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

[faible]

Hawthorn were bottom 2 in the first half of the year, playing overly defensive football, letting sides start with a flurry.




At the halfway mark of the year the gameplan changed(for the third time in the last 5 years), Clarkson was ultimately moved on and our midfield set up was modified for the better with Mitchell’s influence(Newcombe and Nash).

View attachment 1341484

they’re not the stats of a side just catching some good sides off guard.

By the end of the year Hawthorn we’re not a bottom 2 side, by any stretch, but it remains to see what happens in 2022.

Yet the model didn't rate them as a team drastically on the improve for 2022 still.

Maybe they will, maybe they won't. You can create a narrative why they will, the whole point of the model is that it ignores the narrative and just looks at statistics.

No idea why you've got your knickers in a knot about it. Squiggle models Hawthorn as 0.2 wins below the aggregate of all the models, it just rates other teams as more likely to improve than Hawthorn so the ladder position slides.

The model could be right, or it could be wrong. Arguing about it won't magically make the model change what it thinks.

 

[Strapping Young Lad]

Hawthorn were bottom 4 for percentage in season 2021, which is an accurate depiction of the sides quality and depth. Good luck in 2022. You'll need it.

Percentage captures seasonal output, it does not drill down on the detail or trends within a season.

 

[Trav 20]

Percentage captures seasonal output, it does not drill down on the detail or trends within a season.

It's remarkably accurate. The top 5 sides also had the top 5 percentages.

 

[Strapping Young Lad]

Yet the model didn't rate them as a team drastically on the improve for 2022 still.

That’s exactly what the discussion is about.

No idea why you've got your knickers in a knot about it. Squiggle models Hawthorn as 0.2 wins below the aggregate of all the models, it just rates other teams as more likely to improve than Hawthorn so the ladder position slides.

There’s no knickers in a twist, this is literally the thread to discuss the squiggle and I’m doing just that.

 

[faible]

That’s exactly what the discussion is about.


There’s no knickers in a twist, this is literally the thread to discuss the squiggle and I’m doing just that.

And you seem upset that Squiggle hasn't accounted for the narrative you're spinning in the model. You're arguing that percentage isn't a valid indicator.

You're not 'discussing' anything. FS has answered why the model is saying what it's saying, that other teams around Hawthorn have shown a greater improvement over the off-season than Hawthorn has, and that percentage is generally indicative of a teams future performance.

You're trying to spin a narrative as to why Hawthorn will be the exception, and maybe they will, it happens.

 

[Strapping Young Lad]

And you seem upset that Squiggle hasn't accounted for the narrative you're spinning in the model. You're arguing that percentage isn't a valid indicator.

You're not 'discussing' anything. FS has answered why the model is saying what it's saying, that other teams around Hawthorn have shown a greater improvement over the off-season than Hawthorn has, and that percentage is generally indicative of a teams future performance.

You're trying to spin a narrative as to why Hawthorn will be the exception, and maybe they will, it happens.

No, I was agreeing with FS that there are factors that the model does not consider at all.

What I’m discussing is that the model has not been able to capture Hawthorn’s end of year output nor has it carried it through to 2022 modeling, while it seems the other models have, even if only marginally.

 

[Pessimistic]

Turnovers is defensive based game

 

[faible]

No, I was agreeing with FS that there are factors that the model does not consider at all.

What I’m discussing is that the model has not been able to capture Hawthorn’s end of year output nor has it carried it through to 2022 modeling, while it seems the other models have, even if only marginally.

Squiggle rates other sides as improving more over the off-season than Hawthorn. That will naturally have an impact on Hawthorn's modelled performance in Squiggle as compared to others.

You're simply taking issue with the outcome.

 

[Strapping Young Lad]

Squiggle rates other sides as improving more over the off-season than Hawthorn. That will naturally have an impact on Hawthorn's modelled performance in Squiggle as compared to others.

You're simply taking issue with the outcome.

No, again, I’ve been discussing the trend at the back end of 2022 when we performed our best without many of our best players, and that their return doesn’t impact the modelling.

 

[faible]

No, again, I’ve been discussing the trend at the back end of 2022 when we performed our best without many of our best players, and that their return doesn’t impact the modelling.

It does. FS said as much. Just that it impacts it less than the offseason additions of other sides around Hawthorn.

... which is to say, their Round 1 team this year is expected to be better than their Round 23 team from last year, but not by much. In fact, because practically all teams improve over the off-season as they regain players from injury - often by a lot - a 284-pt gain is pretty anemic: the 3rd worst in the league.

And for context as to who the other sides with similar list gains were;

Often the teams who have poor off-seasons were playing finals - they don't have much scope to add more talent for Round 1 because they were sending out their best teams late in the previous season, and didn't get draft picks to spend on recruitment. So the other bottom four teams are Melbourne, Port Adelaide, and Brisbane. It's a bit worrisome that after finishing 14th, Hawthorn have so little list improvement to show as they head into 2022.

 

[Strapping Young Lad]

It does. FS said as much. Just that it impacts it less than the offseason additions of other sides around Hawthorn.



And for context as to who the other sides with similar list gains were;

That was the point regarding O’Brien, Burgoyne and Ceglar leaving, who every man and his dog on here would have said was no great loss, cooked and pretty average, but the squiggle thinks those players plus Cousins are not countered strongly enough by the ins, because the modeling doesn’t really rate Gunston or Sicily due to missing the nest of 12 months of football.

