# USA Furry Convention Attendance - A New Measure of Growth

Furry conventions are great. Where else can you meet (and make) friends from around the world, participate in all manner of weird and wonderful activities, *and* awkwardly ask a stranger to draw your character in a variety of ENTIRELY NORMAL situations? While by no means the be all and end all of the furry fandom, conventions also offer an insight into the size - and growth - of our community, with attendance figures a semi-regular hot topic (at least among those of us for whom numbers can ever be described as a hot topic…). After my last post exploring furry convention statistics, I wanted to dig a little deeper into the distribution of attendees at some of the fandom’s biggest events - and look at a slightly different way of measuring their relative importance.

##### Rawr attendance

With 7,554 attendees this year, Anthrocon continued its reign as the largest furry convention in history. However, that figure disguises a comparatively small growth rate of just 3.33%, suggesting Midwest FurFest may finally take over as the world’s biggest furcon this fall. Regardless of whether you think which convention holds this particular accolade even matters, it would nevertheless represent a watershed moment in US convention history, and we can see this possibility quite succinctly if we plot attendance figures of some of the biggest US cons over the last ten years.

These ‘big’ cons were chosen by the entirely arbitrary metric of whether they’ve had a year with over 2,000 attendees, and while the above plot is kinda messy, we can make out some trends. (Fair warning: I’m going to start using con name abbreviations now to save me some typing.) First up, MFF does indeed look destined to overtake AC in 2017, but there is a not inconsiderable caveat in the shape of Rainfurrest. RF came to an end in 2015, and so you’d expect at least some of those attendees to move to other conventions. This could lead to MFF’s 2016 growth rate being slightly artificially inflated, and is one reason why I’m disinclined to try any formal statistical modelling of these growth curves (although the conclusions, were we to do so, are I think pretty obvious). We can also see that AC’s 2016 growth was a little out of keeping with previous years, which may also be the manifestation of an ‘RF effect’ (but we don’t really have any way of testing this with the available data).

BLFC is the other stand-out performer from the above, with its borderline ridiculous growth seeing it comfortably move into third overall. Further Confusion, meanwhile, has been static (or even in decline), while FWA and TFF have seen broadly similar (and fairly linear looking) growth for the last few years.

##### It’s good to share

Raw attendance figures are all well and good, but I’ll admit I don’t find them particularly interesting. Instead, I wanted to look at an approximation of each convention’s ‘market share’ - in other words a measure of what proportion of *total* US furry convention attendance can be attributed to an individual con. We can estimate total US convention attendance using data from this wikifur page. This lists (or, more accurately, attempts to list) all furry conventions per calendar year with at least 100 attendees. By totalling the number of attendees across all such conventions each year we can get an approximation of total furcon attendance, allowing us to look at overall market share for some of the bigger cons. For example, in 2012 attendance at all US cons with over 100 attendees was 21,850, while AC’s attendance was 5,179 giving it a 20.3% market share, and we can do the same for all cons across all years. Once again, this comes with a big bunch of issues, which I discussed in a couple of weeks back, but for the most part I’m pretty happy this dataset will work well enough for what follows. (One thing to remember is many people attend multiple cons each year, so these figures are total attendance, and *not* total unique attendees!)

Market share allows us to set a convention’s growth in the context of the *overall* increase in furry convention attendance, which tells us something slightly different to simple attendance numbers. For example, if total US convention attendance increases by 10%, but a particular con’s attendance only grows by 5%, that con’s market share will decrease, reflecting the fact that its overall ‘influence’ (so to speak) has declined.

It’s been a while since we had a graph, so have another colourful monstrosity summarizing this market share statistic for the ‘big 7’ conventions along with the smaller ones (bundled under ‘other’).

To my eye, this paints a rather clearer picture of the state of US cons than overall attendance figures. We can clearly see, for example, that despite year-on-year growth, AC’s (and FC’s) market share has been steadily decreasing for much of the last decade. Some of this can be blamed on the increase in other conventions - there were 16 events in 2007 with over 100 attendees compared to 27 in 2016 - but this hasn’t stopped MFF and especially BLFC increasing their respective shares across the same period. In fact, ‘other’ cons have seen barely any increase in their share of total attendance over the last decade. (For the stats fans, a correlation test of no relation between year and market share for ‘other’ cons returns a p-value of 0.02, with an average increase of just 0.5% per year over this period.)

