Games Talks: Measuring Convention Attendance

I stumbled on a very interesting article from Gil Hova on his blog Formal Ferret Games. He is talking about how statistics for conventions come to live and are interpreted.  Original on the link, here is the copy/paste for easier reading:


Anyone who follows the board game convention scene knows that there’s two ways to measure the attendance at a convention. But they’re often used interchangeably by accident, which results in a lot of confusion and head-scratching.

One way is “unique attendance.” This is the number of different people a convention has sold tickets to. So if I buy a ticket to your convention, I count as one unique attendee. Simple enough.

But the more common way conventions report attendance is by “turnstile attendance.” This counts how many people attended the convention each day. So if I attend your convention on three different days, I count as three different attendees.

Conventions love to report their attendance as turnstile, because the number is significantly bigger. For example, if I wanted to misrepresent the difference between Origins and Gen Con, I could report that Origins drew 52,561 people in 2016, while Gen Con narrowly beat them with 60,819. But that statistic is a distortion; the figure for Origins is turnstiles, while the figure for origins is uniques. An apples-to-apples comparison would either report both unique figures (Origins drawing 15,479 to Gen Con’s 60,819) or both turnstile figures (Origins drawing 25,149 to Gen Con’s 201,852).

So let’s take a second and disclaim: none of this is to measure the relative worth of these conventions. I’m not arguing for a moment that Gen Con is almost 10 times better than Origins, since their turnstile attendance is almost 10x greater. Far from it! I’ve had amazing times at smaller conventions, and not-so-great times at larger conventions. Size will change the feel of a convention, but it should never, ever be a sole determining factor in judging one show’s worth versus another’s.

Also, I’ve helped run the board game track at a 300-person convention years ago, and even that is a backbreaking amount of work. All of these conventions require an amount of labor that would make anyone’s hair fall out. Smaller shows usually get by with smaller crews, so I tip my cap to people who run conventions of any size.

That said, it would be useful to have a single metric to measure attendance at these shows. So why not just use unique attendees and ignore turnstiles?

There’s a couple of issues there. It’s rare that a statistic is always useful or never useful. Statistics provide a window into our understanding of an event, but they will usually miss a bit of the picture.

For example, it’s tough to compare conventions using turnstile attendance because not all conventions run the same length. Two conventions may draw about the same crowd per day, but if one convention is 5 days long and the other 3 days, comparing turnstiles will be extremely misleading; the first show will seem significantly larger.

Uniques would seem to be better-suited to compare conventions with, but they can be problematic too. First off, lots of European shows (especially Essen SPIEL) do not release unique attendance figures. Second off, not all shows count uniques the same way. I’ve noticed some shows only count event-long badges as uniques, ignoring day badges.

Third off, if one show has different attendees each day versus another show where the same people return each day, the first show’s unique figures will be much higher, even if both shows draw in roughly the same number of people per day. That’s not to say that uniques are a useless statistic in that situation, but they will not help you judge the relative scale of each event.

So how about another way to calculate attendance? Let’s take a look at convention attendance using this simple formula: turnstile attendance ➗ days.

This gives us a unique window into the scale of a show. It lets us compare conventions that don’t provide unique attendance against shows that do, regardless of the number of days it runs. We don’t have to worry how the show calculates uniques, as turnstile will include all kinds of badges, not just full-event badges.

And as an exhibitor, it gives me a valuable data point: it tells me what the scale of the show is, and how big the crowd should be on an average day. Sure, it’s not going to tell me how large the attendance swings are; if the last day of a show is particularly slow, that will be represented as a dip in the overall statistic. But that’s the risk of summing everything up in a single number (and anyway, most shows do not release daily turnstile figures).

So with that, let’s look at some TPD (turnstile per day) figures! I’ve taken 9 of the largest international conventions with at least some significant board game attendance, with the largest turnstile figures for each individual show. Some shows are board game or tabletop-only (Gen Con, SPIEL), while others are comic book or video game conventions with a fairly good board game board game/tabletop presence (Lucca, PAX, Emerald City). I’ll also opine, when I can, on whether I expect each show to grow significantly or not.

One more thing: convention attendance figures are notoriously unreliable. All of these figures must be taken with quite a few grains of salt, if not the entire shaker!

Convention Year Turnstile Attendence Length (days) TPD
Lucca Comics & Games 2016 271,208 5 54,242
Gen Con 2016 201,852 4 50,463
Festival des Jeux 2014 150,000 3 50,000
SPIEL 2016 174,000 4 43,500
Emerald City Comic Con 2015 80,000 3 26,667
PAX East 2017 80,000 3 26,667
PAX West 2015 70,000 4 17,500
Origins 2016 52,561 5 10,512
UKGE 2016 25,149 3 8,383


Lucca turns out to be the largest show by this metric, even larger than SPIEL in Essen. That’s a bit surprising, but not hugely so; everything I’ve read about Lucca tells me that it’s a madhouse. Keep in mind that it is not exclusively a game convention, so there will be a significant percentage of attendees attending only for comics and/or cosplay.

Gen Con comes in second, and that’s not a massive surprise. Gen Con is not purely board gaming; it has a significant RPG and CCG presence. But board games seem to have grown lately, and I’d imagine that most exhibitors and floor space is dedicated to board games at this point.

Regarding the future, I can’t imagine Gen Con growing significantly larger. Both floor space and hotel room stock are maxed out, and yet they’re locked into their current location for the next few years. (Of course, bigger is not necessarily better, and a show can easily get worse if it doesn’t manage growth well.)

