Behind the Larp Census

Behind the Larp Census

29.751 larpers can’t (all) be wrong

On January 10, 2015, 101 days after launching, the first global Larp Census closed to replies. 29,751 responses were logged from 123 different territories in 17 different languages. The data from this survey is freely available via a Creative Commons license at

Barring death, dismemberment, or debilitating drunkenness, the total results from each question will be revealed in a presentation at Knudepunkt 2015. This article goes under the covers to expose the motivations, methods, and madness of the squishy humans behind the hard numbers.

The Beginning

At Wyrd Con II (a Southern California interactive storytelling convention) in 2011, I was out at a late dinner with some friends. Mark Mensch, a longtime boffer larper, asked me what I thought was needed to unify live action role players.

Without missing a beat, I laid out my

Three Big Ideas

  1. A user-customizable larp map-calendar where people can search for any kind of larp anywhere in the world up to a year in advance.
  2. A digital archival repository of larp events—what was run, by whom, when, where, using what system, and any notes or links to further documentation.
  3. A larp census to track all larpers around the world.

I actually mentioned #3 first, but it’s more dramatic to bury the lead.

I don’t know why I said those three things, and I probably had the ideas before I said them, but that was the first time I voiced them aloud.

Regardless, the conversation turned to other matters and never went anywhere. I kept the ideas in the back of my head, however. I repeated them at a workshop session at Solmukohta 2012, where Claus Raasted and a few others offered help in making the map-calendar: which has since been created, roughly, by and

In mid-February, 2013, New Zealander Ryan Paddy and I started communicating via email after he asked the Larp Academia (or International Larp Academia) mailing list for demographic statistics on larpers. He wanted to know if live action role- playing was “popular” and in which countries.

No one on the list had figures beyond their own larp group’s roster or a few isolated surveys from years past, e.g., Joe Valenti of NERO offered a range from “fifty-thousand to two million.” I again floated my census idea and Ryan took the bait. According to Elizabeth Kolbert[1]Kolbert, Elizabeth,“The Big Kill: New Zealand’s crusade to rid itself of mammals,”The New Yorker, Dec. 22, 2014 it is not unusual to find Kiwis with “a cheerful, let’s-get-on-with-it manner” that she claims she “eventually came to see as very New Zealand.” This is good, because without Ryan, I would still be whining about kooky concepts that nobody builds for me.

We get along well and communication between us, while spotty, has been robust. Ryan edits the English language entry on “LARP” for Wikipedia and has a background in psychology and programming, skills I lacked to get the Census done.

Both of us wanted to know the answers to basic questions about larping worldwide: how big is the community, what are its demographics, how long have people played, what are they playing, and why?

We set out to make the Larp Census a reality.

The Grind

The first choice we faced fell between using a prepared polling system, such as Google’s, or develop our own. Ryan said “Google Forms can only receive a limited amount of data (400,000 answers to individual questions); we wanted more. Also, there were several things we wanted to do it that it couldn’t have achieved. If it was up to the job I would have been happy to use Google Forms.” Thus Ryan did the programming for the Larp Census site.

Next we looked for a website host. We hoped to deliver this baby in an academic institution, but they either didn’t reply or replied in the negative, e.g., University of Tampere. We then sought other entities, leading to one of the Big Mistakes (possibly the biggest).

One of the sites I asked to host was They immediately agreed, as they were already considering doing a similar project, but during the negotiation process I withdrew. I worried about protecting the privacy of the respondents and the data.

A massive email list like what the census would generate is gold to larp businesses; but neither Ryan nor I wanted anyone, including us, to make any money off of it. While discussing things with, I sent over a first draft of the questions. This boomeranged back, and badly. We cut off talks in mid-April and eventually bought the domain with money out of our own pockets.

Most of Ryan’s and my time was spent designing the questions, which proved surprisingly difficult. First we had to decide what we wanted to know. I felt that a self- identifying larper’s location, age, gender, and how long they have been larping gave enough information.

