Certified Legendary Thread Race for the flag, in squiggly lines

Remove this Banner Ad

Final Siren said:
Hmm! Well one way of rating it would be to consider how much better the dynastic team was over their three seasons of dominance than their leading challengers.

This means we rate teams that spent the whole year looking like champions, rather than teams that only peaked at the end.

We probably don't want to compare them to the season average, since that gets skewed by particularly bad teams (and 2013 had two terrible ones). So let's compare the eventual premier to the other seven finalists. That's still subject to skewing by years with unusually poor/strong finalists, but not as badly, and you have to draw the line somewhere.

We'll also smoosh Geelong's three premiership years together as if they were a three-peat. This either gives the Cats too much credit, since they couldn't pull that off in reality, or undersells them, since their dynasty was actually over a longer period, whichever way you want to look at it.

This gives us:

#1 Brisbane 2002

15.3% better than opposition finalists over the course of the year, with only Port (midyear) and Collingwood (late) providing any sort of challenge.

#2 Hawthorn 2015

13.3% better. Never headed but a year with a particularly strong top 8.

#3 Geelong 2011

9.7% better. Played second fiddle to Collingwood for most of the year, but still a lot better than the average finalist.

#4 Hawthorn 2014

9.3% better. Spent the whole year either at #1 or very close to it, but with credible competition from Sydney, Fremantle, and Port.

#5 Brisbane 2003

9.1% better. Always threatening, but never got too far away from Port and Collingwood.

#6 Geelong 2007

9.1% better. Started the year a long way back before going a long way ahead. A season that is done a disservice by averages.

#7 Geelong 2009

8.5% better. Consistently threatening all year, but often behind St Kilda, with late challenges from the Bulldogs and Adelaide.

#8 Hawthorn 2013

8.0% better. One of four strong challengers with Sydney, Geelong, and Fremantle.

#9 Brisbane 2001

4.2% better. Trailed Essendon for almost the whole year and really only got going mid-season.

If you average the three years:
10.2% Hawthorn
9.5% Brisbane
9.1% Geelong

This of course is just according to the particular criteria I've invented here. "Best dynasty" is subjective, so no-one will agree on what it means.
Click to expand...

So...

Hawthorn 》 Brisbane 》 Geelong

Didnt even need to play the 2008 card :)
 
threenewpadlocks said:
Statistically speaking they drag the averages of the league down. It's not like you're adding a league-average team to the competition (which it would be fine to keep them in), rather, you're adding a team which makes the pre-existing clubs seem better than they are, relative to the teams of previous years.
Click to expand...
But the Squiggle is self correcting over time. The more those teams were thrashed the more they had to be thrashed by to get the same movement and the more games that are played the less those earlier games factor into it.
 
Final Siren said:
Hmm! Well one way of rating it would be to consider how much better the dynastic team was over their three seasons of dominance than their leading challengers.

This means we rate teams that spent the whole year looking like champions, rather than teams that only peaked at the end.

We probably don't want to compare them to the season average, since that gets skewed by particularly bad teams (and 2013 had two terrible ones). So let's compare the eventual premier to the other seven finalists. That's still subject to skewing by years with unusually poor/strong finalists, but not as badly, and you have to draw the line somewhere.

We'll also smoosh Geelong's three premiership years together as if they were a three-peat. This either gives the Cats too much credit, since they couldn't pull that off in reality, or undersells them, since their dynasty was actually over a longer period, whichever way you want to look at it.

This gives us:

#1 Brisbane 2002

15.3% better than opposition finalists over the course of the year, with only Port (midyear) and Collingwood (late) providing any sort of challenge.

#2 Hawthorn 2015

13.3% better. Never headed but a year with a particularly strong top 8.

#3 Geelong 2011

9.7% better. Played second fiddle to Collingwood for most of the year, but still a lot better than the average finalist.

#4 Hawthorn 2014

9.3% better. Spent the whole year either at #1 or very close to it, but with credible competition from Sydney, Fremantle, and Port.