 

[faible]

That was the point regarding O’Brien, Burgoyne and Ceglar leaving, who every man and his dog on here would have said was no great loss, cooked and pretty average, but the squiggle thinks those players plus Cousins are not countered strongly enough by the ins, because the modeling doesn’t really rate Gunston or Sicily due to missing the nest of 12 months of football.

So your subjective view is that they’re more important than the statistics value them as. Well done. Squiggle can only value what it can value. It has valued Hawthorn’s improvement as lesser than the improvement of other teams, thus sliding down the ladder more than other models.

Your issue appears to be you want squiggle to have some kind of subjective criteria.

 

[Pessimistic]

Picking around 9 wins seems about right for the hawks. I think there’d be a variable of plus or minus 2

 

[AFLGlicko]

Picking around 9 wins seems about right for the hawks. I think there’d be a variable of plus or minus 2

Far more variance than +/- 2 wins.

 

[Strapping Young Lad]

So your subjective view is that they’re more important than the statistics value them as. Well done. Squiggle can only value what it can value. It has valued Hawthorn’s improvement as lesser than the improvement of other teams, thus sliding down the ladder more than other models.

Your issue appears to be you want squiggle to have some kind of subjective criteria.

My subjective view would be that the addition of Gunston, Sicily, Day, Impey, Jiath, Moore, Bruest into the side is of far greater significance than the loss of Ceglar, Burgoyne, O'Brien and Cousins, but the squiggle doesn't value Gunston or Sicily.

Ask anyone if they'd take those 4 players over the returning 6 and you'd get crickets.

I don't want the squiggle to have subjective criteria, but that doesn't mean that the modelling has to go unquestioned.

 

[Crankyhawk]

No, again, I’ve been discussing the trend at the back end of 2022 when we performed our best without many of our best players, and that their return doesn’t impact the modelling.

on FS numbers, it has Gunstons return being only 70% as good a player as Tim O brien leaving. (150 vs 200 pts)
My guess is that Gunstons points value has dropped due to not getting on the park for the last 2 years.

 

[Pessimistic]

Far more variance than +/- 2 wins.

On that ladder it would put a team between 7th and 17th

 

[AFLGlicko]

On that ladder it would put a team between 7th and 17th

It's a mean ladder. 11 wins wouldn't get you 7th. You are conflating the two.

 

[Final Siren]

on FS numbers, it has Gunstons return being only 70% as good a player as Tim O brien leaving. (150 vs 200 pts)
My guess is that Gunstons points value has dropped due to not getting on the park for the last 2 years.

I just use AFL Player Ratings Pts - I don't have my own system. And yes, players do lose ratings pts if they haven't played for a while.

 

[Final Siren]

It's a mean ladder. 11 wins wouldn't get you 7th. You are conflating the two.

Yeah, you have to be a bit careful about what those predicted ladders are actually saying. There is some explanation on AFL Projected Ladder - Squiggle under "About Ladder Projections," but one common issue is that all teams' predicted wins drift towards the middle, with top teams showing surprisingly few wins and the bottom teams showing surprisingly few losses, as if the season is expected to be very close.

That's actually mostly about uncertainty - in reality, there definitely will be teams who get way out in front, and other teams where the wheels fall off early and stay off. But the models can't be sure which teams they are, so they tend to hedge around the middle.

 

[AFLGlicko]

Yeah, you have to be a bit careful about what those predicted ladders are actually saying. There is some explanation on AFL Projected Ladder - Squiggle under "About Ladder Projections," but one common issue is that all teams' predicted wins drift towards the middle, with top teams showing surprisingly few wins and the bottom teams showing surprisingly few losses, as if the season is expected to be very close.

That's actually mostly about uncertainty - in reality, there definitely will be teams who get way out in front, and other teams where the wheels fall off early and stay off. But the models can't be sure which teams they are, so they tend to hedge around the middle.

"Most early-season projections will predict an unusually close year, tipping too few wins for the top team and too many for the wooden spooner. This isn't an error; it's because "How many wins will the top team have, whoever they turn out to be?" is actually a different question to "How many wins will Geelong have?" — even if we expect Geelong to finish on top!"

 

[tiger_tough]

You're going to hate me, but Richmond:



So far -- can change with new injuries & I haven't done praccy match results yet -- off-season improvement looks roughly like this:

Tons: Richmond, Carlton
Plenty: West Coast, Fremantle, Collingwood, Gold Coast, Adelaide
Some: St Kilda, Essendon, Geelong, GWS, Bulldogs, Sydney, Nth Melb
Not much: Brisbane, Hawthorn, Port Adelaide, Melbourne

Thanks for that. Can I ask how are the points values for players calculated? Aarts rated better than Broad let alone Prestia gave me a chuckle.

 

[Mantastic]

Thanks for that. Can I ask how are the points values for players calculated? Aarts rated better than Broad let alone Prestia gave me a chuckle.

Look 3 posts above yours.

 

[tiger_tough]

Look 3 posts above yours.

Cheers. I posted the reply without reading the last page.

Still, it's an odd rating. Aarts is the definition of average.

 

[darcytiger]

My subjective view would be that the addition of Gunston, Sicily, Day, Impey, Jiath, Moore, Bruest into the side is of far greater significance than the loss of Ceglar, Burgoyne, O'Brien and Cousins, but the squiggle doesn't value Gunston or Sicily.

Ask anyone if they'd take those 4 players over the returning 6 and you'd get crickets.

I don't want the squiggle to have subjective criteria, but that doesn't mean that the modelling has to go unquestioned.

The hawks stink now mate, get used to it.