You may not have heard of the Festival des Jeux. It’s the largest board game convention in France, held in Cannes (yes, the same city as the well-known film festival). In fact, some people will call it “Cannes” the same way SPIEL is referred to as “Essen.” What makes the FdJ notable here is that, according to TPD, it is the largest board game-only convention in the world.

How come so many geeks haven’t heard of it? Probably because it’s very heavily French-speaking. Most attendees and publishers at SPIEL speak English, but FdJ is meant to be a show for France, not so much an international show, at least as it stands right now. Still, if you’re a publisher with a French game, or if your game is language-independent, you have a French rulebook, and you parlez-vous français, it seems a must-attend.

Note that I was only able to find attendance figures for 2014, and that nice round number tells me it’s an estimate, so who knows what the real attendance was? I hear that this year’s show had a turnstile attendance of 200,000, but I couldn’t find any articles to substantiate that. If anyone can point me to a more reliable attendance number for FdJ, I’d hugely appreciate it.

Speaking of SPIEL… who’d have thought it’d place fourth? It’s one of the most highly-regarded shows in the world, after all. But keep in mind: SPIEL is still the second-largest board game-only show in the world (well, mostly board games, but board games make up such an overwhelming percentage of the show that we’ll go ahead and grant it). Also, the fact that it’s very friendly to English speakers means that it has much more of an international flavor, and is much more appealing to people from English-speaking countries. (But I wouldn’t suggest planning to make a ton of money selling a game at SPIEL that’s exclusively English-dependent. This might be experience talking here.)

It’s a shame SPIEL doesn’t present unique attendance figures. I think they’d be fascinating. Anecdotal evidence seems to indicate that Essen would have a surprisingly high unique attendance figure; there are many attendees that attend just for a single day. I would love to compare uniques to TPD, to figure out how much turnover there is from day to day.

As for growth, it’s not unreasonable to imagine SPIEL growing even larger. There’s still about 1-2 empty halls in the Messe, and Essen has a good amount of hotel stock. Assuming the hobby continues to grow, I would expect it to overtake Gen Con soon.

Then we return US soil, with both Emerald City Comic Con and PAX East. Be warned that both these attendance figures are fraught with peril. First off, neither show is exclusively for board games. I’ve never been to ECCC, and while I hear its board game scene is pretty healthy (rare for most comic conventions!), I don’t have any firsthand evidence to back it up.

PAX East, meanwhile, is a video game convention that happens to have a decently-sized board game convention neatly nested within. But PAX is notoriously tight-lipped about its attendance figures, and normally doesn’t even release turnstile statistics. I found these attendance figures on Wikipedia, without citations. Caveat emptor!

With all that said, I was surprised that PAX East’s number was so low. I was expecting it to be on the same scale as Gen Con, but its TPD is about half the size. Once again, that’s not to say that PAX East is an objectively worse show than Gen Con, but I expected its numbers to be a bit larger.

I wish we could have numbers for how many people attend the PAX East tabletop area, which is larger than a lot of entire regional conventions. Just from my experience, I’d peg its TPD at about 5,000, give or take 2,000 people. As Father Guido Sarducci once said, “That’s nothing to sneeze your nose at.”

Immediately after PAX East is PAX West (formerly PAX Prime). The fact that it’s a little smaller isn’t a huge surprise, although it’s tough to get a sense of PAX West’s size just by being there. The Washington State Convention Center is split into several floors, instead of being one or two giant rooms, so you can never get a sense of how large the overall show is.

After that comes the venerable Origins Game Fair. It’s not a massive show, but it’s large enough to boast a 5-figure TPD, which is nothing to be ashamed of. Its size relative to Gen Con is not a surprise either, as both shows are very good about releasing both turnstile and unique attendance.

Origins’ unique attendance dropped a bit in 2016 relative to 2015, but that was mostly because of a CCG tournament that’s no longer held at the show. The TPD shows significant growth, up from 7,843 in 2014 and 8,758 in 2015. I’d expect the show to grow for the next year or two, possibly hitting a floor space limit within the Greater Columbus Convention Center.

The last entry, and one of the most intriguing shows for me, is UK Games Expo. I’d suspected that its size put it on par with Origins, and these numbers seem to back that observation up. One thing to keep in mind is that UKGE is growing like crazy. I expect it to easily shatter the 10,000 TPD barrier this year, and it’s not unrealistic to expect the show to get to 15,000 or even 20,000 in the next few years, if the growth of the hobby keeps up. That’s because the convention is only taking up a couple of halls of an enormous convention center. In terms of floor size, they could get much larger without having to move locations. Their biggest gating factor is hotel stock; rooms sell out faster each year.

And that’s about it! I don’t pretend that this metric is a be-all end-all statistic that replaces both turnstiles and uniques. As a vendor, I do want to know how many total unique attendees there are, and therefore how many potential customers I have. But I think TPD gives the best idea of the scale of a convention, and roughly how many people you can expect at the show on an average day.

Incidentally, I’d love to see turnstile figures for BGG.CON, Dice Tower Con, UnPub, and Geekway to the West. As those shows grow (and all those shows are growing – even BGG.CON is moving to a larger location in a couple of years), I’d be interested to see how close they get to Origins and UKGE in terms of scale.




Leave a Reply

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

You are commenting using your 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 )

Google+ photo

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

Connecting to %s