#5 Brisbane 2003

9.1% better. Always threatening, but never got too far away from Port and Collingwood.

#6 Geelong 2007

9.1% better. Started the year a long way back before going a long way ahead. A season that is done a disservice by averages.

#7 Geelong 2009

8.5% better. Consistently threatening all year, but often behind St Kilda, with late challenges from the Bulldogs and Adelaide.

#8 Hawthorn 2013

8.0% better. One of four strong challengers with Sydney, Geelong, and Fremantle.

#9 Brisbane 2001

4.2% better. Trailed Essendon for almost the whole year and really only got going mid-season.

If you average the three years:
10.2% Hawthorn
9.5% Brisbane
9.1% Geelong

This of course is just according to the particular criteria I've invented here. "Best dynasty" is subjective, so no-one will agree on what it means.
Click to expand...
Thank you very much. That provides clarity to a very subjective topic.
 

Log in to remove this ad.

cryptor said:
Why does the fact GWS and Gold Coast were particularly poor sides in their first few years matter?

Doesn't the squiggle account for this by making it so teams have to utterly demolish them to get the same movement that you'd get from simply beating a good side by a decent margin?

The only sides that would really benefit from them were the ones who played them early before the Squiggle knew it shouldn't rate them.
Click to expand...
Great question! You're right that pretty soon the new teams should reach the position they deserve, and thereafter not move much. But the effect of their entry will still ripple through the whole league and cause everyone else to shift a bit.

This is because the teams that thump the new teams in the first few weeks will move, and after that, any team that plays them will get more credit than they would have for the exact same scoreline.

For example, say that over the off-season Gil has a change of heart and rushes in the Tasmania Devils. As a new team, the squiggle would rate the Devils 50/50. Let's say Round 1 has Essendon vs Devils and Round 2 has Collingwood vs Essendon, with the latter match predicted to be a Pies win 80 to 60.

So in Week 1, Essendon smash the Devils, and as a result move up the chart. The squiggle will now change its prediction to have Essendon performing better against Collingwood. But let's say they don't, because the original prediction was right all along: Collingwood win 80 to 60. This result is better than the new squiggle prediction, so Collingwood move up (and Essendon back, losing some, but not all, of their gains from the Round 1 game). Thus Collingwood are rated higher because of the Devils' introduction even though they haven't played them yet.

Whoever plays the Pies in Round 3 will benefit from the same effect, as will whoever plays Essendon. So after only three rounds, the Devils will have affected six teams: their three direct opponents plus everyone those teams have subsequently played. By the end of Round 5, it's probably spread to the whole league, although by then it will be very diffuse and too small to notice.
 
Last edited:
Final Siren said:
Great question! You're right that pretty soon the new teams should reach the position they deserve, and thereafter not move much. But the effect of their entry will still ripple through the whole league and cause everyone else to shift a bit.

This is because the teams that thump the Devils in the first few weeks will move, and after that, any team that plays them will get more credit than they would have for the exact same scoreline.

For example, say that over the off-season Gil has a change of heart and rushes in the Tasmania Devils. As a new team, the squiggle would rate the Devils 50/50. Let's say Round 1 has Essendon vs Devils and Round 2 has Collingwood vs Essendon, with the latter match predicted to be a Pies win 80 to 60.

So in Week 1, Essendon smash the Devils, and as a result move up the chart. The squiggle will now change its prediction to have Essendon performing better against Collingwood. But let's say they don't, because the original prediction was right all along: Collingwood win 80 to 60. This result is better than the new squiggle prediction, so Collingwood move up (and Essendon back, losing some, but not all, of their gains from the Round 1 game). Thus Collingwood are rated higher because of the Devils' introduction even though they haven't played them yet.

