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

[Yippy Yi Yeo]

No strawman. From me anyway. Here’s your original post:


A fair bit of whining about Squiggle there and claiming of specialness for your flag, including the spurious claim of injury to Gaff.
Bottom line: only Richmond has had an injury free run over the last few years. The West Coast injury toll was fairly normal.

You say I don’t have a clue because I think you’ve overrated Naitanui and underrated Lycett? Nic’s good at hitouts and a couple of contested marks a season, but he’s not a stats hound like Gawn and Grundy ( or Cox before him) and I don’t think he was ever used as a ruck tag like Lycett and Vardy were. We can’t tell what would have happened if the injured and suspended players played, so my guess is as good as yours.
And I’ll say it again, I think the Eagles are well placed with good depth and very good quality players to at least challenge again next year. Squiggle agrees, I believe.

The argument I was making was against people using the squiggle's ranking of our premiership side as an argument when concluding that WC is absolutely one of the weaker flag sides. Not against the squiggle itself, and not sure where I was whinging, nor did I claim it was special. More strawmen. Bottom line is you've still failed to point out a single side in the last 20 years with 3 AA quality players sitting on the sideline, if it was such a normal injury toll to a GF side then I would've thought you'd be able to point me in the direction of at least several. Weird.

I say you don't have a clue because you don't, Naitanui's clearance differential when I last saw it measured was only bested by Nat Fyfe, had the highest hitouts % won, and the highest % of hitouts to advantage of any ruckman. Lycett is a middle of the road ruck. Naitanui doesn't need to be a "stats hound" like Gawn or Grundy because he's a different type of ruckman, and it's bewildering that you still don't understand Naitanui's impact in 2018, after even the most anti-naitanui media commentators like Lyon admitted how wrong they were about him after he returned from injury this year. Obviously he was never used to tag like Vardy and Lycett, because he far more often than not won head to head ruck battles. That statement right there just exemplifies the clear lack of knowledge or understanding you have about West coast or Naitanui.

No we can't tell what would've happened, but I'm sure if we did a statistical analysis of teams with many injuries and teams without injuries, there would be a positive correlation between winning and having less injuries. The side we played against collingwood in the GF probably wins 50% of the time, you add Naitanui, Gaff and Sheppard to that mix and that probably goes up to 65 or 70%.

 

[Clems Knee]

The argument I was making was against people using the squiggle's ranking of our premiership side as an argument when concluding that WC is absolutely one of the weaker flag sides. Not against the squiggle itself, and not sure where I was whinging, nor did I claim it was special. More strawmen. Bottom line is you've still failed to point out a single side in the last 20 years with 3 AA quality players sitting on the sideline, if it was such a normal injury toll to a GF side then I would've thought you'd be able to point me in the direction of at least several. Weird.

I say you don't have a clue because you don't, Naitanui's clearance differential when I last saw it measured was only bested by Nat Fyfe, had the highest hitouts % won, and the highest % of hitouts to advantage of any ruckman. Lycett is a middle of the road ruck. Naitanui doesn't need to be a "stats hound" like Gawn or Grundy because he's a different type of ruckman, and it's bewildering that you still don't understand Naitanui's impact in 2018, after even the most anti-naitanui media commentators like Lyon admitted how wrong they were about him after he returned from injury this year. Obviously he was never used to tag like Vardy and Lycett, because he far more often than not won head to head ruck battles. That statement right there just exemplifies the clear lack of knowledge or understanding you have about West coast or Naitanui.

No we can't tell what would've happened, but I'm sure if we did a statistical analysis of teams with many injuries and teams without injuries, there would be a positive correlation between winning and having less injuries. The side we played against collingwood in the GF probably wins 50% of the time, you add Naitanui, Gaff and Sheppard to that mix and that probably goes up to 65 or 70%.

You keep using the term “strawman”. I’m now convinced you don’t know what it means.
“Arguing against people using the squiggle’s ranking of our premiership as an argument when concluding that WC is absolutely one of the weaker sides” is a strawman because people weren’t making that conclusion. And if they did, talking about the limited usefulness of the Squiggle is not the argument you should have made.

You certainly claimed it was special, carrying on about AA quality players missing from the final series. That’s not a strawman, because I addressed a point you made.

Go ahead. Do a statistical study of WC games with and without Naitanui. Don’t just make numbers up. I reckon you’d be surprised by the lack of difference. Maybe next year without Lycett the difference will be greater.

