Article Header Image
What's With the Polls?
Legends and Lore
Mike Mearls

My name is Mike Mearls, Group Manager for the D&D Research and Development team. Welcome to Legends & Lore, a weekly column where I write about various topics on D&D’s history, how the game has changed over the years, and where it’s going in the future. One of the things I want to do is pose questions and topics to D&D players across the world, because my real purpose is to understand what you think about D&D. Once I know that, then I can better understand the game.

I thought it would be useful to pause this week and talk a little bit about the polls that have been appearing in this column. Some folks think of them as poorly disguised marketing research. In all honesty, they’re simply an attempt to engage in a dialogue. We already have an entire department here at Wizards of the Coast dedicated to collecting data, running official surveys, and so on. Plus, I also took enough statistics in college to understand that a self-selecting audience is by no means a sound foundation for the sort of polling we’ve been running in Legends & Lore.

Instead, think of this column as something similar to a virtual panel at a convention. It gives me a chance to talk about topics that interest me and, hopefully, you. The polls—and the invitation to send feedback to—are the Q&A portion of the panel, your chance to react and my chance to pose a question. If you’ve been to a panel held by D&D R&D at any of the major cons, you’ve probably seen us ask how many people in the audience are DMs, how many own a specific sourcebook, and so on. Think of these polls as something like that. It’s interesting to see the answers, but we’re not about to base any major business decisions on them.

So What Do the Answers Mean?

When I sat down to write this week’s column, I thought I’d look back at each poll’s results and comment on them. (I’m writing this on March 18th, so keep that in mind.) It’s clear that there’s an interesting trend at work here. It’s something you can see from the very first poll, on the difference between grid-dependent mechanics vs. description of an area/DM’s judgment. That poll was split 50.8% to 49.2%. To put that into perspective, at GenCon we have perhaps 1,000 people playing D&D at any given moment. If those percentages held true, of that 1,000 there would be a 16 person difference between one group and the other.

In the poll asking how much content you wanted to see per month, the top choice at 20% was 97 to 128 pages. About 35% wanted more than that, while 45% wanted less. While the overall trend skews low, that’s still one out of three voters who want more. A similar trend showed up on the poll asking about stripping away fighter options. About 25% of voters were perfectly happy sacrificing options for complexity.

Those numbers might seem decisive. However, looks can be deceiving. Determining how you want to play D&D is not an election, with one winner and one loser. Instead, your tastes in D&D have to mesh with those of the rest of your gaming group. This leads to what I call the gnome effect.

Gnomes, Options, and Groups

The idea behind the gnome effect is simple. Let’s say you’re planning on releasing a hypothetical edition of D&D. You want to determine which races are important to the game, so you conduct a poll and find that only 10% of gamers play gnomes. That might make it seem obvious that you can safely cut the gnome without much trouble.

The problem with that line of reasoning is that we don’t play D&D by ourselves. We play with a group, and when looking at rules changes or any other alteration to the game you have to consider its effect on the group. Let’s look back at our gnome example. One out of ten gamers plays a gnome. However, let’s say your data shows that the average group consists of five players (not counting the DM). That means, roughly speaking, half the gaming groups have one player with a gnome character. That number is likely lower, since some groups might have more than one gnome, but it’s a rough approximation that serves to illustrate the larger principle. You cannot measure change and its effects on the individual level. You must look at it on the gaming group level. Delete the gnome from the game, or change it in a way that gnome fans dislike, and you’ve given about half the gaming groups out there a good reason to tune you out.

With that in mind, we can quickly see how all of the options presented in a poll are important. In an ideal world, we would aim our design work at the most popular options but include the ability to slide along the scale from one extreme to the other. In this manner, you can be assured that in a diverse gaming group everyone has the options they’re comfortable with.

Even a topic such as the volume of content released per month falls into this category. Gamers who don’t want more content can easily ignore it or disallow it in their games. A theoretical D&D release schedule could focus on the middle ground of the audience, while something like the open gaming license would allow other publishers to fill in the gaps for those who want even more content. In many cases, the trick to keeping everyone happy lies in areas beyond game design.

At the end of the day, to me the polls show that we have a fairly diverse audience that wants a wide range of experience out of Dungeons & Dragons. Whether discussing our publishing schedule, the game’s complexity, or its use of miniatures, we see a fairly broad distribution of opinions. There are clusters, but so far we haven’t seen a poll that saw one option (or set of related options) overcome the gnome effect.

On that note, here are the resultas from last week's poll....

Legends & Lore: Poll 03/15/2011 Results

4th Edition is:

  • Just right: 67.9%
  • Too complicated: 16.8%
  • Not complex enough: 11.7%
  • I have never played this edition: 3.6%

3rd/3.5 Edition D&D is:

  • Too complicated: 63.2%
  • Just right: 25.3%
  • I have never played this edition: 8.2%
  • Not complex enough: 3.3%

2nd Edition D&D is:

  • I have never played this edition: 27.2%
  • Too complicated: 26.7%
  • Just right: 24.4%
  • Not complex enough: 21.7%

1st Edition D&D is:

  • I have never played this edition: 43.4%
  • Not complex enough: 27.2%
  • Just right: 17.9%
  • Too complicated: 11.6%

Basic D&D is:

  • I have never played this edition: 44.1%
  • Not complex enough: 37.3%
  • Just right: 17.2%
  • Too complicated: 1.4%
Mike Mearls
Mike Mearls is the senior manager for the D&D research and design team. He led the design for 5th Edition D&D. His other credits include the Castle Ravenloft board game, Monster Manual 3 for 4th Edition, and Player’s Handbook 2 for 3rd Edition.