So last night we got a reduced team together for a pre-Canada-Day game night. We didn’t have enough to do any serious play-testing of Soft Horizon (which was the original plan) and I didn’t really feel like working with Hollowpoint since Toph has upstaged all actual play with his son, Jonah. So we pulled out a Random Indie Game that we hadn’t played but had read and that I thought highly of.

I’m not sure why I thought highly of it, now, except that it’s beautiful. It’s really pretty.

In trying to actually play it, we found the rules remarkably vague. I mean, they sound clear when you’re reading it for fun, but when you actually try to extract a game out of it, it fails. I’m pretty sure some of our writing has failed the same way, because when we write we already know what we mean. It tales a third party with no pre-conceived ideas to really notice the failures. We’ve already seen this in some of the third party play-testing we’ve had for Hollowpoint so far, where readers have found really deep failures in explanation that it knocks me over to have missed. But you can’t see in from the inside.

The other thing that failed, once we had a grip on the rules (kinda), was that despite the presence of a lot of mechanism for play, none of those handholds and dials and levers actually helped us have any fun in the game. The game is hard to play. It expects a great deal from every player and if you’re playing badly it’s not fun. So it would be nice if some of the mechanism we invested time learning actually helped make the hard parts easier, but it doesn’t. In fact for the most part the mechanism didn’t do much of anything that flipping a coin wouldn’t do.

To be fair we didn’t play it wrong. Maybe it gets a lot better. There were things about it I like enough to steal, but they are all surrounding prep and character definition and nothing about play.

This isn’t a review, so I’m not naming the game.

It’s cool by me if some games don’t work all that well. I don’t mind buying a game and having a crappy time with it because, as a game developer, I learn from others’ mistakes at least as well as I learn from my own. Probably more, even. So I don’t resent buying it and even before playing it I got a number of ideas just from the layout and typography. Hell I’ve already stolen some of those ideas. It has reinforced a few ideas I’ve suspected were true, and it’s always nice to be right.

First of those is that third-party play-testing is a lot more important than I thought it was. It not only reveals pedagogical failures, but also exposes parts of you game that you think are awesome but are in fact only the awesomeness of your group — seriously, if you play-test only with super-creative and enthusiastic people who already get all there is to get, then you risk having a great time with a game that isn’t what you actually sold anyone. The game, as some say, is at the table, embedded in the people and the food and the drink and the mood of the night, and finally the rules. So if you have everything else going right, the rules might not be delivering all — or even most — of that fun you had.

Second is that the text must teach. It must also act as a reference and I’ve talked before about how hard it is to get both right. They are orthogonal and so if you get it right you’re pretty damned smart. How does teaching text fail? Well it fails in ways that are not obvious when you already know your topic (and so that feeds back into the need for play-testing, obviously). Things like consistency of terminology and repetition: you need to use the same word for the same thing all the time (and not until after you’ve defined it) and you cannot shy away from stating the same rule each time it needs to be stated — referring back is second best, but assuming a rule is understood (or has even been read) by the time its second or third usage is actually broken, because during play the learner will open the book in the middle (and with some desperation) to find clarification. You better give it then and there.

Finally, a pretty book and the accolades of trustworthy people are enough to sell a lot of books. That suggests that a lot of people never play the games they buy, and I don’t think that’s actually all that controversial, but it’s kind of disheartening when you put a lot of effort into delivering a functional text and then realize you could have sold five times the copy if you’d ignored the text and bought more art. You can (and I do) tell yourself that it’s better to be read and played than to be read and displayed (as though that’s a dichotomy — there are plenty of games that manage to fire on all cylinders, delivering beauty and game without compromise) but it’s only partly true. You only know what actually happens to a very small percentage of what you sell, so really, it’s better to sell a crapton of anything you’re proud of.

–BMurray