Certified Legendary Thread Race for the flag, in squiggly lines

Remove this Banner Ad

I'm back! What did I miss?
flame-war.gif
 

Log in to remove this ad.

Simple case of the squiggle rating the bulldogs purely on results, not knowing that everytime the dogs got a key player or two back from injury, they lost another couple, so the squiggle thought that was their level, which was rated OK, around abouts, it has no way of knowing the calibar of players that became available for the finals. 5 changes, Stringer - Libba - Macrae - Roughy - Wood. You bring those 5 into the bottom team on the ladder and they will be twice the team they were. It was also the first time in a long while that the dogs have had their best possible defence since murphy went down.

I think i said it on the bulldog board, but the squiggle needed to be adjusted for those inclusions to be closer to reality, which would be to add 20pts to the dogs and remove 20pts from the opponent for any predicted result. Averages for the finals in regard to points scored and points against for the dogs suggest similar in comparison to H&A season also.

otherwise *insert meme of bulldog taking a turd on the squiggle*
 

(Log in to remove this ad.)

If the dogs win by a small margin, will we be the least liked team by the squiggle to ever win a flag, Max?
Yep, in the last 20 years at least.

Not only that, they will have overcome a pretty good field! Sydney 2005, Adelaide 1997, and Geelong 2009 won flags from relatively poor chart positions, but there weren't too many teams in those years who placed any better. This year, although it's been even, I think there are more teams in better spots.
 
Simple case of the squiggle rating the bulldogs purely on results, not knowing that everytime the dogs got a key player or two back from injury, they lost another couple, so the squiggle thought that was their level, which was rated OK, around abouts, it has no way of knowing the calibar of players that became available for the finals. 5 changes, Stringer - Libba - Macrae - Roughy - Wood. You bring those 5 into the bottom team on the ladder and they will be twice the team they were. It was also the first time in a long while that the dogs have had their best possible defence since murphy went down.

I think i said it on the bulldog board, but the squiggle needed to be adjusted for those inclusions to be closer to reality, which would be to add 20pts to the dogs and remove 20pts from the opponent for any predicted result. Averages for the finals in regard to points scored and points against for the dogs suggest similar in comparison to H&A season also.

otherwise *insert meme of bulldog taking a turd on the squiggle*

That's one of the primary features of the squiggle that it cannot be swayed by team selection. So many times you sit of the fence when tipping a winner and swayed by an important inclusion. Quite often it makes no difference.
 
Grand Final 2016


L8dsLyz.jpg

So let's get this out of the way:

Sydney.png
Sydney 90 - 52 Western Bulldogs
Western%20Bulldogs.png

The squiggle thinks Sydney will win. And not by a little! By a lot. That's a 38-point margin, the biggest tip in a GF since 2007, when the Cats lined up against Port Adelaide. It puts the chance of a Swans win at a whopping 83.9%.

But hopefully that won't happen. The huge margin, at least. That would be dull. And sad.

Both teams this year are in unusual chart positions. Especially Sydney, who as they sit today are the third-most defensively successful team of the last 20 years. (They are currently beaten out, barely, by Collingwood 2011 and Adelaide 2005.) If the Swans win, and it's not some kind of 1989-style shootout, then there will be three notably-defensive cups on the chart, and they'll all belong to Sydney.

The Dogs are also oddly far from past premier positions for a Grand Finalist. But in an objectively bad way, not just lacking balance. Their last three weeks have been outstanding, of course, but they finished the home & away season ranked as a mere mid-tier team. This is reflected in their percentage of 115%: The Dogs did not put a lot of teams away. By way of comparison, Hawthorn's percentage was 119%, the next best was West Coast with 130%, and then there are three more teams in Adelaide, GWS, and Geelong before you get to Sydney on 151%. The Dogs' percentage is even worse than the most logic-defying premier of the last two decades, Sydney 2005 (116%).

So it's hard to make a case for the Dogs, at least going purely by score history, which is what the squiggle does. Strange things happen in footy, but a Dogs premiership would be one of the strangest.

About the only thing I can think of is that Sydney are too defensive, compared to past premiers. There is a reasonably solid history of teams in this area being exploited by more balanced teams. So if the Giants were in the GF, the Swans may have been in trouble. But the Dogs don't look like that balanced team. They are defensively-oriented themselves, but more importantly just weaker across the board.

You could also argue that the Dogs are a different team now than they were before finals. They certainly look like it. But that sounds like the kind of argument everyone makes about their team, and is almost always wrong. West Coast looked pretty amazing for three weeks recently, too. And if we don't believe that past results can predict future performance, there isn't much we can say.

Squiggle's Flagpole algorithm, which is tuned to rate teams based on how much they resemble past premiers, cares more about attacking power, ignores home ground advantage, and places a premium on recent results. Only the last is in the Dogs' favour, though, so they're still rated a fair way off the pace.

