AFL Prediction & Analysis

Are North and Hawthorn bad, or have they just faced a tough draw?

From Hurling People Now:

Almost every credible (and non-credible) football journalist in the country has waxed lyrical about the decline of the Mighty Hawks, and the Considerably Less Mighty Roos. Both sides were finalists in 2016, but have gotten off to tough starts in 2017, a combined 0-8 through a month of footy.

Source: Are North and Hawthorn bad, or have they just faced a tough draw? | Hurling People Now

The Secret Weapon of the AFL’s Best Forward-line – Figuring Footy

Figuring Footy shows just how much extra attacking real estate a side gains when it can kick goals from further out, and looks at how effectively Adelaide are using it:

Because the ground opens out wider the further you get from goal, adding an extra 10 metres to your kicking distance does not just increase your opportunities by 20%, but rather increases the size of your “goal scoring zone” by around 43%. This is a massive increase in the area the opposition needs to defend, both from leading forwards and midfielders/backmen drifting forward.

Source: The Secret Weapon of the AFL’s Best Forward-line – Figuring Footy

Squiggle on Round 4

I’m reading “The Enchanted Wood” to my kids at the moment. In the middle of the Faraway Tree, Moon-Face has a slippery-dip that goes all the way down the inside the tree, which Joe, Beth, and Frannie ride down on cushions. Just like Hawthorn! Wheee!

But as you can tell from the flood of Richmond supporters entering this thread, the big story this week wasn’t that Hawthorn are heading for the soft moss outside the trapdoor at the bottom of the tree, nor that Geelong might actually win the flag this year, and not even that Nathan Buckley is nearing completion of his master plan to dismember one of the greatest teams in modern football. No! IT’S THAT THE TIGERS ARE 4-0.

Like every Richmond fan, I’ve spent the past 48 hours scouring the internet for anyone prepared to say we’re a premiership threat and mashing the LIKE button. I was hoping Squiggle would deliver for me here, but no, it’s still pretty tepid on the Tigers. It is tipping them to make finals for the first time, but only just, and it thinks the next month promises a return to reality, with a series of tough games.

So obviously Squiggle has SERIOUS FLAWS.

Actually, it’s for two reasons. Firstly, Squiggle doesn’t think teams get a lot better or worse in a hurry. It’s pretty conservative in this regard, because historically, teams really don’t get a lot better or worse in a hurry very often. As shocking as it is to see Hawthorn suddenly coughing up consecutive 86-point defeats, this is a team that started exhibiting symptoms of the same disease that claimed Fremantle the year before: pinching a bunch of close games and tailing off into the finals. In 2015 I wrote that Fremantle were the luckiest minor premier ever; in 2016 I wrote that in a fair universe, Hawthorn would have finished 6th, not 3rd. Most of the time, just like when Joe visits the Land of Toys and is turned into a clockwork clown, teams start hinting at what they’re becoming before they actually become it.

And the Tigers were pretty bad last year, finishing 13th with 8 wins, a percentage of 79.5, and no excuses like terrible injuries or a run of close losses. They were just bad. Admittedly, they made finals three years in a row before that, but in the worst possible way you can make finals three years in a row: the way where you don’t win any. This isn’t a premiership contender bouncing back from one dud season.

Secondly, while the Tigers have outperformed Squiggle expectations in all four matches, they’ve only done so by a few goals each time. You can move rapidly on Squiggle if you deliver (or receive) an unexpected thrashing, but the Tigers haven’t done that. They’ve just been a bit better.

So Richmond is rated a middle-ish team with four wins in the bank. Like North Melbourne last year, that looks like a club that will make finals, but not do much more. So, you know, Richmond 2013-15.

And that means that, like last year, there’s a gap between the best seven (Adelaide, GWS, Geelong, Bulldogs, West Coast, Port Adelaide, Sydney) and the rest. Just as the Tigers haven’t risen far because they’ve haven’t delivered a thrashing, Sydney haven’t fallen far because they haven’t received one. So while the Swans’ 0-4 start has virtually destroyed their hopes of a Top 4 finish, they’re probably still a very good team.

Geelong is this week’s big winner, by opening up their Top 2 chances. This is a lot more graphic in Tower of Power form:

Especially if you imagine a little regurge sound. Or produce one involuntarily. Of course, it’s possible that Hawthorn is going full Fremantle 2016 and the Cats aren’t that great; the Hawks are just that bad. After all, even Gold Coast can beat the Hawks by 86 points. But 134-48 is still impressive, particularly the 48 part, where they restrict an opposition to six goals.

It was a bad week for Brisbane, who look increasingly lonely as most of last year’s bottom 6 show improvement (Richmond, Fremantle, Gold Coast, Essendon). I mean, not Hawthorn-bad. Relative to Hawthorn, Brisbane’s week was like visiting the Land of Birthdays, where you get a wish granted with every slice of cake you eat. But still not great.

Flagpole!

Live squiggle!

Why 2017 Results Are Flattering The Cats – Figuring Footy

Figuring Footy ShotPlots are back!

One other method used by most footy fans is considering the number of scoring shots. A team leading 3 goals straight to 1.11 is probably not dominating the game to the standard they’d like. The problem with this method is that goals/behinds doesn’t tell us anything about the difficulty of the chances each team is creating. Maybe a return of 1.11 is about right if all a team is doing is pinging it from the centre square.

Source: Why 2017 Results Are Flattering The Cats – Figuring Footy

Tower of Power after Round 3

What does it look like if you run 10,000 simulations and count how many times each team winds up in each position? This!

At this stage, there’s still lots of uncertainty, so teams are mostly long slashes, capable of finishing in many different positions. As the weeks go by, they tend to flatten out.

Adelaide are already settled in the top 4, though, the squiggle reckons, and Brisbane are very likely for the spoon.

Gold Coast and Freo evacuating those bottom rungs opened up a lot of space for other teams to fall into. That chute opening for Hawthorn all the way down to 17th is kind of terrifying.

Computers are beating the bookies

It’s been a rough start to the year for tipsters. And the start of a season is a tough time for computer models, which generally don’t know about personnel changes during the off-season.

Still, after three rounds, the computers models are doing okay!

Tips Bits
Figuring Footy 18 0.88
Squiggle 18 -0.96
PlusSixOne 17 1.28
Aggregate 17 0.48
Footy Maths Institute 17 -1.84
Punters 16 0.73
Matter of Stats 15 0.16
The Arc 15 0.04

“Bits” from Monash University rewards tipsters for saying a win was likely and punishes them for saying it was unlikely.

Squiggle has been fortunate with a few line-ball calls, doing well on raw tip numbers but badly burned on over-confident predictions that turned out to be upsets.

Figuring Footy has been strong on both. PlusSixOne has been excellent in probabilistic terms, and is currently leading in Bits.