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

[freo_leo29]

Logue $800k!
That's ridiculous for essentially a role player

North have unlimited money rn so its really a free swing for them, better to get a player in instead of spending up on holding onto spuds

but yes it is a wild kind of once in a lifetime offer for Logue that he would have had no choice but to take

also I think as a defender he is better than a role player but thats obviously going to be tested so we will see

 

[Gavin Excell]

Been long of the view that 3rd forward tall types are the most overpaid for actual output.

 

[Strapping Young Lad]

I did forget him.

Fyfe and Omeara hobbling around flexing their muscles doesn't fill me with confidence.

O’meara will be a great get for the Dockers.

 

[fishardansin]

Very generous of Carlton. Underestimating the Pies again.

5th is underestimating? lol

 

[fishardansin]

Squiggle is back.

I challenge Squiggle to a duel



They may have data on their side, but I searched my feelings.

 

[Catsace]

Been long of the view that 3rd forward tall types are the most overpaid for actual output.

McStay certainly wouldn't have knocked back that Pies offer, again for a 3rd tall type in maybe the Dawes mode for the Pies now.

Logue $800K, wow he'd be among the top 3 or so paid at Catland on that coin, just wouldn't happen at Geelong.

 

[tonygeeks]

Old mate is a real club can do no wrong type , Freo screwed it up with Louge we could have signed him to half that 800k but took too long to offer him a deal and by the time we did he had been shopping himself around and some big dough offers came through

That wasn’t even the highest offer apparently

Super athletic dumb ass but there ain’t many who can shut down Cameron ( C ) to Cameron ( J)

 

[Joao]

Old mate is a real club can do no wrong type , Freo screwed it up with Louge we could have signed him to half that 800k but took too long to offer him a deal and by the time we did he had been shopping himself around and some big dough offers came through

That wasn’t even the highest offer apparently

Super athletic dumb ass but there ain’t many who can shut down Cameron ( C ) to Cameron ( J)

If the early re-signing is true, we botched that one.

That said, he really shouldn't be b22 anyway because we need actual forwards in the forward line. So he'd have ended up depth and leaving in a year or two for SFA. Either that or Pearce snaps in half and he plays the next 200 games straight.

 

[Mr MysteryGuest]

Been long of the view that 3rd forward tall types are the most overpaid for actual output.

how much do you think they get paid?

 

[Fadge]

Based on the squiggle, they rate Geelong a better than 50% chance to finish top 2?

I'd love to get even money about them to miss top 2.

Any of about 10 teams could make the top 4 in 2023, and it won't take much for any of them to miss out...

 

[SplintWhisky]

I challenge Squiggle to a duel

View attachment 1582935

They may have data on their side, but I searched my feelings.

 

[Spirit Level]

Re Freo:

Fremantle had a 13-point win with 2 fewer scoring shots, at home (so arguably a loss at a neutral venue given strong HGA), followed by a 20-point loss with 10 fewer scoring shots away (so arguably a worse loss than the scoring margin suggested given Freo only generated 14 scoring shots), to teams that both had H&A percentages only just above 100% on the season. That indicates that they were two pretty poor performances in finals against two other teams that in turn were pretty average on the season.

People love trying to come up with complex reasons to obvious simple things that have an answer. Fremantle across their two most recent matches (any very relevant on form), arguably did not play up to even a top-8 team standard. Strange it hasn't been said so far!

Need some sugar for those grapes?

 

[Purple Suit]

Re Freo:

Fremantle had a 13-point win with 2 fewer scoring shots, at home (so arguably a loss at a neutral venue given strong HGA), followed by a 20-point loss with 10 fewer scoring shots away (so arguably a worse loss than the scoring margin suggested given Freo only generated 14 scoring shots), to teams that both had H&A percentages only just above 100% on the season. That indicates that they were two pretty poor performances in finals against two other teams that in turn were pretty average on the season.

People love trying to come up with complex reasons to obvious simple things that have an answer. Fremantle across their two most recent matches (any very relevant on form), arguably did not play up to even a top-8 team standard. Strange it hasn't been said so far!

So the squiggle just ignores what occurred during the whole season and replaces it with finals performance?

It should examine quarter by quarter performance as Freo were blown away in both games early playing most players first final & then infront of 90k hostile fans. They then steadily and beat the dogs and were coming back against the Pies but left too much to do.

How about Sydney then after the GF? They got belted so that will look bad and only won the prelim by 1 point at home so basically a loss according to you.

 

[Fadge]

So the squiggle just ignores what occurred during the whole season and replaces it with finals performance?

It should examine quarter by quarter performance as Freo were blown away in both games early playing most players first final & then infront of 90k hostile fans. They then steadily and beat the dogs and were coming back against the Pies but left too much to do.