 

[arfadunger]

Haven't visited this thread for a while, what happened. I always thought the squiggle was just a bit of fun, obviously some take it more seriously than others.

 

[iluvparis]

Haven't visited this thread for a while, what happened. I always thought the squiggle was just a bit of fun, obviously some take it more seriously than others.

YYY added to the list of multitude of posters that doesn't rate the squiggle because it doesn't take into account the 'subjective' factors that would clearly show his own team as the GOAT flag side.

 

[nameless]

Yippy Yi Letitgo

 

[Paracleet]

The squiggle is a measure of probability and when the 40% outcome happens the squiggle is 'wrong' or 'underrating a team'? No. Weaker teams win sometimes, it's called chance. If chance didn't exist then why bother to play at all?

 

[RookiePick]

I don't really see where YYY was having a crack at the squiggle. He just pointed out the limitations and how that leads him to put his own weightings on the likelihood of the squiggle predictions holding for a particular match.

Isn't that what everyone does?

 

[the harry]

I don't really see where YYY was having a crack at the squiggle. He just pointed out the limitations and how that leads him to put his own weightings on the likelihood of the squiggle predictions holding for a particular match.

Isn't that what everyone does?

Yeah. But on the whole squiggle is probably better than them, especially for effort put in once all the ground work is done. Additionally, the knowledge of 17 other clubs is probably a flaw in most footy supporters.

The idea that it can calculate a probable outcome for every match of next season is very impressive.

 

[Yippy Yi Yeo]

YYY added to the list of multitude of posters that doesn't rate the squiggle because it doesn't take into account the 'subjective' factors that would clearly show his own team as the GOAT flag side.

Comprehension not your strong point I see. Never said I don't rate the squiggle, I just don't think it should be used in isolation if you're going to make a prediction on something. The reason the squiggle isn't 100% accurate is because it can't account for subjective factors by its very nature. Are you telling me you'd tip a top side to beat another top side because the squiggle said so even if 5 of their top players were going to miss the game? I don't think so.

I don't really see where YYY was having a crack at the squiggle. He just pointed out the limitations and how that leads him to put his own weightings on the likelihood of the squiggle predictions holding for a particular match.

Isn't that what everyone does?

Good to see someone learnt a bit of reading comprehension at school, I thought that was just a bit of common sense.

 

[Yippy Yi Yeo]

You keep using the term “strawman”. I’m now convinced you don’t know what it means.
“Arguing against people using the squiggle’s ranking of our premiership as an argument when concluding that WC is absolutely one of the weaker sides” is a strawman because people weren’t making that conclusion. And if they did, talking about the limited usefulness of the Squiggle is not the argument you should have made.

You certainly claimed it was special, carrying on about AA quality players missing from the final series. That’s not a strawman, because I addressed a point you made.

Go ahead. Do a statistical study of WC games with and without Naitanui. Don’t just make numbers up. I reckon you’d be surprised by the lack of difference. Maybe next year without Lycett the difference will be greater.

You told me I think that Gaff was the difference between a close loss and a slaughter, which I never said. You told me I think the squiggle doesn't respect WC, which I never said. It's an analytical tool, there's no respect involved, however it can underrate a team due to the subjective variables which the model doesn't take into account, which was clearly the case all year given that it didn't even rate us as a top 8 team at the start of the year. I didn't claim the flag was special, another strawman from you. I said we had more high quality talent missing from our GF side than most flag winning teams do, which I stand by. I then asked you in a few different posts if you could name a few examples of flag sides with more talent on the sidelines than WC 2018 and you failed to name a single one, while repeatedly claiming that the amount of talent we had sitting out was normal for flag teams.

My original post was in response to a WC supporter's comment who was using a bit of sarcasm saying the squiggle has vic bias. The only part where I referred to anyone else in the post was this:

"Squiggle doesn't take into account the impact of injuries, which everyone seems to overlook."

That's not a strawman because I'm not misrepresenting any particular persons' argument, nor am I claiming someone said something that they didn't. Do you understand now?

Just for the record, since Naitanui's been our primary ruckman at the start of 2015 our win-loss with him is 38-14, or 73%. Our win-loss without him is 28-16, or 63%. If you want to go back a bit further to his AA year of 2012 onwards the win loss with him is 69-36, or 66%, and without him is 33-26, or 56%.

