You hear it all the time, no matter what the creative endeavor is: before you go out and show it to the public, get some feed back from other people. For art, this involves showing it to them and asking what they think; for books it involves beta readers; and for games, play testing. Now, at present I've done a few Play Tests of my big WIP (mini campaigns), and one or two of NINJA! (and waiting on word from others), but I didn't realize how helpful some of the resposnes would be until I started to implement a couple of the changes I'd heard suggested. So, for today, let's talk about Play Testing.
Involved and Uninvolved
The two basic types of play testing are involved play testing, and uninvolved. Now, I don't know if these are official terms for it, but it's what I call them. The names are fairly self explanatory, but basically you - the designer - are involved in the Invovled type, and are not at all involved in Uninvolved. Doing both is very important for a variety of reasons. Those reasons, at least that I've seen, are as follows.
Involved: Involved is important, because it lets you see the game in action. You really want to do play tests like this from both sides of the GM screen. Why? As a GM you can see how it works and what players will do with the tools that have been presented to them. Maybe you put a system in for hacking bank terminals, and your player uses it for robbing convenience stores. Is that good? bad? Up to you and your game really, but it's not something you're going to find if you don't play the game. As a player, you get the experience of the game from the player's side of things, as well as get to see how a GM interprets the rules and rulings presented. Are they hung up on things? Do they just not get how things work? Also, as a player, you can use systems the way they were intended to be used and see how they hold up there. Do you get a feeling of power where you're supposed to? Or does it slow you down? Can you find a faster way to do it? Try to break your own system, ask your players to try and break the system, and see where it holds up while you watch.
Uninvolved: With uninvolved play testing, not only are you looking for all the same things as before (be sure to ask for responses from people on what happened) but you're also going to find something that is harder to find in involved playtesting. Namely, points of confusion. See, when you're sitting at the table, you can explain how the system is supposed to handle something. Did you forget to include a small section in your "Damage Recovery Section?" well then, you can just tell the people at the table how it works. However, what if you're not at the table? Now you have a point of confusion with no easy work around for it. You will also find points where you aren't clear enough, and other fun stuff like that. The level of detachment can also bring up things you wouldn't hear from other people. Some of the strongest critique you'll get will be from the people who can't see you, and never will have to, and that makes their feed back just so awesome to have.
Now, you've done your play tests, ready for the hard part? Now you have to read that critique. Ask the people involved to write up (and yes, you want it written) what they liked and what they didn't like about the game. Take the responses, and set them aside for a week or so. Then, read them and read them carefully. Look for the points the players didn't like, and see if they said why they didn't like them. Do the same for areas they liked. This gives you a basic feel of what the player thought of the game.
Now, the trick here is to not just go and take all of the critique. Maybe someone didn't like how hard it was to get their fighter into melee combat, but if it is a game about magic or using guns all the time, that might be part of the idea you're going for. Same with someone who doesn't like how unrealistic the gunplay is in a game designed to capture the feel of 80's action movies. Basically, read the feed back, but don't necessarily take all of it as gospel. This is still your game.
Now that you've looked over your feedback, it's time to implement it. This is where a recent batch of feed back rocked my world. One of the play testers from a test game run months ago just recently wrote to their GM with their feed back. In their feed back they had a very simple question "It seems obvious that this stat should be starting at 1, 2, or 3 at these different power levels, so why not just do that instead of having to use complex math?" It's simple, right? And yet, no one had asked me that to my face. Some people had just accepted it, others in my group are just like me and thought it was perfectly fine. But the change (along with a few related changes that'd go with it) makes things a lot simpler, and seems to speed up character creation. Not by a ton, but enough that I like it.
In the end, most of what you'll probably get out of play testing are little changes. Ways to fine tune the experience, or get things working smoother. However, it is possible that you'll find some aspects of the game that are just flat out broken, in which case you'll need to do bigger changes. Either way, try to keep some distance, and remember what you're working towards: a better form of your game.