Playtesting

A game will need many (maybe hundreds) of playtests. This is how I get the most value out of each playtest.

Design with playtesting in mind

Your game must be playable two-player. If you thought it was hard to get one playtester, try getting two (or five) at the same time.

If your game won't work properly with two players, don't make that game. This rules out most voting, trading, negotiation, and hidden identity games. You didn't need to wander into those specific genres anyway.

Make sure there's nothing stopping your game working with higher numbers of players, but leave it until late in development to start actually creating the components and structures for those extra players.

Tabletop Simulator

Once you have your online playtesters, you'll need Tabletop Simulator (TTS). That's what almost everyone uses for playtesting online. It costs $20, and simply provides a 3D space containing a table. You can import cards and other bits into the space, and play all kinds of games in there. It's not hard to use.

I use Discord, to chat with people, and we playtest each other's games. Feel free to get on a Discord server, and ask people more about Tabletop Simulator.

I haven't done real-life testing since 2017.

Rules

No playtester wants to read a rulebook. Just have a tiny sheet, with a very minimal version of the rules on it. Also, have a reminder card, so players can look up the game's icons.

Offer to either explain the game, or have the player read the sheet. Some people strongly prefer one over the other.

Treatment of playtesters

Be as nice as possible to your playtesters.

Don't waste a playtester's time.

End the playtest before they get tired of it. They'll appreciate that. Just test long enough to get a feel for the game. For the first few revisions, I will usually not complete a game.

Always let them off, in gameplay. Let them take things back. If a rule isn't clear, let them adjudicate it. If they get screwed by some design flaw, just annul it. Until they start getting good at the game, always let them win (or almost win).

Do not disagree with their feedback. You can explain to them, if they've misunderstood something. Otherwise, just take it in. When they do give a piece of good feedback, identify it, and make it clear that it's a very good idea.

Thank them, at the end of the playtest.

Always ask yourself "why is this person helping me? What are they getting out of it?" They don't have to help you.

Unless the arrangement is reciprocal, they're doing you a huge favour. 

Cycle through your playtesters. If a playtester has a bad experience with your game, make sure it's in a good place, before testing with them again. Test with someone else.

When you make major changes, or when the game has problems, test with your best-known playtesters. They'll be more forgiving.

At the start and end of a project, you need a small group of experts. There's no point joining someone in at the end, to polish a game they've never played before. You can't make larger changes, and that's what they'll suggest. Only in the middle of the process, do you want a wide swath of opinions, as the game finds its feet.

As with any group, playtesters will drift away, and you'll need a constant stream of new recruits.

If they're a game designer too, and your playtesting is reciprocal, most of this doesn't apply.

Managing playtesters

You must tell your playtesters what to do, and guide them.

Ask them specific questions, and let them know what you're looking for. They do not know how to be a playtester yet!

You want to know about their experience with the game. They are not a technician, who works on your game. They don't know this though.

What you don't want:

• Attempts to break the game (unless it's a very late-stage design).
• Power and balance opinions. Balance is always off, early in design.
• Musings about what might be a problem, but was not actually a problem during this playthrough.
• Solutions.

I did a one-off playtest with another designer. After I explained the rules, he turned the deck over, and began reading through every card, so that he could play optimally.

What you do want:

• The playtester's feelings — what they did and didn't like. How they felt.
• Problems they ran into.
• Ideas.

Homework

Your regular playtesters will often be happy to give you ideas, via message. I ask for ideas for content: new actions/abilities/cards. I use about 5% of these. I tell the playtester this. I also tell them to spam me at any hour, with any half-formed idea that comes to mind. It's very easy to parse all this.

Negative feedback

Positive feedback is fantastic, but I can't usually act upon it, unless I'm not sure where I'm going with a game.

Negative feedback is really the only useful feedback.

Negative feedback is a gold nugget of new insight.

Prompt the playtesters for negative feedback. "What part was the least fun?" is a great opener for negative feedback.

Don't worry so much about player confusion. Prototypes are hard to understand, and you haven't smoothed everything out yet.

If a playtester can identify a problem, that's the best feedback. They most certainly do not need to fix it. They'll start to brainstorm solutions. You can do that much better yourself. They've already done the hard part.

Finding solutions is easy. Finding problems is hard.

Always remember, however, that you're the designer. Playtesters are not capable of doing your job. They're just an opponent, and a source of feedback and ideas, at best. I've done large stretches of productive playtesting with playtesters who provided no useful feedback or input. You are the main playtester.

Bad playtests

Sometimes a playtest will go badly and a playtester has a poor experience.

Keep an even keel, and start asking them for details and suggestions. Agree with them. Say you'll change the rule(s). They'll quickly feel better about things.

You can get good information out of these playtests, and keep your playtester on side.

Some playtesters just don't like the type of game you're working on. If it's not their type of game, just don't playtest with them. Save them for later.

Emotional monitoring

A playtest isn't just for the playtesters.

What should you be doing during the playtest?

Be constantly aware of your emotions during playtests.

This point might sound obvious, like saying "make your game fun", but it's actually a crucial part of my design process. Forgive me if I labour the point.

Normally, we're trying to ignore negative emotions, not monitor them. In day-to-day life, we don't haul a bag of groceries, and analyse ourselves as we do it. We don't think "my hand is sore... my arm is sore... I'm somewhat bored..."

It's a strange thing to do, but if you're experiencing a negative emotion while you're playing your game, other players will too.

You're creating an emotional experience here, and you can test it on a human — yourself.

I can't solo playtest, because it's boring, and the feelings are all different. This means I can't analyse my own feelings.

Earlier in the development of Radlands, I noticed that I had more fun when there were fewer cards on the table. When there were more cards, the game could become a slog, and it was hard to impact the opponent. While the slog was entirely fair, I found it frustrating and boring.

The technical and logical part of game design is easy. Balance, costs, ratios etc., are dead simple. The psychological part is really difficult. Why is something fun?

Why does this game feel weird? Can you turn that question into an answer?

The psychological part is the more important one, and the logic is just there as a functioning structure that serves the psychological objectives.

Fixing problems is easy. Understanding what the problems are is hard.