Ryan wanted more info (much more), which I quickly agreed with. We split the census into two parts: the first page of questions asked for only the required info. Everything else was optional. Tough decisions and some generalizations had to be made for each inquiry. Plus, each question was weighed for informational necessity against the time it would take to answer it, as we wanted to avoid a too-long questionnaire.

One thing was asked of us a few times, “What is your hypothesis?” But we had no thesis going in, nothing we hoped to prove. We merely strived to gather as much data as possible and turn it over to others to see if it confirmed or refuted their hypotheses. My analogy is that we are farmers harvesting data. It is up to chefs—larp scholars, business folk, and independent researchers—to use what we gather and turn it into dissertations and Power Point presentations.

We sent out two iterations of the motivation questions to a few hundred larpers for comments. The first batch had over 50 questions that we edited to below 30. We also asked as many larp scholars as we could manage (herd like cats) to look over the census and provide feedback. One of the comments we received was that it appeared “too American,” something we aggressively trying to avoid. We remained cognizant of the American spelling of words as well as terminology and larp style emphasis. Our goal was to be as broad as possible, to capture something about every kind of larper, straight boffer action to Nordic arthouse and all in between. But this goal, plus the fact that we were talking to larp scholars who stereotypically ha ve a pedantic viewpoint (not anyone at Knudepunkt, of course!), led to some complaints, which I will discuss later.

Remarkably, Ryan was also setting up the website at the same time. Suddenly in the middle of August 2013, we were blindsided: released their own Larp Census.

The census, in my extremely biased opinion, seemed to be heavily based on the first draft of questions we sent earlier. They used a Google poll form, required respondents’ emails, and skewed it to American larpers, e.g., using the U.S. dollar as the only type of currency, and asking a lot of questions that only made sense to campaign players.

I was livid, and immediately began chewing out the new census, until Jordan Gwyther of proved to me in a private mail conversation that I had given them permission to create their own and even promised support:

Jordan: On the census/survey, I think we should go our own directions. We’ll be launching our own here shortly and will have no problem briefly promoting yours when it is ready. We hope that you will do the same for ours. 🙂

Aaron: Yes, of course!


They received just under 4,000 replies, and, according to their own admission, over 17,000 complaints[2]Larp Census FAQ (English version)—I do believe that is an exaggeration, though. Two weeks after their launch, Ryan and I bought our own domain.

Ultimately, seeing the mistakes they made inspired us to tweak and revise our project and make it as good as we possibly could. We dove back into reiterating questions, testing, revising, etc. We were totally on our own, without any group or organization helping, sponsoring, or overseeing us.

Besides the very generous and dedicated handful of reviewers and translators who worked on the Larp Census, everything else was the work of Ryan and sometimes myself. If you’re going to credit anyone, credit Ryan or the other names acknowledged on our FAQ page5. If you are going to blame anyone, blame me.

Securing translations was also partially prompted from the census. In order to avoid making ours “too American,” we introduced alternate currencies and continued that thought into offering the census in different languages. We really wanted to emphasize the global nature of larping. This was irksome because some words ha ve diff erent meanings in diff erent countries. Ryan and I spent at least fifteen Skype minutes debating the definition of “park,” which isn’t quite the same in New Zealand as it is in America.

After weeks and weeks of iterations— although really it was days of nothing followed by bursts of work and conversation—Ryan finally decided to pull the trigger after most of the translations had arrived.

The Larp Census went live on October 1, 2014, but the big launch occurred October 2, nearly 20 months after we began. What we had wasn’t flawless, but it was as good as we were going to get and still have it out in 2014. By the time translations started, the original questions in English were locked— we couldn’t change a word without asking all translators to change their versions, an odious task.

Here’s a secret: from the beginning I knew we were doomed to fail. There was no way we were going to get every larper on Earth to answer the census or even close to it. But we wanted to get as many as possible. I hoped for 100,000 replies; Ryan, one million.

The Run

Once we publicly announced the census, it almost went viral. Here are the numbers of responses that came in per day for the first week, which made up more than half the total[3]Initial week’s totals provided by private email correspondence from Ryan Paddy:

Date Responses

I was smugly pleased to know that in two days we got triple the responses the other census garnered after running more than a year. Great numbers for us, but we never came close to these initial daily figures again. The server even crashed for a brief time in those first hours: but it was up and running again soon, thanks to Ryan and, probably, because we never returned to that level of activity.

