lol jealous much? everyone loves the squiggle
-
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 Race for the flag, in squiggly lines
- Thread starter Final Siren
- Start date
- Tagged users None
🥰 Love BigFooty? Join now for free.
If you waited another week and unloaded on the Suns this week instead of last you would've made money.
Smeg Head
Just take a bite
EFA
Log in to remove this Banner Ad
Crankyhawk
Hall of Famer
YOu need to petition Forward press to reopen your power rankings thread. Because we were going to bet all the remaining beans on the Gold Coast at whatever H2H odds it was on Monday (when you left us) - and we would've collected. That would have rebuilt the bean bank for sure...
To be fair, it's just picked the favourites
- Banned
- #3,457
its pretty evident you are just a troll. We all got a good laugh out of the Power Rankings not because we think youre a comedian, but because we laughed at you and how pathetic your system was.
Now, just give it a rest.
And tipping with yours will send you bankrupt. You're done dude. **** off.
- Aug 18, 2009
- 4,230
- 17,507
- AFL Club
- Richmond
- Thread starter
- #3,459
This isn't a betting thread, Roby. You would have to be dumb as rocks to take the squiggle's expected margins and turn them into line bets on every game, ignoring things like:
- I've always said the squiggle probably won't beat the market, nor an insightful human observer (this is at the top of the FAQ, too)
- Even if that weren't the case, the squiggle is calibrated to tip winners, not margins
- Even if that weren't the case, a punter should only bet when there are significant gaps between predicted margins and betting lines, not blindly on every game
- Apr 8, 2006
- 46,507
- 49,026
- AFL Club
- Hawthorn
- Other Teams
- Dallas Cowboys and Texas Longhorns
Interactive squiggle looking funny. Richmond replaced by Carlton!
Why are you getting so sensitive. I am not taking the expected margins and using them into line bets. I am going against them. Simple really.
That's right, it's not even probably; it doesn't. In fact the system lags so much that it by betting against it yields cash.
For example today we can just bet on the St Kilda and Brisbane lines and make money.
The comment about
is just disingenuous.
p.s. FYI St Kilda should win today. You'll have to wonder why you tipped Essendon instead.
- Dec 5, 2012
- 14,663
- 40,548
- AFL Club
- Richmond
- Other Teams
- Knicks, Giants
There could seriously be an all WA GF...
Engimal
Cancelled
10k Posts
North Melbourne - 2018 Luke Davies-Uniacke and Paul Ahern Player Sponsor
Pantskyle
Minecraft
Shiny Penny
Ruby
Just so we're clear, are you talking about why the squiggle tipped Essendon?
- Apr 4, 2013
- 13,670
- 16,890
- AFL Club
- Fremantle
I think what was best about the Power Rankings thread, was the mini meltdown contained within the last post:
What a complete and utter loser.
Crankyhawk
Hall of Famer
Although he did collect on that one, but how many beans to the dollar?
- Apr 28, 2008
- 13,702
- 10,341
- AFL Club
- West Coast
- Other Teams
- Arsenal Kilmarnock
Out of all the playing teams over the weekend (and given Port played on Thursday night), West Coast would have been the most aggrieved by the Phil Walsh thing. I am surprised that they still managed to win so easily in the end. Flat track autopilot?
I remember that Port and Adelaide were briefly top 2 about 2/3s of the way through 2003, and the WA teams are starting to enter that sort of territory. Long way to go yet, Sydney and Hawthorn (among others) will always be a factor until the fat lady sings otherwise.
I remember that Port and Adelaide were briefly top 2 about 2/3s of the way through 2003, and the WA teams are starting to enter that sort of territory. Long way to go yet, Sydney and Hawthorn (among others) will always be a factor until the fat lady sings otherwise.
Can't wait to see Bombers dive back down to Brisbane territory based off this performance.
Did Fox Footy just use a version of the squiggly lines on their broadcast (Before the Freo/Bris match) ???
Yep, first thing I thought of. The one comparing premiership standard attack and defence.
🥰 Love BigFooty? Join now for free.
- Apr 8, 2006
- 46,507
- 49,026
- AFL Club
- Hawthorn
- Other Teams
- Dallas Cowboys and Texas Longhorns
SJ
Premium Platinum
I think the debate about tipping, betting and power rankings ignores the fact that all these things rely on different attributes.
