Certified Legendary Thread The Squiggle is back in 2023 (and other analytics)

[Magpie Monopoly]

Suns no longer mathematically possible..

Gutted

 

[HTT]

Suns no longer mathematically possible..

|| W | L | D | % \1.| Richmond |18|4|0|146.2% \2.| Collingwood |17|5|0|134.7% \3.| Port Adelaide |17|5|0|130.7% \4.| West Coast |16|6|0|124.7% \5.| Sydney |15|7|0|123.6% \6.| Melbourne |13|9|0|121.4% \7.| Geelong |12|10|0|110.9% \8.| Gold Coast |10|12|0|94.3% \9.|Adelaide|9|13|0|92.3% \10.|Hawthorn|9|13|0|88.6% \11.|North Melbourne|9|13|0|87.8% \12.|St Kilda|8|13|1|89.0% \13.|GWS|8|13|1|84.9% \14.|Essendon|8|14|0|83.1% \15.|Fremantle|8|14|0|78.1% \16.|Brisbane Lions|7|15|0|92.6% \17.|Carlton|7|15|0|85.9% \18.|Western Bulldogs|6|16|0|70.2%

QF1: RICHMOND v West Coast QF2: COLLINGWOOD v Port Adelaide EF1: Sydney v GOLD COAST EF2: Melbourne v GEELONG SF1: West Coast v GOLD COAST SF2: Port Adelaide v GEELONG PF1: RICHMOND v Geelong PF2: Collingwood v GOLD COAST GF: Richmond v GOLD COAST

You've got a lot to learn about unrealistic hopes and football mathematics

 

[Schods]

Suns no longer mathematically possible..

Yes, they are. Carlton were still a possibility up until their loss on the weekend

 

[Billy ray]

QF1: RICHMOND v West Coast
QF2: COLLINGWOOD v Port Adelaide
EF1: Sydney v GOLD COAST
EF2: Melbourne v GEELONG

SF1: West Coast v GOLD COAST
SF2: Port Adelaide v GEELONG

PF1: RICHMOND v Geelong
PF2: Collingwood v GOLD COAST

GF: Richmond v GOLD COAST

You've got a lot to learn about unrealistic hopes and football mathematics

Too easy. Beat the swans on the SCG as they are vulnerable there.
Then west coast at optus where the pressure will be on umpires to not get sucked in by crowd noise.
Then collingwood at MCG will be in the bag as AFL want us in the grand final.
and finally smash the tigers as everyone will be on our side to stop the unbearableness of tiger fans going back to back.

 

[Hobbes]

Sydney +7 v Geelong
Richmond +39 v Adelaide
Brisbane +28 v Carlton
Port Adelaide +31 v St Kilda
Western Bulldogs v Hawthorn +21
Melbourne +35 v Fremantle
North Melbourne +49 v Gold Coast
Essendon v Collingwood +2
West Coast +14 v GWS

8/9. That's more like it. Sydney-Geelong results make less sense every time.

1. Richmond 23.3
2. Collingwood 15.2
3. Port Adelaide 12.4 (+2)
4. Hawthorn 11.3 (+6)
5. Geelong 11.2 (+1)
6. Essendon 10.4 (-3)
7. Melbourne 10.3
8. Sydney 8.0 (-4)
9. GWS 7.4 (+2)
10. West Coast 5.4 (-2)
11. North Melbourne 3.1 (-2)
12. Brisbane 0.4
13. Adelaide -11.7
14. St Kilda -13.2
15. Western Bulldogs -20.1
16. Fremantle -20.5
17. Carlton -30.1
18. Gold Coast -34.7

I'm pretty sure that, in the four years I've been doing this, it's the first time that Brisbane have recorded a positive rating.

Adelaide v Geelong +17
St Kilda +17 v Carlton
Hawthorn +17 v Brisbane
Melbourne +30 v Western Bulldogs
Gold Coast v Essendon +39
GWS v Richmond +11
Collingwood +19 v West Coast
North Melbourne +1 v Sydney
Fremantle v Port Adelaide +20



And the predictive ladder

1. Richmond 17.3
2. Port Adelaide 15.5
3. Collingwood 15.3 (+1)
4. West Coast 14.6 (+1)
5. Sydney 13.9 (-2)
6. Geelong 13.7 (+1)
7. Melbourne 13.57 (-1)
8. Hawthorn 13.55 (+1)
9. North Melbourne 12.9 (-1)
10. GWS 12.5
11. Essendon 11.4
12. Adelaide 9.5
13. Fremantle 7.9
14. Western Bulldogs 7.0
15. Brisbane 6.6
16. St Kilda 6.0
17. Gold Coast 3.7 (+1)
18. Carlton 3.2 (-1)

 

[fidstar]

I know this has probably been explained before, but how does West Coast drop two spots and GWS climb two spots even though the game result played out almost exactly as predicted?

