Remove this Banner Ad

Certified Legendary Thread Race for the flag, in squiggly lines

🥰 Love BigFooty? Join now for free.

I don't have any massive issues about where we are at, our defence looks strong on the chart and our lack of height is reflected that our offense is off the mark so far this year. I am just not sure if the algorithm is a true reflection based on the opposition, when and where you play them so not sure it is a good predictive algorithm or more a captain obvious position based more on basic metrics.

If it is just a basic metric algorithm then a small sample size can significantly vary, if it accounts for quality of opposition, conditions, days break between games vs the opposition and projects forward based on the remaining games then it shouldn't vary significantly if it is semi-accurately predictive.
The algorithm is explained in more detail than you could ever want here.

I haven't tried to model days' break before. I really should.

There's nothing wrong with "basic" algorithms, though. There is a curse in modern statistics called over-fitting, where people throw every variable they have into a computer and have it spit out a model. And this seems like a good idea, because more variables must mean more accuracy, right? But no. This requires a proper post to explain, which I will do if I get distracted enough, but essentially with enough variables, you produce a model that fits past data very well but is terrible at predicting the future, because you're modelling random noise.

As the great mathematician John von Neumann once said, "With four parameters I can fit an elephant, and with five I can make him wiggle his trunk."
 

Log in to remove this Banner Ad

I think this model is flawed because it doesn't account for when umpires gift games to teams that they should not have won.
I could be wrong but this system appears to be completely focused on how teams score and are scored against by other teams with weightings depending on the oppositions respective offensive and defensive strengths. If umpires gift a team a couple goals incorrectly that results in an undeserved win then it wouldn't matter too much as this system doesn't work on straight wins and losses like the official league ladder.

But again, I could be wrong though...?
 
The excuses for the loss are getting funnier by the round. MacMillan broke his leg 10 minutes after Monfries limped off and we had other injured players as well, Cunnington got a knee to the head and was off the ground for a fair while, only ened up playing 74% game time because he was having mandatory concussion break and testing. We were also missing quite a few core players from our side, particularly in the middle with Swallow and Ziebell out of the game and you were over-run in the last quarter.

How much more of a leg up do you want, if you can't roll the B midfield group what are you after, the C team?
I have to admit our 2 losses are holding back any talk of top 4 and for good reason. Let's hope they were anomalies top 4 shouldn't be having too many 'bad' losses like those. Are good will match it with ANY team in the comp but our worst is a train wreak.

Hopefully getting our last 3 best and fairest winners on the park will help with consistency.
 
2014's likely finalists are starting to eerily resemble 2007 for me.

2007: Geelong, Port, North, Collingwood, WCE, Hawthorn, Sydney & Adelaide. Fremantle also 10th following a prelim year.
2014: Geelong, Port, Hawthorn, North, Collingwood, with Sydney, Adelaide & WCE all potential finalists.

Essendon was also up and about in early 2007. If Gold Coast misses and WCE sort themselves out, will be deja vu.
 
2014's likely finalists are starting to eerily resemble 2007 for me.

2007: Geelong, Port, North, Collingwood, WCE, Hawthorn, Sydney & Adelaide. Fremantle also 10th following a prelim year.
2014: Geelong, Port, Hawthorn, North, Collingwood, with Sydney, Adelaide & WCE all potential finalists.

Essendon was also up and about in early 2007. If Gold Coast misses and WCE sort themselves out, will be deja vu.
Bit early to be writing off Fremantle I'd have thought - they're still in a better place than Sydney and Adelaide.
 
I could be wrong but this system appears to be completely focused on how teams score and are scored against by other teams with weightings depending on the oppositions respective offensive and defensive strengths. If umpires gift a team a couple goals incorrectly that results in an undeserved win then it wouldn't matter too much as this system doesn't work on straight wins and losses like the official league ladder.

But again, I could be wrong though...?


Clearly wrong. Please refer to Roby's 'AFL 2014 Power Rankings' for reeducation.
 

Remove this Banner Ad

Bit early to be writing off Fremantle I'd have thought - they're still in a better place than Sydney and Adelaide.

I'm not writing them off, just didn't mention them due to their miss of finals in 2007 (although they weren't far off, finishing 10-12). At this stage they are the outlier to the comparison.
 
In mathematics you don't understand things. You just get used to them.

True.

Many decades ago a uni Maths professor spent two lectures and about 100 square metres of writing explaining to us youngsters precisely why 1 + 1 = 2.

I really didn't understand, but by that age I had already gotten used to knowing that 1+1=2.

But he wasn't as long-winded as the Philosophy professor who spent 6 lectures explaining the difference between "knowing" and "believing". Sort of interesting in a self-flaggelation way.
 
Clearly wrong. Please refer to Roby's 'AFL 2014 Power Rankings' for reeducation.
This is what happens when I post on here late after a long day of work and a couple beers. I miss subtle (or not so subtle) references to taking the piss out of crappy systems. Well played haha :thumbsu:
 
Interesting round, I'm eager to see Gold Coast make some headway on the squiggle sitting at 5-2 but look like a shocking team on the graph.
 

🥰 Love BigFooty? Join now for free.

2014's likely finalists are starting to eerily resemble 2007 for me.

2007: Geelong, Port, North, Collingwood, WCE, Hawthorn, Sydney & Adelaide. Fremantle also 10th following a prelim year.
2014: Geelong, Port, Hawthorn, North, Collingwood, with Sydney, Adelaide & WCE all potential finalists.

Essendon was also up and about in early 2007. If Gold Coast misses and WCE sort themselves out, will be deja vu.

I remember that season well. Geelong and Hawthorn were 1-2 after 15 rounds (back in the good old days when you used to play all 15 sides in the first 15 rounds) before the Hawks lost a string of crunch games (North, Port Adelaide + Sydney) that saw us let slip a seemingly attainable top 4 position.

I guess in a way it was a sign of things to come in the following seasons...

Probably a blessing in disguise as Port made a late run (with back to back thrillers away to Hawthorn + Geelong) and made an unexpected GF. If we made the GF that season we would have been smacked and may well not have recovered. Before they self destructed the logical GF pairing was Geelong + West Coast...

I've often wondered if the Eagles would have provided the Cats with a legit challenge if the shit didn't hit the fan...
 
Last edited:
If Ben Cousins hammy hadn't gone ping in the final against Port it might have been a very different GF. Still Geelong to win though.
 
These are my thoughts as well.

Hindsight is a wonderful thing but Port were on a hiding to nothing in the 2007 GF

I remember the final against West Coast that we won by 3 points. Was there and was absolutely ecstatic at the time, screaming and jumping up and down in my seat, thought it was one of the best days of my life, etc...

Looking back now, and it probably would have been far better if we had lost that game. :(
 

Remove this Banner Ad

Certified Legendary Thread Race for the flag, in squiggly lines

🥰 Love BigFooty? Join now for free.

Back
Top