It’s also interesting to focus on 2016 in terms of market share, where both AC and MFF see jumps that appear slightly out of keeping with trends of the preceding few years. AC had seen steady decline from 2012-2015, but actually increased its market share last year, while MFF’s share - though increasing over a similar period - appeared to show a slight spike. How much of this is attributable to the end of Rainfurrest is impossible to say, but it seems a reasonable explanation for at least some of these patterns. We should know more on that front by the end of the year.

There’s a little less going on with the other conventions over this timeframe, but we can spot a few things. FC, as we’d expect from broadly stationary attendance numbers, is seeing its market share drop, while FWA and TFF are holding steady or increasing a little. BLFC’s ascent, meanwhile, is remarkable however you slice the numbers.

The figures for the ‘other’ cons also highlight a couple of interesting details, with a big drop in 2009 and a slightly smaller one in 2013. The 2009 shift is largely attributable to the first year of TFF (with 542 attendees) and the absence of FA: United (which had seen 381 attendees in 2008), representing a fairly large shift of attendees from the others column to one of the bigger cons. 2013 also saw a new big convention (BLFC), which explains most of the drop in the others market share that year. These shifts, while of mild historical interest, do highlight the somewhat arbitrary nature of how I chose the big conventions, and thus why for the most part we shouldn’t focus too much on any patterns in the market share of the other, smaller, conventions.

##### What about 2017?

While there’s still plenty of 2017 to go, the only big con left is MFF, and so we can get a rough idea of how things may look at the end of the year by using attendance figures for AC, FC, FWA, TFF and BLFC. For this, we’ll have to make a prediction of the 2017 total US convention attendance, which based on my previous post I’m going to set at 49,300. Time for another graph, where I’ve trimmed the time frame a little to make things easier to see. (Another note for the stats fans: we could, if we really wanted to, take the uncertainty of this prediction into account and compute interval estimates, but there’s not really much point.)

There’s nothing hugely surprising here: FC’s share continues to drop, while TFF and FWA post moderate growth. There’s now some evidence that AC’s 2016 growth was slightly out of the ordinary (furthering the case for a Rainfurrest effect), and BLFC shows no sign of stopping, but this figure is more for curiosity than anything else.

##### A fairer measure?

If you’ve made it this far without falling asleep you’re hopefully ready for something a tiny bit more complicated to finish things off. It’s natural to present attendance figures on a calendar year basis, but this overlooks one fairly important consideration. With the furry population (or at least, the population attending furry conventions) presumably seeing continuous growth, calendar-year attendance may be a slightly unfair basis for comparison. For example, MFF (which takes place in November) has a 10 month ‘advantage’ over FC (which takes place in January).

One way to look at this is to instead calculate each convention’s attendance as a proportion of total US convention attendance *since the last installment of that convention*. For example, from the day after Anthrocon 2011 up to and including Anthrocon 2012, there were 22,629 attendees at US furry conventions. The 5,179 attendees at AC 2012, therefore, make up about 24% of all convention attendees since the last Anthrocon. We can calculate this figure for every convention, stripping out the problem that a convention later in the year might have an unfair advantage due to any underlying growth in the fandom’s population.

If we compare this plot with the calendar year market share above (noting that ‘others’ has been removed and the y-axis rescaled slightly), there isn’t a huge difference. What’s slightly interesting is that the 2016 editions of AC and MFF have near-identical year end market shares, suggesting that MFF isn’t deriving much - if any - advantage from being later in the year (at least in terms of underlying population growth). This is good news for me, because this alternative metric is both harder to compute, and rather more questionable (since a convention changing dates can have a major impact on the results).

##### Summing up

There’s a lot that can be taken from the above plots, but for me the main thing is that MFF does indeed look good to take the top spot at the end of this year. My suspicion, though, is it’ll see slightly less growth than it did in 2016, which I’d wager was influenced by the end of Rainfurrest. I should acknowledge that this is largely speculative, and apologize for this post being quite a lot of numbers and not a huge amount of actual statistics. By way of justification, I’m not convinced there’s much value in more formal analyses given the quality (and quantity) of the available data, but graphs are still fun. I also hope I’ve convinced some of you that market share is an interesting (and maybe even useful) measure of a convention’s overall influence in the furry event landscape! That said, I think it’s important not to get too caught up in these figures (there are all sorts of limitations to what I’ve done here, not least the fact that many cons are rapidly outgrowing their current convention space); what really matters is the fandom continuing to grow, and more of us getting to experience the uniquely awesome weirdness furry conventions offer.

As always, comments/suggestions are welcome and you can contact me by email (see ‘About’, above) or on Twitter @tealeafraccoon.