The Finger Points Outwards - No.31

A research report I contributed to on employees earning below the Federal Minimum Wage has been released (Report 3/2011 - Employees earning below the Federal Minimum Wage: Review of data, characteristics and potential explanatory factors).

The Finger Points Outwards - No.30: 65 Greatest Shots in NBA History

Hmm, there's been quite a few basketball posts here recently, but I can't resist a link to this cool visualisation of the NBA's 65 Greatest Shots.

(Also, on the same site - every slam dunk video from the NBA Slam Dunk Contest.)

Game Review - NBA Jam

If you're a junior basketball coach, you'd want to keep your charges well away from NBA Jam, where players are rewarded for showing off and shoving the other team on defence. As a rule-follower myself, I felt a bit uncomfortable at first with this laissez-faire approach to the game J.Naismith built, but after a while you get used to the mania, and how it marks a refreshing change from the 'if it's in the game, it's in the game' attitude of other sports sims. For those unfamiliar with the original, NBA Jam essentially revolves around a fast-paced game of 2-on-2, in which players have abnormally big heads and can leap higher than the backboard to deliver court-shaking dunks. For those already converted to the church of 'Boomshakalaka!' this version updates the players to today's rosters, but you're able to unlock the players from the early '90s version, as well as other basketball legends not named Jordan. There's also a 'remix tour' which involves a slightly different version of 2-on-2 with power-ups (but these don't tend to affect the game that much), as well as a backboard smashing game, a 'domination' game in which you shoot and try to capture seven spots on the floor, and boss battles. And you can still catch 'on fire' if you make three consecutive shots. Like all sports games it tires after a bit, still it's not like you can replace it by going down to your local court and doing backflips from the free throw line. Or growing a gigantor head. Or shoving your opponent to the ground. At least not without some unintended side-effects.

Album Review – Cut Copy: ‘Zonoscope’

What is the difference between OK and addictive? When Cut Copy’s second album, ‘In Ghost Colours’ was released, it seemed like the perfect soundtrack to Melbourne city nightlife (or indeed, any Western city’s nightlife) – swooning vocals, open guitar, pulsating synths - with every track flowing seamlessly into the next and demanding your attention. Most of the same elements are present on their new album, ‘Zonoscope’, and yet it feels much more like scenery than its predecessor. It should work … for example, the 15-minute closer ‘Sun God’ has the potential to be an epic Mancurian-like final track for the ages, with its insistent beat and looping chorus of ‘Are you gonna give me your love/Love will treat you’, and then its winding down for a patient breather before heading into the final flurry. But maybe that’s the problem – it’s all a little bit too much in the head, and built around what should work rather than what does. On ‘In Ghost Colours’, even though a lot of work and craftsmanship undoubtedly went into the music, it also felt like it had been bursting to get out ever since Dan Whitford bought his first Casio. ‘Zonoscope’, in contrast, feels too drawn out, which is not to say that there are not some good tracks on here – ‘Alisa’, ‘Take Me Over’, ‘Pharaohs and Pyramids’ and ‘Blink And You’ll Miss A Revolution’ in particular. But it’s never a good sign when listening to a band’s new album makes you long for their last one. Never mind Cut Copy – I’ll still see you when you tour.

Image copyright Modular

Introducing the AFL Power Rankings - Part Three

OK so far I have introduced the formula for the AFL Power Rankings, which is built around the concept of relative adjusted net margin, where points for versus points against are adjusted for the strength of the opposition and the state where each game is played. I have also talked about what adjustments could be made to net margins, based on the historical results of teams when they travel interstate. Now it is time to calculate us some numbers.

When calculating the numbers, I made one major change to the methodology I had outlined to date. In a previous post, I had proposed using the following formula:

Team Score = 2/3*(Average relative adjusted net margin over past 22 games) + 1/3*(Average relative adjusted net margin over past 5 games)

But I have since replaced it with this formula:

Team Score = Sum over past 22 games of (Weight for game * Relative adjusted net margin for game)

Where weight for game = (1/22) + (1/22) * (11.5 – No. of games ago)/11.5

Essentially the difference is that, in the latter formula, each game has a progressively lower weight the less recent it is. This seemed to me a better weighting pattern than my first formula where, for example, every game from 6 games ago to 22 games ago had the same weight.