How about Sydney then after the GF? They got belted so that will look bad and only won the prelim by 1 point at home so basically a loss according to you.

If the rationale is basing on performances against Collingwood, who were allegedly lucky to even make the top 8, there weren't any teams who played a top 8 standard of football during the finals series, given the premier only beat them by a goal (one more scoring shot) in the Qualifying Final.

So we should be looking at the teams who didn't make the top 8 in 2022 as the stronger teams in 2023. Go big on Carlton, St Kilda, Port Adelaide and Gold Coast!

 

[Bulldog Joe]

Seems a lot are triggered by the analytics.

The squiggle has shown a good accuracy over time but adjusts as every game is played.

There is no emotion in squiggle and I accept it for what it is.

I can still believe it under estimates where my team will be as 2023 is a new year and meaningful data only really becomes available when the season actually starts.

 

[peetoo]

The fixture which everyone acknowledges has an a effect, is factored in. It’s like a bench for each team. Carlton must have one of the easier fixtures

 

[Purple Suit]

I think the model's aggregate ladder and what the media are saying about particular teams is quite different.

View attachment 1341341

I keep hearing in the media about how Carlton are going to play finals, and how Hawthorn are in danger of winning the wooden spoon. Yet, Hawthorn sit above Carlton on the aggregate ladder.

This was what the squiggle was predicting last year before the season.

I think it's safe to say that the Squiggle isn't very good at picking movement between seasons, just like the rest of us.

 

[Fadge]

The fixture which everyone acknowledges has an a effect, is factored in. It’s like a bench for each team. Carlton must have one of the easier fixtures

The fixture (in advance) has zero effect.

The fixture (in retrospect) has negligible effect.

 

[Strapping Young Lad]

Seems a lot are triggered by the analytics.

The squiggle has shown a good accuracy over time but adjusts as every game is played.

There is no emotion in squiggle and I accept it for what it is.

I can still believe it under estimates where my team will be as 2023 is a new year and meaningful data only really becomes available when the season actually starts.

Where it gets interesting for teams who have losses or gained players is how the squiggle rates those players individually.

For example, when Sicily and Gunston were returning from injury last year the squiggle didn’t really rate them as beneficial inclusions that would have much of an impact which I questioned at the time, and this year while it seems to rate the older players we’ve traded out it actually rates the players we were missing from the second half of 2022 due to injury higher, so it’s saying we’ll be better while most people think we’ll be worse.

Dogs lose Dunkley and Hunter but gain Lobb and Jones, is that a gain or a loss according to Squiggle?

 

[peetoo]

The fixture (in advance) has zero effect.

The fixture (in retrospect) has negligible effect.

On prediction?

 

[peetoo]

Funny the opening round has probabilities between 50 and 62%.

The closest being hawks v bombers

 

[Xtreme]

Thou shall not question the almighty Squiggle!

 

[Purple Suit]

Where it gets interesting for teams who have losses or gained players is how the squiggle rates those players individually.

For example, when Sicily and Gunston were returning from injury last year the squiggle didn’t really rate them as beneficial inclusions that would have much of an impact which I questioned at the time, and this year while it seems to rate the older players we’ve traded out it actually rates the players we were missing from the second half of 2022 due to injury higher, so it’s saying we’ll be better while most people think we’ll be worse.

Dogs lose Dunkley and Hunter but gain Lobb and Jones, is that a gain or a loss according to Squiggle?

The squiggle tracks player movement and return from injury but it probably needs to factor in age of the players too

18-25 year olds are probably improving every season on average.
25-30 staying the same
30+ declining

Output based on season performance should have an age demographic incorporated, that might reduce Geelong's run away flag favouritism or Richmond's high mark. It might also predict young rising teams more than it has in the past.

 

[Joao]

So the squiggle just ignores what occurred during the whole season and replaces it with finals performance?

It should examine quarter by quarter performance as Freo were blown away in both games early playing most players first final & then infront of 90k hostile fans. They then steadily and beat the dogs and were coming back against the Pies but left too much to do.

How about Sydney then after the GF? They got belted so that will look bad and only won the prelim by 1 point at home so basically a loss according to you.

Squiggle isn't nuanced enough to capture that sort of effect. I'd argue there probably isn't a mathematical model that could do that as it would be ridiculously complex IF you could even quantify those effects.

Squiggle is good for trends, I would never recommend anyone use it for a one off game.

Squiggle also notoriously doesn't like low scoring teams hence the Freo hate (that and our ordinary 2nd half of 2022).

This was what the squiggle was predicting last year before the season.

I think it's safe to say that the Squiggle isn't very good at picking movement between seasons, just like the rest of us.

100%

It has no way to capture simple things that make a massive difference like natural improvement/decline, quality of pre-season etc.

 

[peetoo]

In short, the prediction is how a team should perform not how ther will perform