry sees the dependence on other results as a flaw in way similar to how you see double chances as a flaw (whether you would say it that way or not). They both have disadvantages from certain points of view (and advantages from other points) - it's a matter of how you weigh them up, and what you want to emphasise in a finals series.

To reward teams mathematically in a knockout system, you need something other than 2,4,8 or 16 teams.

As a mathematician, I'd like to point out that your recurring equation of "mathematically" with "based on calculation of probabilities of winning the flag if each games is qually likely to be won by each team" does a great disservice to mathematics.

On this particular occasion, perhaps you mean that all the ways you can think of rewarding teams in a 2^n team knockout cannot be easily objectively quantified.

It's the same number of games as what they use now, but is technically mathematically fairer. ...
The top seed actually has a genuine advantage under the alternate system.

No, "technically mathematically", you can't say anything about whether this system is fairer without making fairly subjective assumptions about fairness. I'm sure we'd all agree that the higher-seeded team should have less chance of winning in the 50-50 model, but it's not at all clear which seeds should be given more of an advantage than others, especially in the NFL context.