Applying this formula gives us the following rankings after the 2010 AFL home-and-away season:

1. Geelong 35.6
2. Collingwood 34.0
3. Western Bulldogs 17.1
4. Hawthorn 16.0
5. St. Kilda 9.6
6. Carlton 5.9
7. Sydney 4.0
8. Adelaide 3.5
9. Melbourne -5.7
10. North Melbourne -7.6
11. Fremantle -9.4
12. Port Adelaide -19.1
13. Brisbane -20.1
14. Essendon -23.5
15. Richmond -24.1
16. West Coast -24.8

Given what we know transpired in the finals, it may seem that Collingwood and St. Kilda are too low. However, while it may be difficult to remember this fact now, Geelong and Collingwood were roughly equal favourites for the premiership heading into the finals series, with Geelong coming off thumping wins against the Bulldogs and Carlton heading into September, and Collingwood squeaking past Adelaide and losing (albeit narrowly) to Hawthorn. Meanwhile, St. Kilda had a so-so second half of the season, only winning three and drawing one of its last seven games.

The rankings of Hawthorn and Fremantle may also be controversial, given that the Dockers polished off the Hawks in their first final. But leading up to that game Hawthorn were displaying much better form than Fremantle, with the Hawks making up for their poor start to the season, and the Dockers squandering their great start. If you take away Fremantle’s 116-point thrashing in Hobart, where it fielded a second-string side, it would move up to ninth in the rankings, which is perhaps a fairer reflection of its form.

If we factor the 2010 finals in, we come up with the following rankings:

1. Collingwood 39.8
2. Geelong 30.6
3. St. Kilda 12.9
4. Hawthorn 12.0
5. Western Bulldogs 8.2
6. Carlton 4.2
7. Sydney 3.4
8. Adelaide 1.6
9. Melbourne -7.1
10. North Melbourne -10.0
11. Fremantle -11.1
12. Port Adelaide -21.0
13. Brisbane -21.7
14. Essendon -25.4
15. Richmond -25.9
16. West Coast -26.3

Collingwood has now pulled clear of Geelong, and St.Kilda has moved up to third, reflecting the good form of both of those teams during September. In terms of its score, Hawthorn has been punished for its first-up loss in Fremantle, although it remains in fourth position overall. (The scores of non-finals teams also change as all of the relative margins change, but only by small amounts.)

I’m reasonably happy with the post-finals rankings, so I think this formula will stick. That being the case, the Depot will bring you the updated AFL Power Rankings every week throughout the season (though with considerably less discussion). Stay tuned.

Introducing the AFL Power Rankings - Part Two

In my previous post on a proposed ‘Power Rankings’ system for AFL teams, I talked about the concept of adjusted net margin, where the net margin for a team (points for – points against) would be adjusted by the relative advantage/disadvantage from playing home/away. I said there that the adjustments would need to be worked out; in this part I offer some opinions on what those adjustments could be.

To answer this, I calculated the average losing margins of ‘away’ teams at the seven main home grounds of AFL teams: AAMI Stadium in Adelaide, the GABBA in Brisbane, Skilled Stadium in Geelong, the MCG and Etihad Stadium in Melbourne, Subiaco Oval in Perth, and the SCG in Sydney. I did this from 1987 onwards, when West Coast (and the Brisbane Bears) joined the league, but did not bother including the Bears or Fitzroy in my calculations. For the two Melbourne grounds, I assumed that those grounds were ‘home’ grounds for all of the Victorian teams. There were some differences in the Victorian teams' performance between the two grounds, but I think that has more to do with how successful each team has been since Etihad Stadium opened in 2000.

The average losing margins of ‘away’ teams at each ground were as follows:

Skilled Stadium – 22.6
AAMI Stadium – 18.7
GABBA – 18.6
Subiaco – 12.2
MCG – 8.0
Etihad – 7.0
SCG – 6.6

From these figures, it seems that Geelong, the Adelaide teams, and Brisbane have the highest home ground advantages, but one thing that needs to be remembered is all those teams have been fairly successful over the past couple of decades. For example, Geelong’s average net margin since 1987 is itself over two goals. If you look at those teams’ overall average net margins – that is, roughly how well those teams would be expected to do at a ‘neutral’ venue – then I reckon their home ground advantage is worth about two goals in each case. I also think the home ground advantage at Subiaco is worth about two goals, once you account for West Coast’s and Fremantle’s overall performances.

