Note: this content has also been posted on my blog, playing and designing board games.
Okay, so I was having a bit of fun on BGG today, trying to figure out how to get at some of all their awesome data. I mean, yes, you can pretty easily sort by rank, or the number of voters, etc., but did you know you can sort by number of owners? Or by number of people who have the game on their wishlist? By the number of people who want a game in trade?
Anyway, I got to playing around a bit with their stats, specifically the number of plays each game has. And I found something interesting: ignoring the number of plays (since this data can be easily corrupted), you can sort by the number of unique users who played the game. Here’s a link that lists all the games in their database, sorted by the number of unique users who have played the game: http://www.boardgamegeek.com/plays/bygame?sortby=distinctuse.... Pretty cool, huh? Who knew that Carcassonne would have the most unique players, followed by Dominion, Agricola, Puerto Rico, and The Settlers of Catan?
Here’s where it gets really interesting, though: you can limit the time-frame you’re looking at, so you get just the number of unique users who played the game, say, in the year 2000.
You can tell where I’m going with this, can’t you?
First, a link or two: if you want to see how many unique users played Settlers of Catan in, say, 2000, just click here. And you can obviously do a little url hacking to change the year from 2000 to 2001 (that’s easier than picking a new range of dates with their chooser and then choosing to sort by number of unique users, though you can do it that way, too).
And you can go through and do this for 2000, 2001, 2002, all the way up to 2010. What do you get when you do all that?
You get a bunch of web pages, that’s what. But if you look at the top ten games of each year, some interesting trends emerge. And if you look at every game that hits the top ten in that period (there are 34), tracking it until it no longer cracks the top 100, you get a rather interesting little graph. But one that’s very hard to read:
A couple things to note: first, each game is just stacked on top of the next. Second, this looks an awfully lot like exponential growth, at least up until 2010. This might represent the growth of the hobby, but more likely it’s just the growth in BGG users / users who record their plays. Third, I’m not sure why things tail off for 2010 — probably it’s just that fewer players recorded their plays that year (for whatever reason).
Playing with it a little, you can “normalize” the graph by converting all the numbers to percents. Here all that dramatic growth goes away, replaced by an easier-to-read graph where each game gets a slice of 100%:
A caveat: of course there are a number of ways this data might be skewed. First, as stated above, the number of users on BGG has been increasing, changing the nature (and the preferences) of the BGG population. Second, not all users record their plays. Third, there are an unknown but possibly large number of shill accounts. And so on.
Still, I think this is fairly interesting. It gives a pretty good idea as to what the big (popular) names in gaming are, and it does so in an easy-to-read way.
I’d upload the spreadsheet I made so other people can play with the data, too, but Wordpress appears to be unwilling to allow me to post an ODS file. ODT, yes, but not ODS. Anyway, I can post an ODT file with a table if folks are interested. Just let me know.
Edited: title, wording of caveat.