 

[the harry]

I know this has probably been explained before, but how does West Coast drop two spots and GWS climb two spots even though the game result played out almost exactly as predicted?

21 scoring shots to 25

 

[Forward Press]

I know this has probably been explained before, but how does West Coast drop two spots and GWS climb two spots even though the game result played out almost exactly as predicted?

As the harry stated we got a bit lucky - I know I know, bad goal kicking is bad football but we certainly didn't exceed Squiggle's predictions.

 

[Hobbes]

I know this has probably been explained before, but how does West Coast drop two spots and GWS climb two spots even though the game result played out almost exactly as predicted?

Is this addressed to me?

The short answer is that the algorithm moves in mysterious ways.

The long answer is that past results get re-evaluated. For example, West Coast's loss to Adelaide became a worse-looking result, whereas GWS's recent results against Hawthorn and Brisbane were made to look better. Results going back 13 rounds are continually reassessed, but the real effect is that your rating will rise if teams you've played recently do well in other matches.

Really, West Coast's rating dropped a little, but they were overtaken by the rising GWS and Hawthorn.

(The algorithm itself calculated the West Coast-GWS result as a tiny win to GWS, since the margin was smaller than West Coast's notional home-ground advantage against GWS.)

 

[Hobbes]

Adelaide v Geelong +17
St Kilda +17 v Carlton
Hawthorn +17 v Brisbane
Melbourne +30 v Western Bulldogs
Gold Coast v Essendon +39
GWS v Richmond +11
Collingwood +19 v West Coast
North Melbourne +1 v Sydney
Fremantle v Port Adelaide +20

3/9. Still having rotten luck with these 50/50 tips.

1. Richmond 21.5
2. West Coast 14.7 (+8)
3. Melbourne 13.5 (+4)
4. GWS 13.1 (+5)
5. Essendon 11.4 (+1)
6. Sydney 9.5 (+2)
7. Brisbane 9.2 (+5)
8. Port Adelaide 8.8 (-5)
9. Collingwood 7.2 (-7)
10. Geelong 6.3 (-5)
11. Hawthorn 5.9 (-7)
12. North Melbourne 0.8 (-1)
13. Adelaide -5.0
14. St Kilda -6.7
15. Fremantle -15.1 (+1)
16. Western Bulldogs -23.7 (-1)
17. Carlton -35.1
18. Gold Coast -36.7

Massive movement here, as West Coast, GWS and Sydney win key matches, and Port, Collingwood, Geelong, Hawthorn and North don't. Lots of movement in the lower reaches too - Adelaide, St kilda and Fremantle rising, but invisible because the teams weren't bunched enough for places to change much.

St Kilda v Richmond +28
Collingwood +6 v North Melbourne
Sydney +55 v Gold Coast
Essendon +35 v Fremantle
Brisbane +24 v Adelaide
Geelong v Melbourne +7
Carlton v Hawthorn +49
West Coast +47 v Western Bulldogs
Port Adelaide +5 v GWS

A few shellackings predicted here, along with three important matches between contenders.

And the predictive ladder

1. Richmond 16.7
2. West Coast 15.8 (+2)
3=. Port Adelaide 14.51 (-1)
3=. Sydney 14.51 (+2)
5. Collingwood 14.1 (-2)
6. Melbourne 13.8 (+1)
7. GWS 13.4 (+3)
8. Geelong 12.6 (-2)
9. Hawthorn 12.5 (-1)
10. North Melbourne 12.1 (-1)
11. Essendon 11.5
12. Adelaide 10.5
13. Fremantle 9.1
14. Brisbane 7.9 (+1)
15. St Kilda 6.8 (+1)
16. Western Bulldogs 6.5 (-2)
17. Gold Coast 3.5
18. Carlton 2.5

 

[Suma Magic]

Final Siren

Are clubs generally more accurate at home than away? And is this incorporated into the model?

 

[Final Siren]

Final Siren

Are clubs generally more accurate at home than away? And is this incorporated into the model?