To be a good tipper, you need to recognize form, read injuries, draw in a fairly rational and consistent manner. For instance, Essendon's form was bad going into today's match, St Kilda were hot, and coupled with Carlisle, an upset was on the cards. That said footy maths has form baked into its system and it predicted a 28 point win to Essendon. Footy forecaster also predicted an Essendon win, I'm not sure if that system factors in injuries or form but it shows that most systems predicted Essendon. Roby will say that this is proof of his lone wolf genius status but he's got a fair bit wrong too.
To be a good better, you often have to take smart long bets trying to minimize your bets and take calculated risks. You'll often bet against who you think will win based on the betting line. Good betting and good tipping sometimes don't mean the same thing, in roby's 2015 thread, roby often bet one way, and tipped another way because of the value of the upset.
To be good at power rankings, you need to recognize trends and patterns over a long term period, and take into account things like draw hardness, offensive output etc.
These things often don't fit in with each other. Power rankings favours looking at general patterns over years, tipping rewards say noticing form over the last four weeks or an injury, betting rewards say taking a calculated gamble at a 4 or 5 dollar margin. For me, this thread is mainly about power rankings, not tipping or betting. It ignores injuries, not because they should be ignored, but because it's hard to make a power rankings system with them factored in. I think ultimately the best power rankings system would have injuries baked into it, but I think that roby hasn't figured out how to do it accurately and objectively enough. Which is why I see roby's rankings as the first model of a full power rankings, a good idea but full of bugs and in need of massive improvement. But I'm not sure anyone can actually pull off an injury included power rankings successfully. Simply put, we never have data about sport which is devoid of injury distortion, meaning we have no data to be able to figure out what the AFL would look like injury free. It's a nice idea but maybe impossible? All you're left with are truisms like "the suns play better with Ablett and the rest of their best 22 playing" which is the closest that this competition will ever get to a tautology. But what does this mean specifically? If Gold Coast played tomorrow with a full side against the rest of the competition with full sides, how many sides would they beat? No-one can answer this because we have no data to quantify it. Roby has provided an answer: everyone. I'm not convinced. I think that roby's system overrates the suns and that final siren's system underrates the suns.
But here's the kicker.
The difference between final siren and roby is that final siren admits his limitations and never claims that the squiggle is anything it's not. Roby tries to do betting, tipping and power rankings all together and it gets mucked up because the three things aren't driven by the same goals. Final siren's model tells me that even though Richmond fans look at specific results and think that the defence is rubbish and that Richmond only has one decent defender, that over a period, it's actually stood up over time. That's something that 99% of all fans will miss without someone crunching the numbers. This is what I take from this thread, how to measure how good each team's defence and offence is. It's also a very good model for telling us which team is good enough to win the grand final and which teams aren't. Its scope is macro not micro.
To be a good tipper, you need to recognize form, read injuries, draw in a fairly rational and consistent manner. For instance, Essendon's form was bad going into today's match, St Kilda were hot, and coupled with Carlisle, an upset was on the cards. That said footy maths has form baked into its system and it predicted a 28 point win to Essendon. Footy forecaster also predicted an Essendon win, I'm not sure if that system factors in injuries or form but it shows that most systems predicted Essendon. Roby will say that this is proof of his lone wolf genius status but he's got a fair bit wrong too.
To be a good better, you often have to take smart long bets trying to minimize your bets and take calculated risks. You'll often bet against who you think will win based on the betting line. Good betting and good tipping sometimes don't mean the same thing, in roby's 2015 thread, roby often bet one way, and tipped another way because of the value of the upset.
To be good at power rankings, you need to recognize trends and patterns over a long term period, and take into account things like draw hardness, offensive output etc.