AypscJU.jpg

Personally, like everyone else, I'll be barracking hard for the Dogs tomorrow. Nothing against the Swans. But come on. How incredible would it be.

Quick note: If you don't come back to this thread after the GF, because, you know, footy is DONE, then thanks for reading. It's been fun.

Also! This year I found a bunch of great amateur AFL analysts and stats people writing on the web. Some have really interesting GF pieces up, which analyze the teams in a lot more detail and with much deeper data than I have.

I especially recommend Figuring Footy, which has a 3-part series on the GF and a bunch of cool charts centred around the idea of analyzing how teams generate shots at goal.

Hurling People Now have also been analyzing the crap out of the finals series, and I assume a GF piece is coming.

There is also Footy Maths, Plus Six One, Matter of Stats and The Arc, plus a bunch more on Twitter.

Live squiggling!
 
Grand Final 2016


L8dsLyz.jpg

So let's get this out of the way:

Sydney.png
Sydney 90 - 52 Western Bulldogs
Western%20Bulldogs.png

The squiggle thinks Sydney will win. And not by a little! By a lot. That's a 38-point margin, the biggest tip in a GF since 2007, when the Cats lined up against Port Adelaide. It puts the chance of a Swans win at a whopping 83.9%.

But hopefully that won't happen. The huge margin, at least. That would be dull. And sad.

Both teams this year are in unusual chart positions. Especially Sydney, who as they sit today are the third-most defensively successful team of the last 20 years. (They are currently beaten out, barely, by Collingwood 2011 and Adelaide 2005.) If the Swans win, and it's not some kind of 1989-style shootout, then there will be three notably-defensive cups on the chart, and they'll all belong to Sydney.

The Dogs are also oddly far from past premier positions for a Grand Finalist. But in an objectively bad way, not just lacking balance. Their last three weeks have been outstanding, of course, but they finished the home & away season ranked as a mere mid-tier team. This is reflected in their percentage of 115%: The Dogs did not put a lot of teams away. By way of comparison, Hawthorn's percentage was 119%, the next best was West Coast with 130%, and then there are three more teams in Adelaide, GWS, and Geelong before you get to Sydney on 151%. The Dogs' percentage is even worse than the most logic-defying premier of the last two decades, Sydney 2005 (116%).

So it's hard to make a case for the Dogs, at least going purely by score history, which is what the squiggle does. Strange things happen in footy, but a Dogs premiership would be one of the strangest.

About the only thing I can think of is that Sydney are too defensive, compared to past premiers. There is a reasonably solid history of teams in this area being exploited by more balanced teams. So if the Giants were in the GF, the Swans may have been in trouble. But the Dogs don't look like that balanced team. They are defensively-oriented themselves, but more importantly just weaker across the board.

You could also argue that the Dogs are a different team now than they were before finals. They certainly look like it. But that sounds like the kind of argument everyone makes about their team, and is almost always wrong. West Coast looked pretty amazing for three weeks recently, too. And if we don't believe that past results can predict future performance, there isn't much we can say.

Squiggle's Flagpole algorithm, which is tuned to rate teams based on how much they resemble past premiers, cares more about attacking power, ignores home ground advantage, and places a premium on recent results. Only the last is in the Dogs' favour, though, so they're still rated a fair way off the pace.

AypscJU.jpg

Personally, like everyone else, I'll be barracking hard for the Dogs tomorrow. Nothing against the Swans. But come on. How incredible would it be.

Quick note: If you don't come back to this thread after the GF, because, you know, footy is DONE, then thanks for reading. It's been fun.

Also! This year I found a bunch of great amateur AFL analysts and stats people writing on the web. Some have really interesting GF pieces up, which analyze the teams in a lot more detail and with much deeper data than I have.

I especially recommend Figuring Footy, which has a 3-part series on the GF and a bunch of cool charts centred around the idea of analyzing how teams generate shots at goal.

Hurling People Now have also been analyzing the crap out of the finals series, and I assume a GF piece is coming.

There is also Footy Maths, Plus Six One, Matter of Stats and The Arc, plus a bunch more on Twitter.

Live squiggling!
Thanks so much for this season mate, easily my favourite thread on Big Footy. Have started spreading the gospel of squiggle this year to friends and family alike. Will definitely get around to reading some of your books over the summer as well! Take care, hope you keep squiggling far into the future - would love to watch the Lions squiggle their way back to the glory days!
 
yes, thanks Final Siren - another Queenslander here hoping my side the Suns will squiggle on up like we saw with GWS this season (I think we finished about the same place GWS started this year).
Enjoy your off-season.
and I hope you really enjoyed the rocks up in the NT - amazing really - who would have thought? rocks!
 

Remove this Banner Ad

Certified Legendary Thread Race for the flag, in squiggly lines

Remove this Banner Ad

Back
Top