Whoever plays the Pies in Round 3 will benefit from the same effect, as will whoever plays Essendon. So after only three rounds, the Devils will have affected six teams: their three direct opponents plus everyone those teams have subsequently played. By the end of Round 5, it's probably spread to the whole league, although by then it will be very diffuse and too small to notice.
Click to expand...
Actually I had a look at 2012, because it's deliciously virus-like. This is how the effect of GWS propagates:

Round 1: GWS-Syd Ric-Car Haw-Col Mel-Bri GC-Ade Fre-Gee Nor-Ess Bul-WC Por-StK
Round 2: Syd-Fre GWS-Nor Bri-Car Ess-Por WC-Mel Ade-Bul Col-Ric StK-GC Gee-Haw
Round 3: Por-Syd Fre-Bri GWS-Wes Nor-Gee Car-Col Ric-Mel GC-Ess Bul-StK Haw-Ade
Round 4: StK-Fre Col-Por Ade-GWS Wes-Haw Bri-GC Gee-Ric Syd-Nor Car-Ess Mel-Bul
Round 5: Mel-StK Fre-Car Col-Ess GWS-Bul Nor-GC Bri-Gee Haw-Syd Ric-WC Ade-Por

So if GWS were a zombie, and every team it played also turned into a zombie, the whole league would be zombies by the end of Round 5, with the final survivor being Melbourne, succumbing to an infected St Kilda on Saturday night at the G (and losing, too).
 
Final Siren said:
Great question! You're right that pretty soon the new teams should reach the position they deserve, and thereafter not move much. But the effect of their entry will still ripple through the whole league and cause everyone else to shift a bit.

This is because the teams that thump the Devils in the first few weeks will move, and after that, any team that plays them will get more credit than they would have for the exact same scoreline.

For example, say that over the off-season Gil has a change of heart and rushes in the Tasmania Devils. As a new team, the squiggle would rate the Devils 50/50. Let's say Round 1 has Essendon vs Devils and Round 2 has Collingwood vs Essendon, with the latter match predicted to be a Pies win 80 to 60.

So in Week 1, Essendon smash the Devils, and as a result move up the chart. The squiggle will now change its prediction to have Essendon performing better against Collingwood. But let's say they don't, because the original prediction was right all along: Collingwood win 80 to 60. This result is better than the new squiggle prediction, so Collingwood move up (and Essendon back, losing some, but not all, of their gains from the Round 1 game). Thus Collingwood are rated higher because of the Devils' introduction even though they haven't played them yet.

Whoever plays the Pies in Round 3 will benefit from the same effect, as will whoever plays Essendon. So after only three rounds, the Devils will have affected six teams: their three direct opponents plus everyone those teams have subsequently played. By the end of Round 5, it's probably spread to the whole league, although by then it will be very diffuse and too small to notice.
Click to expand...
I wondered if this might be the case when I asked but I thought perhaps it would not matter in the long run.

The reason being that teams only play these sides once or twice in a season while GWS and Gold Coast are involved in games that involve themselves every week. So very quickly they will find themselves "where they belong" while all the other sides spread the impact results had on one side among the others with the result being teams end the season in just about the same spots anyway.

The same effect could be witnessed by a side belting another "unexpectedely" due to injury.

If possible it would be interesting to see a season squiggle of each of the years GWS and Gold Coast entered but with results against these sides excluded and then compared to what actually happened. My gut feel is that teams would over shoot in the actual version but still end up within a reasonably close position to the one with expansion sides excluded.
 
Wbby said:
Not really. Doesn't need to be 100% of the time, but if it happens 90% of the time. You can say that it is more likely than not. So one outlier doesn't change much really
Click to expand...
This record stands again this year. Hawks worst defeat 32 points, incidentally in the QF against their GF opponent, in the H&A it was 22 points. Eagles worst defeat 57 points.
 
