Finding Non-Linear Relationships Between AFL Variables : The MINER Package

/
It's easy enough to determine whether or not one continuous variable has a linear relationship with another, and how strong that relationship is, by calculating the Pearson product-moment correlation coefficient for the two variables. A value near +1 for this coefficient indicates a strong, positive linear relationship between the variables in question, so that high values of one tend to coincide with high values of the other, and vice versa for low values; a value near -1 indicates a strong, negative linear relationship; and a value of 0 indicates a lack of any linear relationship at all. But what if we want to assess more generally if there's a relationship between two variables, linear or otherwise, and we don't know the exact form that this relationship takes? That's the purpose for which the Maximal Information Coefficient (MIC) was created, and recently made available in an R package called MINER.
Read More

What's Easier - Predicting the Home or the Away Team Score?

/
Consider the following scenario. You're offered a bet in which you can choose to predict the final score of the Home or of the Away team and your adversary is then required to predict the final score of the other team.
Read More

Measures of Game Competitiveness

/
All this analysis of victory margins, and a query from Dan about a recent blog post, has had me wondering about victory margin as a measure of the competitiveness of games. Within a given era - say 10 years or so - during which the average points scored per game won't vary by too much, victory margin seems to be a reasonable proxy for competitiveness, but if you want to consider a broader swathe of AFL history, it strikes me as being deficient.
Read More

Modelling AFL Team Scoring

/
Today's blog is the first in a series that will look at statistically modelling the scoring behaviour of teams in the AFL. If you're profoundly reductionist about it, you can think about a team's footy score as being the product of the number of scoring shots it creates and the proportion of those scoring shots that it converts into goals.
Read More

Winners' Share of Scoring

/

You might recall from seasons past my commenting on what I've claimed to be a startling regularity in AFL scoring, specifically, the proportion of scoring shots recorded by winning teams.

In 2008, winning teams racked up 57.3% of all scoring shots, while in 2007 the figure was 56.6%, and in 2006 it was 56.7%. Across the period 1999 to 2008 this percentage bounced around in a range between 56.4% and 57.8%. By any standard that's remarkable regularity.

I've recently come into possession of the scores for the entire history of the VFL/AFL competition in a readily analysable form - and by now you surely now how dangerous that's gotta be - so it seemed only natural to see if this regularity persisted into earlier seasons (assuming that it makes sense for something to persist into the past).

Below is a chart showing (in purple) the percentage of scoring shots registered by winning teams in each of the seasons 1897 through 2008. (The red line shows the proportion of goals that they scored, and the green line shows the proportion of behinds.)

Winners_Scoring.png

So, apart from the more extreme dominance of winning teams in the first decade or so of the competition, and a few other aberrant seasons over the next two decades, we have certainly seen remarkable stability in the percentage we've been discussing. Indeed, in the period 1927 to 2008, the percentage of scoring shots registered by winning teams has never been outside the range 55.0% to 59.6%. That surely almost establishes this phenomenon as a Law of Footy.

For those of you who prefer to digest your data in tabular form (preferably taken with meals), here's a decade-by-decade summary of the data.

Winners_Scoring_Table.png

The recent peak in winning teams' share of scoring was witnessed in 1995 and it came not as a consequence of a spike in 6-pointer dominance but instead from a spike in winning teams' share of behinds. In 1995 winning teams scored 57% of all behinds, which is about 2-4% higher than anything we've witnessed since. 1995 was the year that Carlton won the minor premiership kicking 317 behinds, Geelong finished runners-up kicking 338, and Richmond and Essendon, finishing in 3rd and 4th, kicked 600 more between them. By way of context, that's almost 75 more behinds than the top 4 of Geelong, Hawthorn, Western Bulldogs and St Kilda managed in 2008.

Regularity also aptly describes the history of the percentage of goals kicked by winning teams across the seasons (the red line in the chart). Again looking at the entire period since 1927, this percentage has never strayed from the righteous range of 57.0% to 61.8%.

Winning teams' share of behinds (the green line) has been, relatively speaking, quite variable, ranging from 51.9% to 58.2% in the period 1927 to the present, which once again demonstrates that it's goals and not behinds that win footy games.

A Little AFL/VFL History

/

Every so often this year I'll be diving into the history of the VFL/AFL to come up with obscure and conversation-stopping facts for you to use at the next social event you attend.

For example, do you know the most common score in AFL history? It's 12.12 (84) and has been a team's final score about 0.88% of the time (counting two scores for each game in the denominator for that percentage). What if we restrict our attention to more recent seasons, say 1980 to 2008? It's 12.12 again (84), only now its prevalence is 0.98%. Last year though we managed only a single 12.12 (84) score, courtesy of St Kilda in Round 14.

While we're on the topic of scores, which season do you think produced the highest average score per team? It was 1982 and the average was 112.07 points. The trend since that season has been steadily downwards with the nadir being in 1997 when the average was 90.37 points.

Average_Team_Scores.png

From season averages to individual game scores, here are a couple of doozies. In May of 1919, Geelong took on St Kilda in a Round 5 clash at Corio Oval. The first quarter failed to produce a goal from either team and saw Geelong lead 0.6 to 0.2. St Kilda found their range - relatively speaking - in the second quarter to lead 3.4 to 0.9 at the main break. One need speculate only briefly about the thrust of the Cats' half-time speech from the coach.

The speech clearly didn't help, however, as Geelong continued to accumulate only singles for the remaining two quarters, finally emerging goal-less and defeated, 0.18 to 6.10.

Just over two years later, in July of 1921, St Kilda swapped roles and matched the Cats' ineptitude, eventually going down 0.18 to Fitzroy's 6.8 in front of around 6,000 startled fans.

If you're looking for more sustained inaccuracy you'd be after the South Melbourne team of 1900. They managed 59.127 for the entire season, a 31.7% accuracy rate.

In contrast, in 1949 the Hawks put on a spectacular display of straight kicking at Glenferrie Oval, finishing with 7.0 for the game. Regretably, their opponents, Essendon, clearly with no sense of aesthetics, repeatedly sprayed the ball at goal finishing 70 point victors by bagging a woefully inaccurate 16.16.

Again, turning from the single game to an entire season, plaudits must go to the St Kilda team of 2004, who registered 409.253 or 61.8% for the season. But, as the Hawks discovered, accuracy does not preordain success: St Kilda went out in the Preliminary Final to Port by 6 points.