What is playtesting exactly?

There's nothing that you "need" to do, but of course it depends on your goals as a designer. If you just want to make a game that is fun to play with your friends, and the game you've designed is fun, then you're already there!

But when we speak of playtesting here, we're not talking so much about "do they like it?" which is an important question, but about that more important question "does the game work?" If the answer to that one is "no", then whether they like it or not, it doesn't matter -- the game will go no where.

There is all sorts of conventional wisdom about needing to get your game "blind tested", and all sorts of good reasons why you need to do that -- test your rulebook without you there to teach and explain the rules, get the opinion of people who aren't biased by being friends with you, etc. I won't treat you to all that conventional wisdom here.

I'll simply observe that playtesting with an eye towards developing a working game system that has no obvious holes, that can't be broken by players making moves that you hadn't anticipated, that the game doesn't have a lot of common design flaws, etc, is essential, and is WAY more important than "do they like it?"

My impression based on hearing a lot of first-time designers, and having once been one myself, is that everyone's friends likes the games that they make. And immediately, the designer thinks, "how can I sell this?" The designer feels his game is the newest and best thing in the world.

I've heard of no cases yet where that turned out to be true -- my own included!

The reality is, just like anything else, game design requires patience and growth. And so, interestingly, does playtesting! I've tested several games with my regular group, and just as I'm getting better as a designer, they're getting better as playtesters -- giving me feedback beyond "this is fun" or "this is not fun" and more like "this system needs to be modified, because it's horribly imbalanced". And, they're much more brutally honest than they used to be, because they know they can be/need to be! I've also found that I am in general my own most reliable critic -- that ultimately, I need to be the one to point to the problems in the game, and try to hit on a solution to it. This also takes time.

So, my advice to new designers (speaking as an only slightly experience designer myself!) is first to play a lot of games, and if you can't play, read about some games using web resources like www.boardgamegeek.com. Identify what you like and don't like. Find out what is "great" about games that are described as being "great." Then, start designing. Expect to throw away a lot of your early designs, or at least need to revise them pretty heavily. That's ok! When you're happy with something you've got, start talking to people about it. Don't be hyper-secretive. No one wants to steal your idea -- most people have their own. Consider putting it up on the Game Design Workshop. Get the game tested by someone other than your own circle of friends before you make a big decision like going ahead and self-publishing.

I think there are several well-known examples of games with some mild to serious design flaws that the designer believed in and went ahead and self-published, to bad critical reception. Some will no doubt recover, but some will not. And since it's hard enough to sell a game anyway, you want to make sure your game is a good one!

So, the long and short obviously is, "yes, playtesting is important!" Good luck with game design!

-Jeff

I couldn't agree more, however I am a little bit suprised to hear that from Jeff :)

- Seth

Just a quick one about 'alpha-testing' (you know, when you spend hours each night playing the game by yourself and forgetting to feed the cat)....I use 2 methods:

1) 'mono-strategists' by all but one player. The remaining player can make any choice of action. This checks if any strategy is too dominant and needs tuning.
2) 'free for all', all players choose the (perceived) best action. This checks that all players have viable multiple choices open to them as much as possible.

Addtional, I recommend writing down each "player's" choice of action (and the immediate outcome if applicable) so that you can review the turning points/dominant actions some 4 hours later (when the cat has gone to your neighbours fishpond for its tea).

Dave.

I contemplated using a non-disclosure when playtesting, but then decided against it. For me and the people I typically play-test with, it seemed a little much. That being said, I do communicate my expectations concerning confidentiality to play-testers when they sit down to play. I've also performed remote testing (mailed mini versions of games) for feedback and in such cases have provided my confidentiality expectations in writing to play-testers (via e-mail) and asked them to respond to acknowledge receipt. I didn't have them "sign" anything though.

If I were play-testing at a convention with a lot of people I didn't know, or who weren't referred to me by friends, perhaps I'd be a little more formal about it. One thing I do ask of people when referring playtesters, or in the case of remote testing where people play with neighbors, friends, etc. is that they don't allow anyone that works for a game/toy company to play or see the game. I obviously can't always control such things. I find people typically respect your wishes though and I haven't had any problems yet.

Also when testing, I always test games using a different name than the one I intend to use when the game is produced. In fact, aside from my wife and a few close friends, I didn't disclose the name of my newest game, Coopetition, for over two years. I'm also working on another game and have no intention of publicizing the name until the details of the game are worked out. (A name/trademark, not necessarily the mechanics of the game itself, is something you can truly protect.)

Chip