Final Siren said:
Actually I had a look at 2012, because it's deliciously virus-like. This is how the effect of GWS propagates:

Round 1: GWS-Syd Ric-Car Haw-Col Mel-Bri GC-Ade Fre-Gee Nor-Ess Bul-WC Por-StK
Round 2: Syd-Fre GWS-Nor Bri-Car Ess-Por WC-Mel Ade-Bul Col-Ric StK-GC Gee-Haw
Round 3: Por-Syd Fre-Bri GWS-Wes Nor-Gee Car-Col Ric-Mel GC-Ess Bul-StK Haw-Ade
Round 4: StK-Fre Col-Por Ade-GWS Wes-Haw Bri-GC Gee-Ric Syd-Nor Car-Ess Mel-Bul
Round 5: Mel-StK Fre-Car Col-Ess GWS-Bul Nor-GC Bri-Gee Haw-Syd Ric-WC Ade-Por

So if GWS were a zombie, and every team it played also turned into a zombie, the whole league would be zombies by the end of Round 5, with the final survivor being Melbourne, succumbing to an infected St Kilda on Saturday night at the G (and losing, too).
Click to expand...
This would look awesome in a diagram.
Modelling a virus spread throughout the AFL teams, not in an exact exponential growth which is cool.
 
Final siren,

Can you explain what makes this years top 8 so strong I'm just curious, because looking at the 8 teams it's hard to work out what made the team's so strong. Would you say it's an anomaly of your system? As looking at the games and the results of the finals say it was anything but a strong final 8. If you do think it was a strong top 8. Which teams do you believe fit into the category?

Cheers!
 
thebarry111 said:
Final siren,

Can you explain what makes this years top 8 so strong I'm just curious, because looking at the 8 teams it's hard to work out what made the team's so strong. Would you say it's an anomaly of your system? As looking at the games and the results of the finals say it was anything but a strong final 8. If you do think it was a strong top 8. Which teams do you believe fit into the category?

Cheers!
Click to expand...

Strong is a relative term in this case. The "top 8" were comparably better than the bottom 10 than in other years - ergo the top 8 is "strong".

Replace "strong" with "dominant" and you may have a better description.
 
Simon_Nesbit said:
Strong is a relative term in this case. The "top 8" were comparably better than the bottom 10 than in other years - ergo the top 8 is "strong".

Replace "strong" with "dominant" and you may have a better description.
Click to expand...
Yes, this. Many years, there's not much separation between teams, so you see a big clump where they're all close to the middle. But by the end of this season, we hardly had a middle at all: instead we had a very credible top 10 all the way down to Geelong, then an enormous gap to the bottom 6, with only Collingwood and GWS in between.

Compare for example 2002, where a 13-9 record got you top 4 while 9-13 got you bottom 4! The Swans and Bulldogs finished 11th and 12th with percentages of 107.4 and 104.0 respectively, while Melbourne, North, and West Coast all made the finals with percentages below 100. Really 6th-12th spots were a raffle where anyone could have made it or missed out.
 

(Log in to remove this ad.)

A little table I just made up based on Squiggle overall rating and 2016 fixture.

Difficulty = Average of all opponents (home and away) - HGA 12 points.
Wins/% = Pretty obvious
Code: 
                  Difficulty                     Percentage                Wins
Team        Home     Away    Overall       Home    Away   Overall    Home  Away Overall
Hawthorn    60.01    61.18    60.60        155%    104%   129%        11    6    17
Sydney      61.39    59.41    60.40        137%    94%    115%        11    5    16
West        59.04    61.20    60.12        151%    98%    124%        11    4    15
Port        60.28    58.44    59.36        137%    94%    115%        11    4    15
North       61.19    60.49    60.84        135%    91%    113%        11    4    15
Richmond    58.58    62.72    60.65        144%    90%    117%        11    3    14
Adelaide    60.15    62.40    61.28        132%    85%    109%        11    3    14
Western     61.86    58.08    59.97        127%    90%    108%        10    3    13
Fremantle   60.69    59.85    60.27        128%    86%    107%        11    2    13
Geelong     60.76    57.95    59.35        127%    87%    107%        10    1    11
Collingwood 60.14    60.59    60.36        121%    78%    100%        10    0    10
GWS         63.08    58.49    60.79        109%    76%    92%         7     0    7
Essendon    58.30    60.88    59.59        112%    67%    89%         7     0    7
St          57.80    61.59    59.70        111%    66%    89%         7     0    7
Melbourne   60.45    60.75    60.60        108%    68%    88%         7     0    7
Gold        57.65    60.80    59.23        112%    67%    89%         6     0    6
Brisbane    62.22    58.83    60.52        103%    68%    86%         5     0    5
Carlton     59.70    59.65    59.67        103%    64%    83%         5     0    5