Interesting question, but no, there's no significant difference. Since 2011, home teams have converted at 52.98% while away teams have converted at 52.90%.

To put that 0.08 percentage point advantage into perspective, it means the home team has converted one additional attempt out of every 1,250 shots. Or put another way, once every 50 games, the away team has missed one shot that they would have been expected to convert as the home team.

Something that small is random noise, the equivalent of flipping a coin a thousand times and getting 501 heads and 499 tails.

 

[Suma Magic]

Interesting question, but no, there's no significant difference. Since 2011, home teams have converted at 52.98% while away teams have converted at 52.90%.

To put that 0.08 percentage point advantage into perspective, it means the home team has converted one additional attempt out of every 1,250 shots. Or put another way, once every 50 games, the away team has missed one shot that they would have been expected to convert as the home team.

Something that small is random noise, the equivalent of flipping a coin a thousand times and getting 501 heads and 499 tails.

Thanks for the info! I'm surprised that there isn't a difference, but I'm not sure that I should be...

 

[Final Siren]

Thanks for the info! I'm surprised that there isn't a difference, but I'm not sure that I should be...

Probably because West Coast usually have much bigger home vs away disparities than average:

This year, the Eagles have converted at 60.71% accuracy at home and 48.82% away!

Since 2011, they're 55.32% at home and 52.57% away. That means their home advantage converts an additional one out of every 36 shots (not 1 out of 1,250, like the league average).

 

[Joao]

Probably because West Coast usually have much bigger home vs away disparities than average:

This year, the Eagles have converted at 60.71% accuracy at home and 48.82% away!

Since 2011, they're 55.32% at home and 52.57% away. That means their home advantage converts an additional one out of every 36 shots (not 1 out of 1,250, like the league average).

That's nuts!

Goalkicking of affirmation perhaps?

 

[RookiePick]

Probably because West Coast usually have much bigger home vs away disparities than average:

This year, the Eagles have converted at 60.71% accuracy at home and 48.82% away!

Since 2011, they're 55.32% at home and 52.57% away. That means their home advantage converts an additional one out of every 36 shots (not 1 out of 1,250, like the league average).

Yowzers!

 

[Billy ray]

Probably because West Coast usually have much bigger home vs away disparities than average:

This year, the Eagles have converted at 60.71% accuracy at home and 48.82% away!

Since 2011, they're 55.32% at home and 52.57% away. That means their home advantage converts an additional one out of every 36 shots (not 1 out of 1,250, like the league average).

just for shits and giggles, can you do gold coast?

 

[Final Siren]

That's nuts!

Goalkicking of affirmation perhaps?

Must be all those free kicks in Perth from the goal square.

But no, you expect bigger home/away differentials in non-Vic teams because almost all of their games involve serious travel by one team or the other. And Perth requires more travel than anywhere else.

That said, looking at the data, there doesn't seem to be any such trend, so it could be all random:

|Home|Away|Home Accuracy Bonus
\West Coast|55.32%|52.57%|+2.75%
\St Kilda|52.39%|50.05%|+2.34%
\Carlton|52.42%|50.54%|+1.88%
\Geelong|54.23%|52.91%|+1.32%
\Gold Coast|51.51%|50.35%|+1.16%
\Greater Western Sydney|52.23%|51.10%|+1.13%
\Essendon|52.32%|51.34%|+0.98%
\Melbourne|52.94%|52.49%|+0.45%
\PortAdelaide|53.27%|53.30%|-0.03%
\Western Bulldogs|52.21%|52.30%|-0.09%
\North Melbourne|54.15%|54.34%|-0.19%
\Hawthorn|55.78%|56.16%|-0.38%
\Collingwood|52.03%|52.89%|-0.86%
\Brisbane Lions|52.22%|53.23%|-1.01%
\Adelaide|52.96%|54.09%|-1.13%
\Fremantle|52.29%|53.91%|-1.62%
\Richmond|51.34%|52.97%|-1.63%
\Sydney|52.41%|55.61%|-3.20%

Probably the most interesting thing out of this is how accurate Hawthorn are! Gold Coast are the least accurate.

 

[Joao]

Must be all those free kicks in Perth from the goal square.

But no, you expect bigger home/away differentials in non-Vic teams because almost all of their games involve serious travel by one team or the other. And Perth requires more travel than anywhere else.