I'd say that a 10% difference over 164 games is probably pretty statistically significant, wouldn't you?

 

[JZTIME!]

I think its pretty obvious with the evenness of the competition that its going to be rare to have a side dominate to an extent they match the power teams of years gone by. IF we revisit in 10 years I would think that there will be more premiership sides closer to the last couple of years. It is just that even now.

 

[Yippy Yi Yeo]

I think its pretty obvious with the evenness of the competition that its going to be rare to have a side dominate to an extent they match the power teams of years gone by. IF we revisit in 10 years I would think that there will be more premiership sides closer to the last couple of years. It is just that even now.

Definitely think this will be the case. Talent spread over 18 teams makes it harder to get the level of talent necessary, plus greater freedom of movement that's come with FA and trades makes retaining enough depth in an elite side much harder.

 

[inbetween]

You told me I think that Gaff was the difference between a close loss and a slaughter, which I never said. You told me I think the squiggle doesn't respect WC, which I never said. It's an analytical tool, there's no respect involved, however it can underrate a team due to the subjective variables which the model doesn't take into account, which was clearly the case all year given that it didn't even rate us as a top 8 team at the start of the year. I didn't claim the flag was special, another strawman from you. I said we had more high quality talent missing from our GF side than most flag winning teams do, which I stand by. I then asked you in a few different posts if you could name a few examples of flag sides with more talent on the sidelines than WC 2018 and you failed to name a single one, while repeatedly claiming that the amount of talent we had sitting out was normal for flag teams.

My original post was in response to a WC supporter's comment who was using a bit of sarcasm saying the squiggle has vic bias. The only part where I referred to anyone else in the post was this:

"Squiggle doesn't take into account the impact of injuries, which everyone seems to overlook."

Like you said; squiggle by design does not factor in the subjective 'variables'. It cant even truly under-rate anything since it has no opinion. It is a mathemetical function and a relatively simple one at that. It uses only a few data inputs as metrics to create a fairly straightforward ratings system.

As with most systems ; these inputs are based on past data and results. Basically the rating systems always lag behind actual results; they're reflective by nature, not predictive. Squiggle is good because it deals only in tangibles.

Squiggle didnt rate us top 8 at the start of the season because it went by results of previous seasons. It did the same for Richmond of the previous season. Its not wrong. It cant predict how a team will improve; it will only reflect that it has improved.

So many people who fail to grasp these concepts. If you want nonsensical ranking systems which account for subjective data like injuries; a quick search for robys power rankings will have you appreciate the squiggle more.

 

[the harry]

Despite it not predicting, it is better at predicting than an awful lot of people

 

[Clems Knee]

You told me I think that Gaff was the difference between a close loss and a slaughter, which I never said. You told me I think the squiggle doesn't respect WC, which I never said. It's an analytical tool, there's no respect involved, however it can underrate a team due to the subjective variables which the model doesn't take into account, which was clearly the case all year given that it didn't even rate us as a top 8 team at the start of the year. I didn't claim the flag was special, another strawman from you. I said we had more high quality talent missing from our GF side than most flag winning teams do, which I stand by. I then asked you in a few different posts if you could name a few examples of flag sides with more talent on the sidelines than WC 2018 and you failed to name a single one, while repeatedly claiming that the amount of talent we had sitting out was normal for flag teams.

My original post was in response to a WC supporter's comment who was using a bit of sarcasm saying the squiggle has vic bias. The only part where I referred to anyone else in the post was this:

"Squiggle doesn't take into account the impact of injuries, which everyone seems to overlook."

That's not a strawman because I'm not misrepresenting any particular persons' argument, nor am I claiming someone said something that they didn't. Do you understand now?

Just for the record, since Naitanui's been our primary ruckman at the start of 2015 our win-loss with him is 38-14, or 73%. Our win-loss without him is 28-16, or 63%. If you want to go back a bit further to his AA year of 2012 onwards the win loss with him is 69-36, or 66%, and without him is 33-26, or 56%.

I'd say that a 10% difference over 164 games is probably pretty statistically significant, wouldn't you?

You’ve lied about what I’ve said. Check what I wrote and retract if you are capable of reading. This argument is tedious. I’m done.

 

[iluvparis]

Despite it not predicting, it is better at predicting than an awful lot of people

Yes and people tend to think that the 'subjective' things that squiggle overlooks (ZOMFG we had no Gaff or Nic Nat!!!!) are a lot more relevant than they actually are.