Some early conclusions - Geelong, Port, Essendon, Carlton and Gold Coast have great draws - though it doesn't help Carlton too much!

Predicting a Hawthorn and Interstate top-4 again.
Compared with 2015 end positions, Richmond to drop, Port to climb, Melbourne to slide back a bit.

Do we over-rate the Strength of Schedule?
 
Seppo said:
Fixture been put long enough to get 2016 squiggle going Final Siren ?

I'm on my honeymoon at the moment and have found some Internet to check this thread for an update...
Click to expand...
Good to see you've got your priorities straight Seppo.

Although if I'd known how it was going to come out, I would have done it sooner!

2016 Ladder Predictor
yo5crkE.png

Play with live squiggle here - same as last year, you can drag teams around to new locations and generate new predictions. Also you can click through every match of 2016 to see the squiggle tips.

This is all based purely on 2015 match results plus the 2016 fixture. That is, there is NO accounting for trades, delistings, retirements, drafting, injuries, returns from injury, or anything else that happens in the off-season. (Which is why it's similar to the 2015 ladder for most teams.) You have to factor that stuff in as and where you think it's significant.

Fremantle is low because the squiggle genuinely thought the Dockers were that average by the end of 2015, based on the scorelines they were delivering. Likewise, it thought Port were pretty good, despite missing finals. Although the Power may be Top 4 only because they got to finish their season off by thumping Fremantle's B-team.
 
Geez, big call on Freo. As much as I would love to see the purp-a-derps languishing in the bottom half of the table I still think they'll play finals next year. Probably a 4th-6th side in reality.
 
Final Siren said:
Hello all! Are you, like me, suffering through another content-free off-season? If so, good news! Next week there will be a brand new squiggle addition.

Can't say too much but it should provide relief for those in desperate need of some footy-based time-wasting.
Click to expand...

You had me at footy based time wasting....
 
Maybe its who squiggle thinks is training the house down?

Will it fall for the Liam Jones types? Or predict the next break out?
 
Final Siren said:
Hello all! Are you, like me, suffering through another content-free off-season? If so, good news! Next week there will be a brand new squiggle addition.

Can't say too much but it should provide relief for those in desperate need of some footy-based time-wasting.
Click to expand...

Jade said:
You had me at footy based time wasting....
Click to expand...

you had me at 'hello'
 
You must log in or register to reply here.

Similar threads

PieLebo87
  • Poll
The race for the 2023 flag
5 6 7 8 9
Replies
220
Views
10K
nahnah
nahnah
Havoc
Flags won in the digital television era
2
Replies
36
Views
2K
Freomaniac
Freomaniac
Lsta062
Analysis Best forward lines of 2024
2 3 4
Replies
82
Views
4K
Lsta062
Lsta062
T
Rolling Flag Contenders 2024
2
Replies
48
Views
2K
Robroy22
Robroy22
Orange crush
Analysis Best back lines in 2024
2
Replies
45
Views
3K
Corn Cobbers
Corn Cobbers
Matchu
  • Poll
Would a 2023 flag for the Giants be more impressive than the Bulldogs' 2016 flag?
Replies
16
Views
692
Pocket Terrace
Pocket Terrace
Taper
  • Poll
Who's closer to a flag? Brisbane or Gold Coast?
Replies
8
Views
134
Son of Purple
Son of Purple

Remove this Banner Ad

Back
Top