That said, looking at the data, there doesn't seem to be any such trend, so it could be all random:

|Home|Away|Home Accuracy Bonus
\West Coast|55.32%|52.57%|+2.75%
\St Kilda|52.39%|50.05%|+2.34%
\Carlton|52.42%|50.54%|+1.88%
\Geelong|54.23%|52.91%|+1.32%
\Gold Coast|51.51%|50.35%|+1.16%
\Greater Western Sydney|52.23%|51.10%|+1.13%
\Essendon|52.32%|51.34%|+0.98%
\Melbourne|52.94%|52.49%|+0.45%
\PortAdelaide|53.27%|53.30%|-0.03%
\Western Bulldogs|52.21%|52.30%|-0.09%
\North Melbourne|54.15%|54.34%|-0.19%
\Hawthorn|55.78%|56.16%|-0.38%
\Collingwood|52.03%|52.89%|-0.86%
\Brisbane Lions|52.22%|53.23%|-1.01%
\Adelaide|52.96%|54.09%|-1.13%
\Fremantle|52.29%|53.91%|-1.62%
\Richmond|51.34%|52.97%|-1.63%
\Sydney|52.41%|55.61%|-3.20%

Probably the most interesting thing out of this is how accurate Hawthorn are! Gold Coast are the least accurate.

Is there anything suggesting particular grounds are better than others?

Hawthorn doesn't surprise me. At their peak they just kicked through you for an easy uncontested mark 35m out on a slight angle. Much easier than the 45m boundary kicks they'd make us take to score.

 

[Freomaniac]

Some things never change with the squiggle. We won the flag and it STILL doesn't think we're a chance.

No body beats the squiggle, NOBODY!

 

[Billy ray]

Yay.

 

[Pessimistic]

Slow day was perusing historical squiggles

Noticed only three premiers came from a starting point in that year in the latter part of the squiggle field

Richmond in 2017. Geelong in 2007. Adelaide in 1997. Also Essendon 1993 to an extent

The other characteristics? Apart from being exactly a decade apart, Adelaide went b2b and geelong won three of the next five. Richmond look like there’s at least one more flag capability, and this ‘omen’ bears that out

 

[Roobs321]

Interesting that squiggle predicted the Lions to beat the Crows by 5, shows how far they've caught up, as there must have been a chasm between them in 2017.

 

[Hobbes]

St Kilda v Richmond +28
Collingwood +6 v North Melbourne
Sydney +55 v Gold Coast
Essendon +35 v Fremantle
Brisbane +24 v Adelaide
Geelong v Melbourne +7
Carlton v Hawthorn +49
West Coast +47 v Western Bulldogs
Port Adelaide +5 v GWS

5/9.

1. Richmond 23.7
2. GWS 18.9 (+2)
3. West Coast 17.5 (-1)
4. Collingwood 14.4 (+5)
5. Essendon 13.2
6. Melbourne 10.1 (-3)
7. Hawthorn 9.5 (+4)
8. Brisbane 6.0 (-1)
9. Port Adelaide 4.3 (-1)
10. Geelong 3.4
11. Adelaide 0.4 (+2)
12. Sydney -3.8 (-6)
13. North Melbourne -8.1 (-1)
14. St Kilda -11.0
15. Fremantle -16.1
16. Western Bulldogs -25.3
17. Gold Coast -26.5 (+1)
18. Carlton -40.2 (-1)

GWS, Collingwood and Hawthorn rise, Sydney plummet.

Essendon +22 v Sydney
Richmond +9 v Collingwood
Geelong +3 v Brisbane
GWS +35 v St Kilda
Gold Coast +21 v Carlton
Adelaide v Melbourne +4
North Melbourne v West Coast +16
Western Bulldogs v Port Adelaide +23
Fremantle v Hawthorn +17



And the predictive ladder

1. Richmond 16.8
2. West Coast 16.2
3. Collingwood 15.1 (+2)
4. GWS 14.4 (+3)
5. Port Adelaide 13.6 (-2)
6. Hawthorn 13.0 (+3)
7. Melbourne 12.94 (-1)
8. Geelong 12.9
9. Sydney 12.7 (-6)
10. Essendon 11.9
11. Adelaide 11.6 (+1)
12. North Melbourne 11.2 (-2)
13. Fremantle 8.8
14. Brisbane 6.9
15. Western Bulldogs 6.6 (+1)
16. St Kilda 6.3 (-1)
17. Gold Coast 5.2
18. Carlton 1.9

 

[Forward Press]

Not a good weekend for Sydney fans.