We didn’t have much of a marketing plan, if any. Social media such as Facebook worked best, while the ability to email your friends (once) was hardly used. Ryan and I are both old, so the new-fangled youth methods of communication are lost on us. Plus, we had no budget to do any ad buys—remember, this was just the two of us.

Some translations required minor corrections in the first two weeks, which Ryan repaired with aplomb. We accepted offers to translate the census into Danish, Swedish, Japanese, and Hebrew, though we only completed the first three.

We did give a few interviews on larp sites, and our push was always to larpers and larp groups. I sent press releases to mainstream geek sites like io9, Boing Boing, and Kotaku, but they didn’t reply. If only we had associated with College of Wizardry.

All things considered the run went well even though we didn’t get the amount of responses we hoped for.

The Lessons and Casualties

Irrespective of the data, I learned a few things just from the census existing.

First, there is no way to ever make everyone happy, ever. This should be obvious, but the point was nailed home after we received specific complaints from four people. Two said the census skewed tow ards boff er combat, and two said it favored theater-style.

It even prompted one newcomer to write, “I’m a little turned off to larping as a consequence of filling out this survey.” By making sure every larp style was represented, we shrunk the spotlight on one person’s particular larp preference, which, to them, seemed like a slight.

Second, race and racism are not the same in America as other countries. On the first page of the census we asked respondents to self-describe their race or ethnicity. I don’t know how it translated out of English, but the question upset a few people. Even asking about race offended them.

On the other hand, for many Americans, to not ask the question would be seen as racially insensitive. Although it appears to be a Catch-22 situation, I hope to repair the issue in subsequent censuses with the phrasing “Please describe your racial and/or ethnic heritage. We understand this question may be offensive to some, and it is not our intention to do so. You can refuse to answer.” Or something equivalent.

Third, and more positively, the census is provoking exactly the kind of discussions and issues we hoped it would. A long thread on LARP Haven spun out of Christopher Amherst’s analysis of the preliminary American statistics[4]Amherst, Christopher Preliminary Analysis of American Larp Census data. The original poster noticed the male-female ratio in the U.S. is roughly 60% – 36% (with about 3% genderfluid or not answering) and wanted to know why more women weren’t participating in larps. A boisterous conversation ensued.

Although I am aware of the dangers of relying too much on statistics, especially ones pseudo-scientifically generated, having nearly 30,000 larpers respond to the Larp Census will at least plant a few guideposts toward a deeper understanding about this art, hobby, or sport we enjoy. I am proud to know that our Census will finally provide some factual basis to confirm or refute a few Internet arguments while spawning hundreds more. This, I feel, is a Good Thing.

By the way, we’re going to ask if you consider larp to be a sport, hobby, or art in the next version, coming up in about five years.

For more information and to receive the data from the Larp Census, go to or find us on Facebook.

No one on the list had figures beyond their own larp group’s roster or a few isolated surveys from years past, e.g., Joe Valenti of NERO offered a range from “ fifty-thousand to two million.”


This article was initially published in The Knudepunkt 2015 Companion Book which was edited by Charles Bo Nielsen & Claus Raasted, published by Rollespilsakademiet and released as part of documentation for the Knudepunkt 2015 conference.

Cover photo: “Hollerith Census Machine pantograph” by Marcin Wichary is licensed under CC BY 2.0.

References   [ + ]

1.Kolbert, Elizabeth,“The Big Kill: New Zealand’s crusade to rid itself of mammals,”The New Yorker, Dec. 22, 2014
2.Larp Census FAQ (English version)
3.Initial week’s totals provided by private email correspondence from Ryan Paddy
4.Amherst, Christopher Preliminary Analysis of American Larp Census data


Aaron Vanek
Aaron Vanek survives Los Angeles with the help of his wife Kirsten, also a larp designer, and Missekat, a pussyfoot named in honor of Nordic larp. For a longer bio and free stuff, check out
%d bloggers like this: