Dunbar's Number is 150, which is the "average maximum" number of friends with which a typical individual can sustain meaningful social relationships. Yes, this is how many actual friends you can have. None of this Facebook friends nonsense! The number, which is actually 148 if you don't round up, is based on a statistical analysis of brain size and social group size across various species of primates, along with other sources of data. It is believed to be a cognitive limit, based on the assumption that the human brain evolved mainly to manage and keep track of social resources and alliances. Dunbar’s number is an extremely interesting idea. I'm not sure I entirely believe it- and it ignores the issue of individual differences in cognitive ability- but I find it intriguing. In fact, I've spent plenty of time wondering if I could come up with my own number about something, but no luck so far.
Robin Dunbar and his colleagues have further suggested that various levels of relationship exist within a nested hierarchical structure. The smallest subgroup is 3 to 5 individuals, which corresponds to our closest friends and family members. Beyond that is a "sympathy group" of 12 to 20 individuals who we see on a regular basis. Beyond that is another clustering of 30-50 individuals which is a typical "band" of hunter gatherers, according to anthropologists. Then we have Dunbar's number of 150 "friends," followed by a larger group of acquaintances of roughly 500 people, followed by the "tribal" group of approximately 1500 individuals. Beyond that, people are just strangers to each other. Note that these levels differ by a scaling ratio of approximately 3. Coincidence?
This research has been influential and somewhat controversial, but the bigger question that has inexplicably been ignored by scholars up to this point is this: How does Dunbar's principle apply to the board gaming hobby? The core, innermost group of 3-5 people is just right for an afternoon of gaming. This might represent the ideal number of close gaming friends. Any more than this, and it starts to get difficult to schedule and organize a session of gaming. The range of 3-5 also happens to be ideal for playing most games, and about the right number of people to fit around most tables. Coincidence?!
You might have a small group like this that meets regularly, probably in somebody's home. Another way to do it is to have a larger, less tight-knit group that is more likely to meet in a public place, such as a library or a coffee shop. How big is this group going to be? There is a group near me that meets twice a month in the local community center. Attendance varies, but it is usually enough to have two or three games going at once. These people are all friends, but they are in that next layer out, perhaps that sympathy group of 12-20.
As you keep going outward in increasingly larger social circles, friendships become less close and intimate. These are still people you know and are willing to play games with, but I would imagine that you are less likely to loan them games, or invite them over for dinner. I don't know for sure, but I would imagine that Alan Moon's Gathering of Friends maxes out at about 500 people. It's by invitation only, so it is a different kind of event than BGG Con, or really any other gaming convention you might name. At a certain point, you don't really know people anymore. Interactions are governed more by rules and broad social norms, rather than informal friendship arrangements.
Dunbar's Number of Games?
I feel fairly confident that Dunbar's number applies to different kinds of gaming groups and meetups. I wonder if it has any relevance to the number of games you can have a relationship with. Sure, you can collect games and own hundreds or even thousands of titles. But I don't think you will be able to play them all, much less know the rules to all of them. You don't really have a relationship with them if they are just stored in a closet somewhere. These games are like all those Facebook friends we never interact with. Why do we even have them?
If we follow Dunbar's reasoning, we might be expected to have a core group of games which we know well and play often. These are our favorite games. How many is this going to be? I don't know about you, but I can think of maybe twenty or thirty games that would fit this description. That's something like the sympathy group level in Dunbar's taxonomy. If I go higher than that, I have games that I don't play as often, that I might have to read the rules for if I were to play them again. They are in my collection, but they are just friends, not my closest friends. For me, this is about 100 games. I actually own more games than that, but a lot of them are just taking up space.
Is there any way we can calculate a maximum number of games that would correspond to Dunbar's number? This could be useful in helping people to avoid excessively bulky and expensive collections, which is a frequently discussed problem on BGG. In fact, Daniel Karp recently posted a GeekList on how to keep game collections manageable. He has kept his collection at about 300 games by following the policies described in his list, even after many years of playing and buying games. That actually seems like a pretty reasonable number. Obviously if you go much higher than 300, it becomes difficult to play them all every year, assuming an average rate of one game per day, which is a pretty high standard.
I did an informal survey on this recently and found that the median number of games people have in their collection is between 125 and 150 games, whereas the optimal number of games people say they would like to own is between 150 and 175. Could we actually use Dunbar's number here as our frame of reference and call it an even 150? It's worth noting that there was a huge amount of variability in people's responses.
Another way to go about this is to look at the average number of unique games people play per year. If I don't play it every year, do I really need to own it? That seems excessively strict. How about every three years or every five years? We don't need a precise number. Just a range that could serve as a guideline for what constitutes a reasonable collection.
I'm going to keep thinking about this. If you have any ideas, let me know. I'd really like to call this Butler's Number, all right?
Statistics and Speculations on the Behavioral Science of Board Gaming
11 Jul 2018
- [+] Dice rolls