Gydafud
Premium Platinum
Probably the most negative movement I have seen Geelong have after a win.
-
Mobile App Discontinued
Due to a number of factors, support for the current BigFooty mobile app has been discontinued. Your BigFooty login will no longer work on the Tapatalk or the BigFooty App - which is based on Tapatalk.
Apologies for any inconvenience. We will try to find a replacement.
Certified Legendary Thread Squiggle 2017
- Thread starter Final Siren
- Start date
- Tagged users None
🥰 Love BigFooty? Join now for free.
Power Raid
We Exist To Win Premierships
The dons and swans are creeping up as dark horses
- Aug 18, 2009
- 4,230
- 17,507
- AFL Club
- Richmond
- Thread starter
- #854
To add to the other replies:
Even if no-one cared about the future, we'd still want to pay attention to prediction, because that tells us whether we're looking at the right things.
For example, say your only interest was in the Geelong v Fremantle match. You don't care about who will win next week or anything. But you do want to know all about what happened: who won, how, why, and what that means.
The most basic piece of information is: "Geelong won." But you'd want to know the scores, at least, which reveals: "Geelong won 74 to 72." Now you better understand what kind of match it was. And that understanding improves with more information: "Geelong scored 10.14 to Fremantle 11.6," "The match was played in Perth," "Geelong had 57 Inside 50s to Freo's 47," and the score worm that shows Freo getting out to a 34-point lead in the second quarter before falling behind early in the last.
Squiggle will add to that by showing that Geelong's Attack fell down while Fremantle's Defence was a lot better than expected, and that Geelong would normally be expected to win by more.
Now some of the above isn't very useful unless you know it has predictive power. For example, if there was no correlation between scoring shots and final margin, we wouldn't care very much that Geelong had 24 scoring shots to Fremantle's 17. And ditto with Inside 50s. They might suggest something about the style of the match, but not about how close it was, where it was won or lost, or how typical. That's also true with squiggle: if squiggle couldn't tip, there would be no reason to trust its ratings. But because these things do have predictive power, we know we can rely on them to suggest something informative.
So even for a model that isn't designed to beat the bookmakers (squiggle isn't), how good it is at prediction is still very relevant.
Log in to remove this Banner Ad
I liken the squiggle to the expert who is capable of noticing macroeconomic trends rather than the guy who can tell you which stocks to buy and sell this week. The problem with games is that there's so much variables in a single match to be perfectly predictive, but general trends tell you information.
Not only is the fixture uneven, its also lopsided. Squiggle helps you make sense of the 'true' pecking order which can also be affected by other things. Key outs, only the top 8 progressing to finals, any team being capable of beating certain others regardless of position are layered on top of that
iluvparis
Import Whisperer
- Apr 1, 2005
- 50,055
- 40,656
- AFL Club
- Carlton
- Other Teams
- Calgary Flames, Man Utd
If you a know a stock picker who nails 100% of his calls please PM me
TheFVK
Norm Smith Medallist
- Mar 21, 2008
- 7,504
- 27,014
- AFL Club
- Port Adelaide
I dont think they are the same as macro models though. Macro models are off in the short run because large structural breaks (such as unpredictable public policy and irrational herding behaviour) infiltrate all the data moving results well away from trend in the same direction. However in the long run good macro models have some predictive power as macro variables return to fundamentals (ignoring ones associated with economic growth ofcourse as they are completely reliant on made up technology assumptions).
I dont see any structural breaks in football that can influence the majority of games results in one direction away from the model in a 3 month period of time. Can you? Football results even in the short run should have a reasonably normal deviation around the mean of a model. Thus models should have some predictive power over a year unlike macro models. If it doesnt i dont know what it tells you.
- Aug 18, 2009
- 4,230
- 17,507
- AFL Club
- Richmond
- Thread starter
- #860
2017 Round 14
That's a whole lot of teams nowhere near the premiership area!
It was an open field this time last year, too, but at least we had a few contenders. Most years, there's usually at least one team snuggled in among a few premiership cups. But this year no-one's been able to deliver sustained performance.
That makes for a volatile season, and the chance for someone to make a late run, or even do a Bulldogs 2016 and win it from way back. Something dramatic like that.
It's not 1993-even. But it's pretty even.