The best thing about squiggle is that it shows you most of the time you can ignore all the subjective bs

 

[Final Siren]

Squiggle doesn't take into account the impact of injuries, which everyone seems to overlook. I dare say we were also one of the most injury riddled premiership sides in the last 20 years as well, and not just throughout the season which impacted our win/loss and %, but in finals as well. I can't think of many teams missing 3 players the quality of gaff, naitanui and shep throughout the entire finals series who went all the way.

Jk and Ryan missed half a season, Shuey, Darling and Barrass missed a quarter, Naitanui missed 10 games, Gaff missed 6, Shep missed 3, Jetta missed 5 and Mackenzie missed the whole season.

Considering we still finished 2nd and won the flag despite all of this, I'd say we were one of the stronger premiership sides of the last 20 years. Imagine if we'd had Richmond's run of injuries, or lack thereof.

Squiggle did actually become player-aware halfway through this year, i.e. it now does account for injuries.

But teams' chart positions don't describe how strong they would have been with no injuries; they describe how strong the team actually was. So in that sense, the Eagles are indeed rated lower than they would have been with no injuries. (Same as with any team.)

Squiggle hasn't made a 2019 prediction yet because the fixture came out while I was on holidays, but once I've had a chance to run it through, it will account for personnel changes, including returns from injury.

 

[Art OF war]

Squiggle did actually become player-aware halfway through this year, i.e. it now does account for injuries.

But teams' chart positions don't describe how strong they would have been with no injuries; they describe how strong the team actually was. So in that sense, the Eagles are indeed rated lower than they would have been with no injuries. (Same as with any team.)

Squiggle hasn't made a 2019 prediction yet because the fixture came out while I was on holidays, but once I've had a chance to run it through, it will account for personnel changes, including returns from injury.

How can it account for injuries objectively though? Wouldn't it be better without this?

 

[eaglespremiers]

As for losing Naitanui, I think West Coast played better as a team afterwards. Their premiership winning tactic of tagging the dominant ruckman only developed when Vardy and Lycett teamed up.

But that too can be attributed much to Naitanui's tactical abilities on the sidelines as he was our ruck coach for the last quarter of the year.

 

[HTT]

How can it account for injuries objectively though? Wouldn't it be better without this?

From memory it uses player ratings, which isn't perfect, but achieved better results than previous models so seems to hold some merit.

 

[Final Siren]

How can it account for injuries objectively though? Wouldn't it be better without this?

As harrythetiger says, it uses AFL Player Ratings to approximate the value of an individual player. This is used to adjust a team's rating based on its Ins/Outs.

For example, if it's Richmond v Collingwood, but the Tigers have lost Martin this week and replaced him with a 4-gamer, then Squiggle will rate Collingwood as more likely to win (or to win by more).

And Collingwood's chart movement will incorporate Squiggle's knowledge that Richmond was Martin-less, so the Pies will get less positive movement (or more negative movement) than they would have otherwise.

 

[The Gov]

You’ve lied about what I’ve said. Check what I wrote and retract if you are capable of reading. This argument is tedious. I’m done.

To be fair you asked him to go through games with and without Naitanui and said the difference wouldn't be much, he did and it was about 10% difference which is fairly significant. Not sure what would have happened in the GF, maybe we would have lost if NN had played, who knows, but on statistics alone we would have been more likely.

 

[Engimal v3]

You’ve lied about what I’ve said. Check what I wrote and retract if you are capable of reading. This argument is tedious. I’m done.

Interesting. Because you said:

Go ahead. Do a statistical study of WC games with and without Naitanui. Don’t just make numbers up. I reckon you’d be surprised by the lack of difference.

And then he said:

Just for the record, since Naitanui's been our primary ruckman at the start of 2015 our win-loss with him is 38-14, or 73%. Our win-loss without him is 28-16, or 63%. If you want to go back a bit further to his AA year of 2012 onwards the win loss with him is 69-36, or 66%, and without him is 33-26, or 56%.

I'd say that a 10% difference over 164 games is probably pretty statistically significant, wouldn't you?

Which is exactly what you asked him to do.

 

[Clems Knee]

Interesting. Because you said:


And then he said:



Which is exactly what you asked him to do.

That’s not what he lied about.

 

[Yippy Yi Yeo]

That’s not what he lied about.