How I learned to tell stories on purpose in Frog Fractions 2

In this article, I’ll explain how I went from having basically no idea how to construct a story to making players cry with my story (in a good way this time). There are big spoilers for the Frog Fractions Hat DLC below, so maybe play it first (or keep reading until an explicit warning about spoilers comes up).

When I ran the Kickstarter for Frog Fractions 2, most people probably guessed that I had no idea what I was going to make. I had made Frog Fractions entirely improvisationally and I figured I could just do that again.

The trick is, when you get something right the first time, you haven’t learned anything. You have no idea which elements were due to your innate talent and which were accidental. The most important accident, I discovered much later, was that I built Frog Fractions in chronological order, and I designed each scene to follow naturally from the previous ones.

By contrast, I started work on the sequel before I knew where I’d be hiding it, so there was no previous scene to work from. Instead I started building gameplay vignettes that were individually entertaining. It turned out to be very difficult to fit these together into something that felt cohesive, and I feel like I only partially succeeded.

I had no idea how much of the success of the first game—even to me, a not-particularly-story-focused player—stemmed from it being at heart a buddy comedy, the story of two friends going on an adventure together.

I started work on “Hop’s Iconic Cap” with these intentions:

• Like Frog Fractions and Glittermitten Grove, I wanted to build it improvisationally. It’s more fun that way, and leaving the design loose means you can reshape it on the fly as you learn more about the game you’re building.
• Like Frog Fractions and unlike Glittermitten Grove, I wanted the game to flow easily, like watching a movie, which meant all the minigames should be easy and, if possible, they should be recognizable riffs on existing games that the player already knows.
• Unlike Frog Fractions and Glittermitten Grove, I wanted to figure out how to tell a meaningful story.

With storytelling on the brain, I replayed The Secret of Monkey Island and noticed that it doesn’t have a story so much as it has “there is an antagonist” and “there is a love interest.” The lesson being that your characters don’t really need to be complicated to be compelling. They basically just need to want stuff, and the story can be those wants playing out, and charming dialog can paper over a lot of storytelling weaknesses.

(Warning: spoilers start here!)

So I started by spewing character traits into a Google Doc. Here’s the section for “Frog Kid,” the character that became October:

• 🔥🔥🔥 Talks using lots of emoji. ⚡️⚡️⚡️
• Goes on road trip with you. Wishes they were with mom, who does all the cool stuff.
• Playing Game Boy literally all the time.
  • You revisit old locations from FF1 together, but they’ve seen it all and act bored.
  • Kid turns out to be more engaged than you thought. Texting mom about your cool adventures—mom thinks kid likes you more.
  • Use Game Boy at the end to hack ad agency servers?
• Rides Draggy’s kid because that’s what you do?
  • Draggy’s kid has a Wonderswan??
• Or maybe rides on your head.
• TURNS OUT TO ACTUALLY BE YOUR DAD??

(Draggy doesn’t have a kid. Draggy’s a lifelong bachelor.)

Then, while I was doing the technical work of porting Frog Fractions 1 to Unity, I spent a bunch of time talking to friends who knew about storytelling, about how to take these ideas and turn them into a story. I found Laura Michet and Xalavier Nelson, Jr. to be especially informative resources.

The opening scene of the game came to me as I was thinking about the kind of person who’d be replaying Frog Fractions after eight years. I decided that Hop should embody the player who was looking for adventure but didn’t know where to find it, so they were revisiting old adventures. I made Hop run a nightly reenactment of the events of the first game for a live audience. Since I ended up writing Hop in my own voice, this got muddled thematically with the idea that Hop was instead embodying me, making the same video game yet again. But you know what? That interpretation is fine. You can believe that if you want.

Fun with dialog trees

By the time the FF1 port to Unity was ready to go, I had a story skeleton I was happy with, and I started implementing it, like FF1, in chronological order. In a sense, the problem I was solving got flipped on its head—in the first game, I was trying to figure out what should happen next. In this one, I was trying to figure out how to get from where I was to the next scene.

I ended up leaning heavily on dialog trees. Ron Gilbert has talked about discovering, while making Monkey Island, that playtesters hated lengthy expository scenes but they had no problem with exhausting every option in a dialog tree and absorbing the same amount of information in the same amount of time. Even a tiny nod to interactivity can have a huge psychological impact. Humor helps, too—LucasArts-style dialog trees can be extremely dense with comedy because each option they present to you can be a separate punchline to the same joke. If the player is laughing, that means they’re not tired of reading yet.

As it turns out, I love writing dialog trees. I am really annoyed that I only discovered this after turning 40. I could’ve spent those forty years writing dialog trees!

I worked very hard to make the dialog as punchy as possible, especially after realizing just how much of the game was going to be dialog trees. I made it a priority to get to the next story beat fast, in as few lines as humanly possible, and also be as entertaining as possible along the way. After playtesters showed reading fatigue when presented with too many possible responses, I started being much more stringent about how good a joke had to be to get added to the list of response options.

I also added a little fast-forward icon to mark the fastest way through a conversation, for people who were in a hurry to get away from the wall of text.