Making Games

Previously:
- 1.1 overview; parameters and desired player experience.
- 1.2 research, inspiration, comps.
- 1.3 design document and writing your rules.
- 1.4 rapid prototyping.

Next:
- 1.6 iterative design.
- 1.7 release! (fin)

Playtesting!

The reason we want to get to a rapid prototype as quickly as possible (see 1.4) is that No Game Rules Survive First Contact With The Enemy The Player. Notably, gamers are assholes creative.

Also notably, observe; don't interact or answer questions. You want to see what the players do without your intervention. Ideally, they should forget you're there.

And you should run through your own game if possible, although the more eyes the better.

- Ten insightful playtest questions [Game Developer].
- 20 Playtesting Questions That Set Players Up To Give Great Answers [Entro Games].

Let me give you an example not even of a complete rules set, but of a proposed "game balance" change to a game that was a Crysis Wars mod. (Mechwarrior: Living Legends, to be exact, which is sanctioned now?!) MWLL had a notoriously steep learning curve.

At the beginning of the game in one of the most common modes, every player would start with the same small amount of money, so you'd all be stuck in (inevitably) an Owens C or mayyyybe a Raven. Most of us learned not to get into the paper-armor Osiris unless trolling. But basically, because you earned money by getting kills or capturing bases, good players would earn money to buy more powerful mechs, suck players (hi!) would languish forever.

I remember being in the forums when a well-intentioned gentleman proposed that the devs figure out some way to implement a "detect n00b player" and give them extra money to buy a nicer starting mech to even the odds. This sounds great except implementation matters: he specifically said (keeping in mind this was state of the art for coding about ten years ago) it could "detect n00b" by e.g. when the player, being bad at piloting, bangs into the side of the hangar three times in a certain time period.

People IMMEDIATELY pointed out that this would just incentivize everyone (including non-n00b, highly skilled players) to bang into the side of the hangar three times for the extra cash.

This might also suggest to you, correctly, that "game balance" is a can of worms in itself, but we're getting ahead of ourselves.

Some players are great at min-maxing and will BREAK your system given half an opportunity. (My husband.)

Some players find symbols hard to parse and will CRY if your game rules rely on learning a bunch of symbols. (This is me. I literally cried the first time I played Race for the Galaxy. I'm someone who stares at the Dock in panic because I can't figure out which funny symbol is Chrome, which is why I have a keyboard launcher installed.)

Some players will always play a fighter no matter what other character classes are on offer unless forced into a corner. (This is also me.)

Some gameplay preferences don't have to be accommodated. I'm still remembering the friend's wife who radioactively would not handle anything combat-themed, which meant a CCG loosely based on samurai warfare was never going to work for her. Someone who only plays RPGs may not be into your abstract Eurogame.

But at other times, watching players, or reading/listening to their play reports, may reveal areas where your game design, rules set, or prototype could be improved. This also gets into issues of accessibility, which is a large topic; see the resources for some readings.

Practical note: My experience is that unless you're bullying your family into playtesting for you (HI HUSBAND AND DAUGHTER), I expect something like a 50% flake rate for people who volunteer to playtest over the internet. People have lives, the unexpected happens, etc. Don't take it personally. Just ask twice as many people as you think you'll need. :)

Challenge 1.5
Round up at least one playtester (two or three if we're allowing for FLAKE RATE) and ask them to test your game, whether in person or remotely. (You can always ping me unless your game involves graphic, hopefully fictitious catten harm!)

Next up: 1.6 iterative design, 1.7 release.