These things often don't fit in with each other. Power rankings favours looking at general patterns over years, tipping rewards say noticing form over the last four weeks or an injury, betting rewards say taking a calculated gamble at a 4 or 5 dollar margin. For me, this thread is mainly about power rankings, not tipping or betting. It ignores injuries, not because they should be ignored, but because it's hard to make a power rankings system with them factored in. I think ultimately the best power rankings system would have injuries baked into it, but I think that roby hasn't figured out how to do it accurately and objectively enough. Which is why I see roby's rankings as the first model of a full power rankings, a good idea but full of bugs and in need of massive improvement. But I'm not sure anyone can actually pull off an injury included power rankings successfully. Simply put, we never have data about sport which is devoid of injury distortion, meaning we have no data to be able to figure out what the AFL would look like injury free. It's a nice idea but maybe impossible? All you're left with are truisms like "the suns play better with Ablett and the rest of their best 22 playing" which is the closest that this competition will ever get to a tautology. But what does this mean specifically? If Gold Coast played tomorrow with a full side against the rest of the competition with full sides, how many sides would they beat? No-one can answer this because we have no data to quantify it. Roby has provided an answer: everyone. I'm not convinced. I think that roby's system overrates the suns and that final siren's system underrates the suns.
But here's the kicker.
The difference between final siren and roby is that final siren admits his limitations and never claims that the squiggle is anything it's not. Roby tries to do betting, tipping and power rankings all together and it gets mucked up because the three things aren't driven by the same goals. Final siren's model tells me that even though Richmond fans look at specific results and think that the defence is rubbish and that Richmond only has one decent defender, that over a period, it's actually stood up over time. That's something that 99% of all fans will miss without someone crunching the numbers. This is what I take from this thread, how to measure how good each team's defence and offence is. It's also a very good model for telling us which team is good enough to win the grand final and which teams aren't. Its scope is macro not micro.
Last edited:
I'm not going to get into debate about everything you've said otherwise will be here for a very long time.
There is a reason why I posted in the power rankings thread tips against the Squiggle, because I knew how it performs. I was probably hitting Max's website two or three times a day which I bet (excuse the pun) is more than anyone here. I could suggest an improvement for the Squiggle's algorithm quite easily with one little tweak, but why should I given Final Siren has been a dick for the last couple of years. I never actually reported him because I really didn't care in the end. But I knew it was ultimately just jealously.
His model sits in bottom six of the age tipsters. This week by betting against it you would have return nearly double your money with 7 of the 8 lines coming up. If you think it's an oddity, it's not.
It's not very good at predicting the Grand Final because last year he tipped Sydney when nearly every other model tipped Hawthorn. Hawthorn didn't just win, they won by 10 goals. If it is ultimately a race for the flag in squiggly lines and if it can't even pick the one game it matters what does actually do well?
- Is it good at tipping?
- Does it beat the market?
- Is it good with margins?
I found the above point you made interesting.
Most tipsters tipped Essendon and Essendon were favourite. Having said that only two tipsters in the age tipped a higher margin for Essendon than Squiggle. Even leading tipsters tipped Essendon this week.
You can't just say there is limitations to a model when most people tipped Essendon. It's too simplistic of an assessment and doesn't add anything. The reason why Essendon were favourite is because Essendon were much more competitive against the premiership favourite and the measuring stick of the competition. St Kilda lost 63 points to the Hawks while the Bombers only lost 38.
The problem lies much deeper. You see you get one thing incorrect and everything falls. Bulldogs are not rated at the moment and I understand why but Saints played well against them. Neither is Melbourne. I'm guessing that they will now beat Essendon next week. Again I checked the squiggle and it's tipping Essendon over the Dees. But the Bombers are still very injured they played like shit against a team that the Dees nearly beat, is it so hard to create a model that will tip Melbourne next week?
Alan Turing said that a computer could be considered intelligent if it could deceive a human into believe it was human. Roby is actually a robot I wanted as kid when I was growing up.
I can tell you that I am an awful tipper. If the model didn't predict outcomes I would not bother posting anything. My model is designed to imitate human intuition and common knowledge. It's a model that learns, but a even a learning model has it's flaws. Intuition is wrongly attributed as been opposed to rational thinking. The human brain is a machine; a quantum computer. There is no intuition, or free will. Intuition is logical pattern of thought which has not been empirically studied due to it's complexity.
This year nearly every week the Squiggle changes it's mind on who will play in the Grand Final. It's still got Fremantle making it. They won't get there with a 10 foot pole. So what happened? Wasn't it a race for the flag? Surely it won't get it wrong two years in a row?
Similar threads
