Certified Legendary Thread Race for the flag, in squiggly lines

[Gralin]

Another question for relative noobs - I saw above a mention of running at 72%. Is there a way to know what it runs at when lower placed teams get up (i.e. an upset)?

Running at 72% is getting it right 72% of the time regardless if who the public favorite is.

 

[iluvparis]

Example!

Not even the Freo hating squiggle could have predicted just how bad they would be!

 

[barmy44]

Another question for relative noobs - I saw above a mention of running at 72%. Is there a way to know what it runs at when lower placed teams get up (i.e. an upset)?

lower placed teams do cause upsets, so i would say it runs at about 72%...

 

[tommyk72]

Running at 72% is getting it right 72% of the time regardless if who the public favorite is.

Thanks - but that's not what I mean. In footy you have two teams play - one is higher placed. The rule of thumb is that the higher placed team on the ladder, more often than not, wins (oversimplification, but you get my point).

When the squiggle does its prediction, it occasionally suggests that a lower placed team will beat a higher placed team. I'm talking about this scenario. Of the occasions that the squiggle picks the lower placed team, what proportion does it get right? 72% is the average rate of ALL games (regardless of the ladder position).

Make sense?

 

[tommyk72]

lower placed teams do cause upsets, so i would say it runs at about 72%...

See my explanation - maybe I wasn't clear.

 

[Gralin]

Thanks - but that's not what I mean. In footy you have two teams play - one is higher placed. The rule of thumb is that the higher placed team on the ladder, more often than not, wins (oversimplification, but you get my point).

When the squiggle does its prediction, it occasionally suggests that a lower placed team will beat a higher placed team. I'm talking about this scenario. Of the occasions that the squiggle picks the lower placed team, what proportion does it get right? 72% is the average rate of ALL games (regardless of the ladder position).

Make sense?

Well ladder position is often not a true reflection of form or team strength so are you asking about tipping upsets vs favorites or just tipping based on ladder position because they aren't the same

 

[tommyk72]

Well ladder position is often not a true reflection of form or team strength so are you asking about tipping upsets vs favorites or just tipping based on ladder position because they aren't the same

Fair call. I guess I'm talking upsets based on favouritism. I'm defining that as ladder position, but I suppose you could define it as betting odds as well. You can't really define it on who the squiggle says will win, as then we come back to your earlier point - it'll be 28%!

On whether I can rely on this to help me tip upsets in betting/tipping comps. I'd like an objective way of measuring how often it gets the upsets right.

 

[Gralin]

Fair call. I guess I'm talking upsets based on favouritism. I'm defining that as ladder position, but I suppose you could define it as betting odds as well. You can't really define it on who the squiggle says will win, as then we come back to your earlier point - it'll be 28%!

On whether I can rely on this to help me tip upsets in betting/tipping comps. I'd like an objective way of measuring how often it gets the upsets right.

as an example, GWS vs Adelaide which is 2nd vs 6th. In NSW tip GWS in SA tip Adelaide. You have to look a bit further than ladder position to pick the favorite for the game.

I think Final Siren may have posted something previously on the Squiggle record vs upsets or at least the teams the Squiggle had the most trouble with, worth looking up if you are interested

 

[Final Siren]

Not even the Freo hating squiggle could have predicted just how bad they would be!

That's right! Turns out the squiggle actually loved Freo, relative to where they're going to finish.

 

[Suma Magic]

I remember a comment from a Freo fan a couple of weeks before the season started.

Something about them looking forward to another season of the squiggle not understanding Freo.

Turns out they were right, sort of...

 

[Bloody]

I remember a comment from a Freo fan a couple of weeks before the season started.

Something about them looking forward to another season of the squiggle not understanding Freo.

Turns out they were right, sort of...

Funny how he disappeared once first few rounds had taken place.


Sent from my iPhone using Tapatalk

 

[beert]

On whether I can rely on this to help me tip upsets in betting/tipping comps. I'd like an objective way of measuring how often it gets the upsets right.

The was a post earlier in the year that compared the squiggle with other strategies (home teams, higher on ladder, bookies favorites) over a number of years.