It was a great week for two teams, Port Adelaide and Melbourne, who put together impressive victories while their Top 4 competition in Adelaide and Geelong stumbled. Huh, it feels a little weird to be putting "Melbourne" and "Top 4" in the same sentence. What a world we live in. Anyway, the only thing that could have gone better for the Demons would have been if Fremantle slotted that last goal against the Cats, just to open that door a little further.
Squiggle is very disappointed in Geelong, because it doesn't rate Fremantle much. So it was a dishonourable victory for the Cats, and they slide toward the middle of the pack.
Melbourne's rise isn't great news for Sydney, who play them this week. That one should be a great match.
Port get reward for a solid victory over Collingwood, again rolling out a very strong defensive effort in keeping the Pies to 9.8 62. Collingwood really have gone nowhere this year: they're almost exactly where they started, and haven't journeyed anywhere in the meantime. West Coast are similar.
Essendon, on the other hand, get better each week! Losing to the Swans in Sydney by a single point is good enough to send the Bombers flying up, up, again, like they have each week since Round 7 with the sole exception of their Round 10 loss to Richmond.
Things are very close in a few key areas there, and the Tower of Power (which is generated by a slower, more thorough series of simulations) disagrees, thinking GWS are more likely to snag an extra win and finish ahead of the Crows, Geelong will hold onto a Top 4 spot, and the Bulldogs will make the finals at West Coast's expense:
Adelaide's loss hurts their chances of finishing top quite a lot, partly because of missing the four points, and partly because what does it say when you lose at home to Hawthorn. It says you might not win enough games to finish top of the ladder, that's what.
Meanwhile, no-one wants 17th! Every time Carlton or Hawthorn look like locking it down, they pull off an upset win.
Meanwhile on flagpole, it's the same as always.
Live interactive squiggles!
More squiggle stuff!
LeverPuller
BigFooty Tanker
- Jun 23, 2011
- 35,113
- 40,649
- AFL Club
- Melbourne
- Other Teams
- Newcastle United Seattle Seahawks
It's not just you.
Squiggle needs to do something about injuries. Both prior to the game and within game.2017 Round 14
That's a whole lot of teams nowhere near the premiership area!
It was an open field this time last year, too, but at least we had a few contenders. Most years, there's usually at least one team snuggled in among a few premiership cups. But this year no-one's been able to deliver sustained performance.
That makes for a volatile season, and the chance for someone to make a late run, or even do a Bulldogs 2016 and win it from way back. Something dramatic like that.
It's not 1993-even. But it's pretty even.
It was a great week for two teams, Port Adelaide and Melbourne, who put together impressive victories while their Top 4 competition in Adelaide and Geelong stumbled. Huh, it feels a little weird to be putting "Melbourne" and "Top 4" in the same sentence. What a world we live in. Anyway, the only thing that could have gone better for the Demons would have been if Fremantle slotted that last goal against the Cats, just to open that door a little further.
Squiggle is very disappointed in Geelong, because it doesn't rate Fremantle much. So it was a dishonourable victory for the Cats, and they slide toward the middle of the pack.
Melbourne's rise isn't great news for Sydney, who play them this week. That one should be a great match.
Port get reward for a solid victory over Collingwood, again rolling out a very strong defensive effort in keeping the Pies to 9.8 62. Collingwood really have gone nowhere this year: they're almost exactly where they started, and haven't journeyed anywhere in the meantime. West Coast are similar.
Essendon, on the other hand, get better each week! Losing to the Swans in Sydney by a single point is good enough to send the Bombers flying up, up, again, like they have each week since Round 7 with the sole exception of their Round 10 loss to Richmond.
Things are very close in a few key areas there, and the Tower of Power (which is generated by a slower, more thorough series of simulations) disagrees, thinking GWS are more likely to snag an extra win and finish ahead of the Crows, Geelong will hold onto a Top 4 spot, and the Bulldogs will make the finals at West Coast's expense:
Adelaide's loss hurts their chances of finishing top quite a lot, partly because of missing the four points, and partly because what does it say when you lose at home to Hawthorn. It says you might not win enough games to finish top of the ladder, that's what.
Meanwhile, no-one wants 17th! Every time Carlton or Hawthorn look like locking it down, they pull off an upset win.
