To those who have designed their own base classes, I’m curious: how do you prefer to go about testing & polishing them?
Do you just run an arbitrary handful of sessions and look at the overall feel? Do you have a particular style of session you like to run – dungeon crawl, wilderness adventure, urban crime, etc? Do you control party compositions – heavy on similar classes (Ninja+Thief+Bard), different ones (Ninja+Paladin+Fighter), or the “classic 4” party? Or do you just eyeball the moves for clarity & completeness and call it a day when you think it looks good-ish?
Obviously there’s no One True Way – especially since DW sidesteps the need for fiddly evaluation of combat effectiveness – but I’m interested to see what people have settled on.
Alex Norris Adrian Thoen n Trenton Kennedy
Having a good eye for design/balance helps tremendously, but that’s something you can only acquire through practice and getting feedback from others.
I’d say my effort spent designing vs. effort spent playtesting ratio is probably 80/20 – I start by making sure there’s nothing bad in the moves I’ve designed (by obsessively going over my own moves and making a million tiny wording changes over 2-3 weeks), then I’ll just play or run the class in a few games (~10 or so four-hour games does the trick) to make sure no issues come up in play (mostly around people who aren’t me understanding triggers/effects, because my wording isn’t always perfect, or around unforeseen interactions between my stuff and other playbooks).
It helps (and I would say it’s necessary) to have played a good amount of Dungeon World, both as a player and as a GM. Beyond that, it’s a lot of playtesting.
A perfect set of playtests would be with different types of players, different types of sessions (as you said, dungeon crawl, city based, etc.), with different group compositions, and different advancement moves chosen.
But it can be difficult to find players willing to do all that. When we were testing all the Grim World classes, we did soooo many one-offs. It burnt a few of us out on playing for awhile (so be careful of that).
For the first playtest or two, don’t be afraid to adjust things on the fly. If a move isn’t working, call for a quick break and tweak it. Something that looks great on paper may not work in play. Likewise, something on paper that looks a little strange may actually be great in play.
It’s super useful to be able to play as the class yourself too, if you have access to another GM. Just make sure everyone’s cool with you trying a half-baked class.
I recommend bumping the playtester up a few levels fairly quickly every so often (in the same session even). Give them a few minutes to pick a new advanced move. Ask them to talk about what they’re thinking. If their choice is easy for them, that can be a red flag. Ideally choosing should be tough for them. That is, tough because there’s a bunch of great options (not tough because nothing seems interesting).
Finally, here’s a few rules of thumb I follow in playtesting (both with tabletop and video games):
1. STFU. Seriously. Be quiet. When this class/game is shipped out, you are not in the box. Players will not have you there to explain things. Hand over the playbook and let them figure it out. When they inevitably ask you a question, respond with something like “How do you think it’s intended to work?” If what they say is incorrect, then you probably have a problem.
2. Shut upppppp. This is worth repeating. It’s fun and helpful to answer questions, right? Sure! AFTER the playtest. Until then, keep your mouth shut. (Obviously, don’t be a dick about it, explain to your players what you’re doing and why.)
3. When someone complains, pay attention but don’t automatically do their suggestions. Where there’s smoke, yeah, there’s a good chance there’s fire. But sometimes playtesters don’t have the full picture or are having a knee-jerk reaction to something. Pay attention, but look for the actual fire, look for what actually might be causing the problem. (And don’t forget, when you see smoke, sometimes players just like to smoke.)
4. Ultimately, it’s your class. You’re designing it. Sometimes people aren’t going to like it. Yeah, it’s great when people love it. That’s an admirable goal. But not everyone is going to dig it. That’s okay. That’s great actually. People have different tastes and different playstyles.
5. Finish the damn class at some point. Iterate on it, tweak it, make improvements, playtest it, repeat a couple times. Then shove it out the door! Nothing is ever perfect. With Grim World, I could have tweaked the classes forever. And I learned a ton from making them, so of course there’s more changes I could make. BUT, if you want it to see the light of day, you’re going to have to let it go eventually.
Okay, that was a lot. Hopefully that helps!
Great Post!
Spoken like a true artist Trenton Kennedy
Alex and Trenton pretty much covered everything already. I find the best playtesting feedback I get is questions. If someone asks a question, then I need to make sure I have been clear. Sometimes that means that means being clear about a move’s intention, and sometimes it means being clear that the choice is the player’s to make.
I like to get my playbooks out in the wild and tested by groups without me there as soon as possible. This is the fastest way to find out what works and what doesn’t as soon as possible. I also expose the playbook to peer review as soon as possible. There are a lot of people without whose feedback my classes would not have been very good at all.