The bookies favourite was a pretty clear winner and the squiggle was generally second.

The squiggle is not particularly accurate - there are better systems out there. What makes the squiggle great is

a) it is purely factual - it relies only on data which everyone agrees on.

(b) it is transparent - it is easy to see and understand how it works, anyone could reproduce it (

c) given the simplicity, it is surprisingly accurate.

 

[Final Siren]

How does it look if you run it in order of teams likely to finish highest (squiggle ladder)? e.g. Hawks on left, then Cats, then Crows, etc...

I like the current version, just curious.

It is definitely more orderly!

 

[MC_9]

It is definitely more orderly!

I like it.

 

[Suntory Boss Coffee]

I won't be happy until hawthorn can only finish top 2

 

[Headless]

I won't be happy until hawthorn can only finish top 2

Yeah same.

... far out.

 

[Mantastic]

It is definitely more orderly!

I think you got yourself the winner there, mate!

 

[Sirlemsip]

It is definitely more orderly!

The other thing I want to suggest is how deep vertically the club logos are compared to the tower.

Surely it will be a bit easier to read if you increase the size of the logos so that the entire tower is exactly 18 logos deep?

For reference it is currently 25 logos deep.

 

[Final Siren]

The other thing I want to suggest is how deep vertically the club logos are compared to the tower.

Surely it will be a bit easier to read if you increase the size of the logos so that the entire tower is exactly 18 logos deep?

For reference it is currently 25 logos deep.

Heh, yes, I actually did that already, but didn't post it because I thought no-one wants to see 15 different Tower of Power versions as I fumble my way toward a good version. Also I'm supposed to be working today, not playing with logos. But it does make it easier to read:

I realize the logos don't all line up vertically for some reason yet.

 

[Dirty Bird]

The was a post earlier in the year that compared the squiggle with other strategies (home teams, higher on ladder, bookies favorites) over a number of years.

The bookies favourite was a pretty clear winner and the squiggle was generally second.

The squiggle is not particularly accurate - there are better systems out there. What makes the squiggle great is

a) it is purely factual - it relies only on data which everyone agrees on.

(b) it is transparent - it is easy to see and understand how it works, anyone could reproduce it (

c) given the simplicity, it is surprisingly accurate.

On these 2 points.
Do people here think that the success of the squiggle had something to do with how subjective Roby's rankings were.
If that was actually any good would people have been hooked on this alternative so fast?

 

[Mningasimisonda]

On these 2 points.
Do people here think that the success of the squiggle had something to do with how subjective Roby's rankings were.
If that was actually any good would people have been hooked on this alternative so fast?

I think a lot of the attraction to The Squiggle and Final Siren's other predictors like the flagpole and the Tower of Power is the visualisations. Obviously, no one would care about them if they weren't good predictors, but the visualisations make them something awesome.

On the other hand, most of the attraction to Roby's predictions is the comedy and strange methods that are used. If Roby's predictions actually made sense and worked well, it would be boring.

You look at the squiggle to know what is going on, you look at Roby's thread for light entertainment.

 

[TheGhostOfJohnBelushi]

Huh, well there you go. They really spare no expense on this one.

You'd think they'd update the trophy's AFL font... that's the 1990s version! O.O

 

[Dirty Bird]

You'd think they'd update the trophy's AFL font... that's the 1990s version! O.O

i thought it was a mix of the 90s a and the 00s fl

 

[TheGhostOfJohnBelushi]

i thought it was a mix of the 90s a and the 00s fl

Nah, it's the A from the logo and the FL is just in a similar font to mimic the look. The current logo doesn't have those flattish portions.

 

[Simon_Nesbit]

Heh, yes, I actually did that already, but didn't post it because I thought no-one wants to see 15 different Tower of Power versions as I fumble my way toward a good version. Also I'm supposed to be working today, not playing with logos. But it does make it easier to read:

I realize the logos don't all line up vertically for some reason yet.

Could it be less of a "Tower" and more of a "block"? Making it wider will make it easier to interpret quickly.