My notes on the How to find amazing game ideas GMTK's video.

Where do game ideas comes from

Bottom Up

Start to create a game with the mechanics / Systems / rules, then add story, theme later.

Other games

• Take an existing game and move it to another perspective
• Shift a game into a different theme / context
• Switch medium (from board game to video game for example)
• Bring game back from the dead

Don't reskin / clone existing title - explore the unique design space offered by the change in perspective / theme / medium or simply by bring you unique perspective.

Everything anyone ever makes is inspired by what's in their head --- what they've played or read or encountered or thought a lot about.
Creation doesn't happen in a vacuum --- everything is a remix!
You start with the ideas you love and obsess about, and go from there to make something that's uniquely your own.
Terry Cavanagh

Using Genres

• Can act like a template / recipe: use a set of ready made mechanics or convention and put your own spin on it.
• Find a problem in a genre, and fix it.
• Combine multiple genres
• Add something brand new to a genre
• Remove something from a genre
• Take a mechanic of an existing genre, but swap the metaphor for something completely new (example, change the first person shooting mechanics to something like taking a picture, or a power washer). Reduce the genre to its smallest possible DNA, then build outward from there.

Making mechanics

• Real world activities / experiences
• Find inspiration through the way we control games
• Isolate one specific mechanic or system from another game and spine that out to do something completely new

I read a lot, I like learning new things, and at some point I'll just trip over a subject or some material that I find particularly fascinating.
So it's not like I sit down and say, okay, I'm going to come up with a new game idea.
It's more like I'm kind of just exploring, browsing the world, then it's like oh, maybe I can make a game out of this.
Will Wright

Top down

Start with the experience, then derive the gameplay

Experience

• Fantasy: what sort of role do you want the player to inhabit?
• Start with a theme
• Highly personal experience / autobiography

Constructing a game idea

Story, Gameplay, ... are not game ideas: they are starting point.
You need to turn it into something that can actually be played.

One common structure:

Ask yourself, in an individual level or scene:

• What is the player's goal / the win state?
• What is pushing player away from this win state / what are the obstacles?
• What are theses obstacles pushing the player toward, what is the fail state?
• What is the player doing to overcome those obstacles and reach the win state / what are the player's actions?
  

General ideation tips and tricks

1. Break games down into their smallest element, this will help you pick / mash-up / changes / and remix this elements into new forms
2. Really interrogate the games you play, consider their pros and cons with a critic's mentality. Flaws can be opportunities for new games
3. Challenge convention: take what is expected of a genre and look at what will happen if you flip it on its head
4. Play everything
5. Impose a limitation or restriction on yourself
6. Enter game jams with interesting theme

What's something that normally takes a lot of space that would be interesting or unique if you game it less space?
Tim "Walaber" Fitzrandolph on Ludum Dare 54 theme "Limited Space"

7. Open your editor and start exploring
8. Surround yourself with music and art
9. Do literally anything else than designing / brainstorming
10. Keep your ideas small - think of a game idea as a seed to be cultivated and grow during the entire development process

Is this idea any good

Don't make the first idea you think of. Ask yourself few questions

Can you make it?

If you can make the gameplay prototype in one or two days, then you can make the game in one or two years
Jonas

Are you passionate by the idea? You'll probably work on it for years, so you better love it.

Will this game stand out?

Hook

Does the game idea has a hook?
Something new and different.

Some interesting bit of information about the game that compels people to try it, or to discuss it
Ryan Clark

Imagine what the headline of an article about your game be?
If you can't write a catchy headline from your game idea, it probably doesn't have a hook.

Anchor

But don't get too stressed about the hook either

You can't go overboard and wow them with a hook that is so out there, so original, so revolutionary that potential buyers have no idea what your game even is.
If it sounds too risky to them, they are going to pause, say 'looks cool but I don't trust it'. Then move on.
Chris Zukowski

Think about an anchor: something that makes the game feels familiar or safe.

Make a game that is simple, with something unexpected. For example a tower defense game, where towers can be damaged by your own bullets.
Simple gets people in the door, unexpected gets them hooked
Zachary Richman

Appeal

Jonas recommends to focus less on hooks, but more on what he calls appeal.

Appeal is everything that draws you into the game, before you play the game.

Different kind of appeal:

• Fantasy appeal: "I want to be this"
• Exploration appeal
• Toy appeal: "I want to touch this, to interact with this"
• Challenge appeal
• Nostalgic appeal
• ...
  To see if you your game is appealing, ask yourself:

How would you market it?

When I'm even thinking about the idea, I'm thinking "How can I express this in the trailer?". If I can't imagine right now a cool trailer for this, then it's probably not worse pursuing.
Lucas Pope

Thinking about the trailer makes you ask: Can I show the world what makes this game idea interesting?

When you start a project you don't necessarily spend a lot of time thinking about the Steam store assets you're going to end up with, and maybe you should.
Tom Francis

Is the game idea actually fun?

Our brain are terrible video games simulators: all ideas seems fun in our heads, but once we actually make it, we realize it's boring, too complicated, and has disastrous design problems.
The only way to know for sure is to build a prototype.