If you’re into the AFL and follow their account on Twitter, then every week you’ll see a cryptic message like the following:

AFL Hashtags for round 11

Those familiar with Twitter will of course recognise these as hashtags.  A hashtag is simply an agreed piece of text people add to their tweets so that they can be found by others interested in the same thing when using Twitter search.  For example – to find everyone talking about the AFL (using the #afl tag), you could search like this.  The intent is that people following the match on the weekend can use the hashtag so they can find each others tweets.

When I first saw these my immediate reaction was “this will never work”.  Let me explain why:

1. As a Cats fan, I’ve been using the #gocats tag last season, along with many other passionate cats fans on Twitter.  For us at least, it wasn’t broken – there is no incentive to move.
2. They look overly complicated – the teams are in the order of home then away.  With the limited grounds the AFL plays on these days, it’s a bit arbitrary as to which team comes first.  At least I know with Geelong that if they play at Skilled Stadium it’s a home match, but when they are at the MCG for example, it’s not always an away game.
3. They are a little long – with 140 characters, shorter is sweeter as far as hash tags go. At least once this year we should see #aflbombersdemons – 17 characters is quite lengthy in the world of hash tags.

Now we’ve reached half way through the footy season, I thought it would be interesting to use Tribalytic to explore these and see in fact if my immediate reaction was correct or not.

To do this I’ve taken a few different approaches.

1. Look at the performance of #gocats on game day Vs. the “official” hashtag.
2. Compare this against the performance of #gopies on game day Vs. the “official” hashtag
3. Look at how the use of the official hashtags is going round by round.

#gocats Vs. official AFL tag

Results from Tribalytic for Official AFL Tag Vs. #GoCats

There you have it – Fans using GoCats and their Tweets Vs Fans using the AFL recommended tag and their tweets.

Clearly the official AFL tag is out performing the user generated #gocats one.

#gopies Vs official AFL tag

I chose this for comparison because not only do Collingwood have a passionate supporter base, but also because Sean Callanan of SportsGeek who’s a mad Pies fan said he was one of the first using it.

Results from Tribalytic for Official AFL Tag Vs. #GoPies

Interestingly while #gocats fans are clearly MUCH more passionate than #gopies fans, it’s also clear the official AFL hashtag wins hands down in both cases.  Overall, the usage of #gocats and #gopies is generally trending down.

Collingwood fans have taken to the official hashtag and enjoy an edge with the greater number of Twitter fans, which isn’t surprising given their membership numbers, BUT the cats fans are more engaged – on average over the season Cats fans have tweeted 4.2 times to the Pies fans 3.7.

The Official AFL match hashtags round by round

Use of the official AFL match hashtags, round by round

You can see that there is a general trend upwards for the number of fans using the hashtag, and an even greater trend upwards on the number of tweets.

So I think we can conclude that the AFL match hashtags are generally a success, with increased numbers of fans using them on game day and tweeting more when they do.

Other observations

It’s clear that there are ups and downs in tweeting behaviour.  The first and most obvious is that blockbusters matter – the bigger the match, the more people want to engage with it.  For Collingwood this was most obvious during Round 5 (Anzac Day clash with Essendon), Round 8 (Battle between Freo and Pies for Top spot) and then the big Pies Vs Cats battle for top spot in Round 9.

The drop off in Round 4 and Round 6 is not easy to explain.  I tried mapping it to crowd sizes.  There appears to be no correlation between the crowd size and the number of Tweets or Fans tweeting, so it’s not that.

To get to the bottom of this, I had to do something I didn’t want to do – map the each clubs Twitter performance round by round.  So many hashtags!  Here it is, the aggregate number of Unique Twitter fans across rounds 1 – 11 for each AFL club.

It’s not a perfect analysis, obviously when #aflpiescats happens, both Pies and Cats fans are counted, still it does show quite a difference in performance.  As the Herald Sun said today – Where are all the Kangaroos fans!

Still, looking at Round 9 in context of these figures, not only was it the battle between one and two, it was also the battle between the two teams with the most active Twitter followers!

It’s hard to really piece together why Rounds 4 and 6 drop off so much, but in general I think it’s because irrespective of ladder positions or the crowds, two teams with a high Twitter fan base meeting generate a lot more noise.  Even the ground the teams play on doesn’t matter – when the Pies, Cats and Blues all played away in Round 8 they still generated a lot Twitter talk.

Conclusions

Looking at the figures, you have to conclude that the AFL hash tags are proving to be successful.

There is evidence that they are somewhat confusing to fans as in this example where you can see two peaks for #afldogspies – one when the Pies had the home match against the Dogs and again later in the season where they had the away match against the dogs.

A double peak where people have confused the teams

But Twitter isn’t meant to be perfect and it’s only a very small number (only 7 people got it backwards).

The lessons to take away are:

1. It doesn’t matter how perfect your hash tag is, it matters more that you tell people what it is.  The AFL cross promote this tag on their website, through tweets and have the support of the radio stations who use it during their broadcasts as well.
2. Consistency – stick with it and the audience will grow.
3. It doesn’t matter how many supporters or Twitter fans the Pies have, they still won’t win the flag this year! Go Cats!

A little bit of background on Tribalytic

Tribalytic is a market research tool for Twitter that constantly samples Australian twitter users. You can search and drill into references of terms over the last three months and see the most relevant users, their influence and segment and filter on related keywords and locations.  You can sign up online and access a free seven day trial.

Edit: Corrected reference to Round 7 to correct Round 8 in other observations.