This past weekend, I took my prototype game, Legends of the Wandering Crows, to ConQuest Sac’s Protospiel to get some playtest feedback. I was lucky that I got a lot of feedback, and it was… constructive. The playtesters had a lot of positive things to say about the game, but they had more constructive things to say about the game. So I have had a lot to think about for the past few days.

It is really tempting to focus on the negatives, to only hear the criticism. I know intellectually that most game developers spend years developing their games; but knowing that doesn’t make it easy when it’s your game. It’s also really tempting to just ignore their feedback, to say, “Oh, what do those bums know?” But I don’t think that’s true, either; their feedback actually illuminated some issues that I’d been feeling, but didn’t have the words for. This was honest playtesting feedback, and if I’m serious about game development, I can’t ignore it because it’s inconvenient.

So I’m taking a page from my GM playbook. I’m going to try to “fail forward.” Failing forward is a term that I’ve heard a lot of DMs use (Justin Alexander and Mike Shea, among others) to describe a move that GMs can use to keep a story moving forward, even when players fail a die roll. Essentially, failing forward describes a success “with consequences.” Some common examples of “failing forward” include:

• The player gets what they want, but the GM starts a clock that counts down until something else bad happens. E.g., if the players are using stealth, the characters might still be hidden, but their enemies have found their trail and will find them soon.
• The player gets what they want, but suffers some other cost or penalty. E.g., if the player is trying to pick a lock, they succeed, but take damage or break their lockpicks. If the player is trying to decipher a magical text, they can read it but take psychic damage from its terrible revelations.
• The player doesn’t get what they want, but gets a clue towards another option. E.g., if the player is asking the bartender about the thieves’ guild, the bartender says that they don’t know anything, but a member of the thieves’ guild overhears and attacks the party later, giving the party a new avenue of investigation. Or the bad guy gets away, but leaves behind a medallion that helps the party track them down.

These techniques are probably familiar to you from books and film. In Raiders of the Lost Ark, Indiana Jones fails to steal the plane carrying the Ark, but also destroys the plane. This forces the Nazis to move the treasure by truck, and gives Indy a second chance to steal it. Indy failed forward. In The Empire Strikes Back, Luke Skywalker and the rest of the heroes escape the Empire, but at terrible cost: Luke loses a hand, and Han is captured and frozen in carbonite. This sets up the first arc of Return of the Jedi, when the heroes bust Han out of Jabba’s palace. Authors and directors use this technique all of the time to create tension without actually stopping the story’s forward momentum.

As a GM, I believe that the best time to fail forward in the beginning and middle of a scenario. Failing forward is, essentially, a mixed success; and early in a story, especially in a mystery, that is a really useful move to have available. The closer we get to the climax of a story, however, the higher I want the stakes to be, so I’m less likely to “fudge” the outcome of a single die roll in this way. However, I might still use this technique at the end of a scenario to support a larger story, if we’re still in an early story arc.

If I choose to look at my playtesters’ feedback through this lens, what do I see?

• The playtesters actually responded quite positively to my overall strategy. In their opinion, it’s only my execution that needs work. Okay, that’s promising.
• The playtesters validated some of my key game mechanics. For example, my card game uses cards as both inventory and hit points. They liked that mechanic for representing combat damage and healing. That’s not all of my mechanics, but it’s one of my core ideas.
• Perhaps more importantly, the playtesters helped me see really clearly game elements that simply don’t work the way that I wanted them to. Great! I’m glad that I get to learn that early on, in a small, low-stakes environment like ConQuest Sac, instead of someplace else.
• Besides my game feedback, I got to meet and game with some really great people. All of the Patricks, all of the Johns, James, Velgus, and even Floyd! And they’re all going to be at Kublacon next month, so I’ll get to play with them again. Yay to new friends!
• And most importantly, even accounting for the feedback, I had fun! In addition to my prototype, I got to partner with John Brieger for the game design “contest,” which meant that I learned a lot. Then we playtested one of his prototypes, which was also a fun learning opportunity.

So here’s to failure. I didn’t walk away from ConQuest Sac rich and famous (as if that was ever in the cards), but I learned a lot, and my game will be much better as a result of this experience.

What about you? How do you use “failing forward” either in your GMing or game design process?