Certified Legendary Thread Race for the flag, in squiggly lines

[cryptor]

No score predictions this week
I love beating the squiggle while other teams dissappoint.

ROUND 12 TIPS

Geelong 103 - 80 Carlton

Hawthorn 113 - 64 West Coast

Port Adelaide 123 - 56 St Kilda

Greater Western Sydney 75 - 105 Essendon

Western Bulldogs 98 - 79 Brisbane Lions

Gold Coast 72 - 102 Sydney

Fremantle 89 - 58 Adelaide

North Melbourne 84 - 66 Richmond

Melbourne 56 - 98 Collingwood

 

[Final Siren]

Awarding teams partial wins would result in a more realistic ladder.
But I would advise against rounding the final figure, and simply having the teams with a non-integer win tally. Though the ladder would have most, if not all, teams with a fractional number of wins, this way you can ensure that the total number of wins available in a season remains consistent with the number of games.

If you were to round, you could potentially increase/decrease the total number of wins (league-wide) to a figure that's actually greater/lower than the actual number of games played. A very basic example between a 3 team league to demonstrate this:

Team A (80%) v Team B (20%)
Team A (80%) v Team C (20%)
Team B (50%) v Team C (50%)

Team A: 0.8 + 0.8 = 1.6 wins = 2 wins (rounded)
Team B: 0.2 + 0.5 = 0.7 wins = 1 win (rounded)
Team C: 0.2 + 0.5 = 0.7 wins = 1 win (rounded)

As you can see there were only 3 games played but when rounding, the final ladder would ultimately award a total of 4 wins from the 3 games.
However, the actual ladder itself doesn't use decimal values and partial wins, but with the decimal values, the predictor will actually rank the teams in the correct order. i.e. a 16.9 win team is ranked above a 16.6 team.

This is all exactly right! And indeed it's what I did in an earlier pass. However, this method has a couple of drawbacks, too.

Most importantly, it would cause the predictor to ignore percentage. Imagine it's Round 22. Geelong and North Melbourne are both on 15 wins, but the Roos have a better percentage. So in Round 23, all North need to do to finish higher is beat Melbourne. Assuming that Melbourne don't improve quite a lot between now and then, the predictor should say that is going to happen. But if it ranked purely on likelihood of wins, as you describe, it would predict Geelong to finish higher, because they play Brisbane at home, and that win is a bit more likely than the Roos'. Which would be flat-out wrong, because the chances of Geelong winning while North lose - the situation required for Geelong to finish higher - are pretty small.

The other thing is I feel this is an appropriate point to collapse the probabilities down into hard wins or losses. The finals matches are done by tips, not probability distributions, so it's really happening anyway.

But you are certainly correct to observe that the predictor can calculate an impossible ladder, especially earlier in the year. For example, after a game it might bump up a team's predicted wins for the year by one, as they tip over from, say, a predicted 8.4 wins (rounded down to 8) to 8.6 (rounded up to 9)... without any other team's predicted wins changing! So where is that extra win supposed to come from? But I still think this is a better way to do it, since although the entire ladder may not be strictly possible, the prediction for each team within it is more likely to be right.

Anyway, I do plan to add a thing you can click to make the predictor show all the nitty-gritty details: every tip for the rest of the season plus the unrounded values. So you will have access to that data if you want it.

 

[Final Siren]

I imagine Final Siren's forecaster works the same.

Yep, exactly. If Hawthorn had to play its next 10 games against Essendon, I would tip the Hawks for each individual game, because they're a better team. But I wouldn't expect them to win all 10, because upsets happen. I just don't know when they'll happen.

Getting out the maths again... if it's 80% likely that the Hawks will beat Essendon, then for the Hawks to win all 10 games, an event that is 80% likely needs to happen 10 times out of 10. The odds of that are 0.8 x 0.8 x 0.8 x 0.8 x 0.8 x 0.8 x 0.8 x 0.8 x 0.8 x 0.8 = 0.107. Which is only 10.7%.

Another way to look at it is to imagine all the different pathways to a particular outcome. The only way to reach the outcome of Hawthorn going 10-0 is if every match ends in a Hawthorn win. By contrast, there are ten different ways for the Hawks to go 9-1: they could drop the first game, or else the second, or else the third, or else the fourth... all those different pathways add up to make it quite likely overall.

It's similar to how if you roll two dice, you're more likely to roll 7 than a 12. The only way to roll a 12 is to get a 6 on both dice. But there are lots of ways to roll a 7: 1 & 6 or 2 & 5 or 3 & 4 or 4 & 3 or 5 & 2 or 6 & 1.

 

[Seppo]

Freo predicted 89 - 58.
Actually score 90 - 50.

Bam.

 

[Heeney2Franklin]

Wow the score was almost spot on for our game.

 

[The Speaker]

Squiggle's been on fire this round.

 

[Shane Hird]

The squiggle is King.

 

[Hawkk]

Looks like the Sydney and Hawthorn squiggles are 'locked in' on their 2012 and 2013 flag positions...

 

[Pessimistic]

2012 sydney v 2013 hawthorn

Who wins ? What woudl franklin do

 

[Leather Poisoning]

Looks like the Sydney and Hawthorn squiggles are 'locked in' on their 2012 and 2013 flag positions...

We need Luke Darcy to tell us it would be a great story.

 

[WWSD]

We need Luke Darcy to tell us it would be a great story.

Say that again!

 

[Iron]

 

[The King!]

2012 sydney v 2013 hawthorn

Who wins ? What woudl franklin do

Whoever has spangher normally wins

 

[WWSD]

Whoever has spangher normally wins

That's true. He's currently 7-0 as a Hawk.

 

[Final Siren]

Tigers hit their predicted score like it was a brick wall.

Predicted: North 84 - 66 Richmond
Half-time: 31-66
15 minutes later: 82-66

 

[Final Siren]

Freo predicted 89 - 58.
Actually score 90 - 50.

Bam.

I was filthy at that umpire when it was 89-50 with two minutes to go and he denied an Adelaide mark in the goal square.

 

[Chunderboy]

I was filthy at that umpire when it was 89-50 with two minutes to go and he denied an Adelaide mark in the goal square.

Roby would have accounted for that and nailed it

 

[Final Siren]

Round 12, 2014​

Interactive squiggle

Not much movement this week, because the squiggle's tips were good.

The main mover was Collingwood, which gets credit for keeping Melbourne to three goals. Although not as much as you might expect, since Melbourne's attack is so weak.

What this does is open up even more of a gap between 7th and 8th. This will probably change, as one of the leading teams gets the wobbles, and/or someone makes a break from the middle. But right now, the squiggle reckons the difference between 7th and 8th is huge.

 

[SJ]

Pretty cool when teams like the Dogs and Blues barely move in over half a season. You know you've got a pretty good read on them.

 

[Roobs321]

Gold Coast has almost equaled West Coast on the squiggle, and they play this week. Perfect statement time for the Suns.

 

[Seppo]

Should see Freo fly up and to the right if its only predicting we beat Richmond 76-69.

 

[Epilegs]

Melbourne are drawing boobs

 

[MF]

Tells a pretty good story...

The Hawks and Swans both sitting around the mark where they snagged their last premierships. Geelong going backwards.

North and Port both sitting just outside the premiership strike zone with port needing to improve defensively while north need to improve offensively.

Freo doing circle work on the extreme defensive side of the chart where only Sydney have managed to win a premiership recently, where Ross Lyon was an assistant at the time.

 

[drummond2fevola]

Melbourne are drawing boobs

I reckon it will end up being a penis

 

[Leeda]

gosh I am totally guttered with that..

eyes above the horizon please