How equal is equal? Part Two (hey mate, good game)

In my last post I discussed measures of equality in the AFL. As forecast, I couldn’t just let one analysis be enough.

Below are charts showing the proportions of games decided by 6, 12, 18, 24 and 30 points or fewer. In short, this is a measure of how often players get to say “good game” and mean it more than Wiley Wiggins in Dazed and Confused (see from 1 minute mark or so).


Margins rolling

From the first two charts, you could certainly get the impression (along with retina damage) that the proportion of good games is on the decline. The year to year figures jump all over the place as you’d expect, but the rolling averages show the stark difference in the proportion of “good games”.

They didn’t get many games each year in the nineteen-teens (just 49 games in 1917) but the games they did get were good ones – an amazing 85% of games decided by under 5 goals in 1916, compared to below 50% for most of the last 30 years.

Interesting to note, too, the most of the decline has been at that top end; the proportion of “very good” games is lower than the peak, but by a lesser margin than any of the other categories.

However, let’s not leave it there. In line with the last post, in a second experiment each of the margins has been adjusted to accommodate the prevailing scoring norms of the day. An adjustment factor has been calculated using 2013 scoring as the benchmark. (Hypothetical example: if the average score per team in 2013 is 100, and in 1970 is 95, the factor is 100/95 = 1.052. Each margin in 1970 is then multiplied by √1.052. I tried it without taking the square root, and the results seemed a little… dramatic.)

See the results below – once again, sunglasses recommended before viewing the charts.

Adjusted margins

Adjusted margins rolling

This second set of charts de-emphasises the apparently golden era of close football games in the early 20th century (a 15-point win probably isn’t really as close when it’s 46 points to 31). It also echoes the first, in that it shows the direction that the AFL was heading before the 1987 introduction of the salary cap – one potential explanation for the flattening out of some of these lines, which seem oddly in line with 0.1, 0.2 etc up to 0.5.

Without access to huge swathes of data that I don’t have, it’s impossible to say whether this makes the AFL more or less competitive (by this measure) than other leagues across the country. However, I’ve always assumed that (like me) other AFL-watchers enjoyed close(ish) games. Is a fifty-fifty probability of seeing one satisfactory?

Reader feedback on this, as ever, is welcome.


2 thoughts on “How equal is equal? Part Two (hey mate, good game)

Click here to like, or tell me whether this is genius or gibberish

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s