The SCG result is a little surprising, even accounting for the fact that it includes some games where two ‘away’ teams played (for example, when North Melbourne and the Bulldogs played ‘home’ games there). My first thought was to say that the MCG, Etihad and the SCG all have a home ground advantage of about nine points against all interstate teams. But looking closer at the figures, I noticed that Sydney tends to perform better in Melbourne than other interstate teams, and Victorian teams tend to perform better in Sydney than other interstate teams. Therefore, I reckon a better approach is to say that playing at the SCG is worth one goal advantage to Sydney against Victorian teams, and two goals against all others, and that playing at the MCG or Etihad is worth one goal to Victorian teams against Sydney, and two goals against all other interstate teams. (If you hadn’t already worked it out, I’m setting the advantage of playing in Melbourne against another Victorian team as zero.)

Those results seem fairly reasonable to me, so now I have pretty much all I need to go calculate some ranking points.

Movie Review - Easy A

Usually I hate 'teen movies', and those set in California even more. But this one is really good, and smarter than your average flick. The title of the movie itself is a reference to Nathaniel Hawthorne's 'The Scarlet Letter', and there's a few Huck Finn jokes as well. Also, jokes take a while to pay off, which means you have to pay attention for more time than it takes to tweet. Emma Stone is excellent as the high school girl who pretends to be a skank in order to help boost the reputations of the socially awkward - her character is a clever mixture of self-awareness and naïveté, cynicism and hope. I also picked up two phrases I will have to use in future - 'terminological inexactitude' (as in 'there was a lot of terminological inexactitude around the office today'), and 'you burn through a lot topics really quickly' (as in 'shut up'). An even nicer surprise than 'Mean Girls'.

Introducing the AFL Power Rankings

Over at ESPN.com each week, columnists Marc Stein and John Hollinger produce their NBA Power Rankings. The aim of these rankings is to ‘look through’ the standings to see where each team really stands in the scheme of things. Stein’s rankings are based on his opinion, while Hollinger’s rankings are based on a statistical formula.

I have wondered if Hollinger’s approach could be translated across to the Australian Football League. After some pondering, I think a formula that could be used is as follows:

Team Score = 2/3*(Average relative adjusted net margin over past 22 games) + 1/3*(Average relative adjusted net margin over past 5 games)

Let me explain:

1) Like Hollinger, I think the strength of each team should be based on their average net margin (points scored less points conceded) rather than their wins and losses. I have not checked if winning or losing margin is indeed a better predictor of a team’s future success in AFL than wins and losses, but if makes sense that it would, given teams that keep winning close games are riding their luck to an extent.

2) Average net margin needs to be adjusted for the strength of teams played, hence giving us the concept of average relative net margin. Average relative net margin is calculated as the average over all a team’s games of their net margin against each opponent less the average net margin by all teams against each of those particular opponents. So if Team A beats Team B by 3 points but if, on average, teams beat Team B by 13 points then relative net margin for Team A from playing Team B is -10 points. Over the long-run, as they play most of the teams, a team’s average relative net margin should more or less simply approximate a team’s average net margin.

3) Average relative net margin in turn needs to be adjusted for the advantage from playing at home and/or the disadvantage from playing interstate, giving us the concept of average relative adjusted net margin. To do this, all the net margins for a team would need to be adjusted by the relevant advantage/disadvantage (which would need to be worked out) from playing home/away. So if a Team A wins by 3 points when playing interstate but the disadvantage from playing interstate is 18 points then the adjusted net margin of Team A is 21 points.

4) I have put two-thirds weight on a team’s success over the last 22 games, and one-third weight on their success over the last 5 games. That seems to me a reasonable estimate of how people generally weigh up a team’s performance in recent matches.

If/when I can put some figures together, I’ll try and put this system into practice.