Meanwhile on flagpole, it's the same as always.
Live interactive squiggles!
More squiggle stuff!
Yes, small unpredictable events, irrational hearding, unpredictable policy decisions, how is this different to rain, injuries or teams playing flat?
My point is economics can often be hard to predict in short term because of the factors which can impact on a specific moment, and it's easier for a team to fluke a win than fluke a premiership. All your examples, such as unpredictable policy decisions and irrational behavior sounds like the kinds of random unpredictable behavior which generally spoils short term predictions yet is able to be explained through macroeconomic trends.
Again, there are economists who are fairly good at playing the long game but are useless at playing the stock market game. Presuming that economists must be able to play the short and long game is unreasonable and ignores how economics differs in macro and micro trends. So why should the squiggle be any different?
The squiggle loves high scoring teams. That's perhaps it's main weakness; that in ignoring the likes of Fremantle, which it was smart to, it credits sides like Adelaide who score a lot but aren't great defenders a little too much. It's one thing to be able to score, it's another thing to be able to still defend, especially away. That's something Hawthorn had but Adelaide don't.
GreyCrow
Make me an Admin!
- Mar 21, 2016
- 92,112
- 155,710
- AFL Club
- Adelaide
- Other Teams
- Sturt, White Sox, Tasmania
David Mackay likes this
Nah adelaides best is probably better than anyone else. Your problem is that your worst is also a lot worse than teams around you. Gotta find some consistency.
Part of the beauty of the squiggle is it uses the most basic stats that have been around forever (raw scores and home team for prediction), but throws out modern hyperbole and gives a more reasoned perspective.
🥰 Love BigFooty? Join now for free.
evo
Let's hit the f*ken road!
Squiggle still hates us. Especially the flagpole. 
North are second bottom, 4 games behind us, but apparently a lot better chance of the flag.
North are second bottom, 4 games behind us, but apparently a lot better chance of the flag.
- May 6, 2009
- 10,877
- 26,895
- AFL Club
- St Kilda
The other day I had a look at the saints draw for the rest of the year and my first thought was.
Geez... we're stuffed.
It's nice to know the squiggle and I are on the same wavelength.
Geez... we're stuffed.
It's nice to know the squiggle and I are on the same wavelength.
So it looks like every team other than Freo, Carlton and Brisbane have a greater chance of winning the flag than Richmond do. I guess we should be 15th on the ladder then instead of 6th?
iluvparis
Import Whisperer
- Apr 1, 2005
- 50,055
- 40,656
- AFL Club
- Carlton
- Other Teams
- Calgary Flames, Man Utd
Flagpole ignores ladder position
- May 25, 2014
- 10,120
- 14,798
- AFL Club
- Western Bulldogs
I think this has been explained a few times before, but probably many pages ago. Something like:
- If Sydney make finals, flagpole thinks they're a good chance for the flag
- Squiggle thinks Sydney will make finals
- Flagpole therefore gets to consider Sydney as a decent possibility for the flag and thinks they're the second best chance to win it.
I think.
Whereas most of us would probably take the actual prerequisite of making finals into account when making the calculations - i.e. if Sydney make finals I think they're a better chance of winning the flag than Adelaide, but they're only 50/50 to make finals whereas Adelaide are a lock. So Adelaide are actually the better chance to win the flag at just over halfway through the season.
I've started to look at this year a little bit like this, given the relative fragility and inconsistency of most of the top teams. Who do you think can win at least three finals in four weeks, including a GF at the MCG? Who has the proven form to go on a streak like that, and who do you trust in finals-type intensity multiple weeks in a row? Which of these teams actually look like they're playing good footy, and potentially building to September?
I'd have GWS first, then probably Sydney second tbh. Then I'd take probably Melbourne, then Geelong. Then maybe Port, then Richmond, then Adelaide. Adelaide's best is maybe the best in the comp, but I just couldn't trust them not to throw in a shocker at some point. Dogs are a wildcard if they can actually pick up some form, but it doesn't look likely at the moment.
So while I can look at Sydney's draw and wonder if they can even make it, I'd be comfortable backing them as one of the main fancies if they happen to sneak in. GWS are solid, but Adelaide as second favourites looks fraught with danger and the rest lack decent and/or consistent form.