Failure Is Fuel, Not a Dead End

Most dice-based games have a failure state that goes like this: you roll, you don’t hit the number, the GM says “nothing happens,” and the table waits for someone else to try something. I’ve sat through a lot of those moments over the years, and they all share the same problem — the roll consumed time and tension and gave the table nothing back. That’s not failure as drama. That’s failure as a stall.

I wanted Fate’s Edge to treat failure as fuel instead. Not “fuel” in a softening, everybody-gets-a-trophy sense — a Miss in this system is still a real setback, and it should feel like one. But it should also hand something back to the table immediately, so the story keeps moving instead of idling.

The loop: a Miss pays two people at once

Here’s the actual mechanic. Roll your dice pool. Each die reads as a success (6–9), a crit (a 10, worth two successes), or a wash. Hit your target number of successes and you do the thing. Fall short and something happens on both sides of the table at once. There are only ever four answers, and all four of them move:

Note what’s not in that table: a row where nothing happens. There isn’t one. That’s the whole design.

You, the player, get Boons — two of them on a Miss, one on a Partial. Boons are yours to spend: reroll a die, bump your Position a step, fuel certain Talents, or bank them toward XP if you don’t use them. A bad roll just handed you a resource you didn’t have thirty seconds ago.

Meanwhile, every 1 you rolled hands the GM a Story Beat. Story Beats are the GM’s half of the same economy — spent to introduce a complication, escalate a threat, or push a timer forward. The two currencies come out of the same dice, on the same roll, and they’re built to be spent, not hoarded. I’ve said elsewhere that the climax of a session should be made of the Story Beats spent getting there, not one saved up and cashed in all at once at the end. A GM sitting on a pile of unspent Story Beats going into the final scene has been playing the economy wrong.

Why the GM isn’t allowed to just take dice away

There’s a rule underneath all of this that matters more than it looks like it should: the GM never reduces a player’s dice pool. Not for a bad plan, not for unfavorable circumstances, not for “realism.” If a situation is genuinely harder, that gets expressed through Position (Dominant, Controlled, or Desperate — which changes what a success or failure costs, not how many dice you’re rolling) or through a higher target number, never by quietly shrinking what’s in the player’s hand.

I added that rule on purpose, because “just take a die away” is the easiest bad habit for a GM to fall into, and it’s corrosive in a way that’s hard to notice from the GM’s side of the screen. It moves the tension away from the fiction and into a number the player didn’t choose and can’t see coming. Position and Story Beats do the same job — making a moment feel dangerous — without ever touching the thing a player actually built: their pool.

What this asks of a GM

The practical effect is that a GM’s job stops being “decide whether the player succeeded” and starts being “decide what happens next, regardless of the outcome.” A Miss isn’t a wall, it’s a fork: the door doesn’t open, and now the torch goes out, or the guard down the hall heard that, or the rope frays a little more than it should have. None of that is punishment for its own sake — it’s just what a Story Beat is for. The system hands you the currency and trusts you to spend it on making the scene more interesting, not on making the table feel bad for rolling low.

That’s a real shift in what “being a fair GM” means. Fair doesn’t mean neutral outcomes. It means the dice and the fiction are doing honest work together — a Miss costs something real, but it never costs the story its momentum.

The same four outcomes, at every size of problem

Here’s the part I didn’t plan and now think is the most important thing about the loop: it doesn’t care how big the problem is.

That same matrix resolves a character picking a lock. It also resolves a character talking a magistrate out of a ruling, a ship crossing a strait, a season of downtime, and — this is the one that surprised me — a war. When I wrote a mass-combat module recently, I went in expecting to need a wargame layer bolted underneath. I didn’t. An army turned out to be a Follower with a Capability rating and two clocks (Supply and Morale); a general’s order turned out to be an ordinary roll whose successes fill an objective clock two segments, one, or none, by exactly the four outcomes above. There is no second engine anywhere in that module. There’s one engine, pointed at something bigger.

I’ve started treating that as a design test rather than a happy accident. If a new subsystem needs its own resolution mechanic, I’ve found a bug, not a feature. Almost every time I’ve reached for a new mechanic, what I actually needed was one of four things I already had:

• a roll (Attribute + Skill against a DV, at a Position)
• a clock of 4, 6, 8 or 10 segments
• the outcome matrix above
• the two-currency economy — Boons to the player, Story Beats to the GM

Crafting is a roll that fills a clock. Travel is a roll that fills a clock. A siege is a roll that fills a clock. A faction’s slow collapse over a campaign is a roll that fills a clock. The scale changes; the grammar doesn’t. And because the grammar doesn’t change, a player who has understood one lockpick has understood the war, which is worth more to a table than any amount of subsystem depth.

Why this is the part I’d defend hardest

If I had to pick the one mechanic in this whole system I think does the most work for the least complexity, it’s this loop. It doesn’t require a GM to be clever on the fly about pacing — the dice hand you the pacing. It doesn’t require a system of fail states and success states that fork into different rulebooks. It’s the same four outcomes every time, and every one of them moves the scene forward instead of pausing it.

I built the rest of the system — Position, timers, the Deck of Consequences, the Legacy Tracker — as extensions of this same idea: that a roll should never be a dead end, just a decision about which direction the story bends. But this is the load-bearing piece. Everything else is decoration on top of “failure pays out, and it pays out to both sides of the table.”

And that’s why it scales. A rule that only says “you failed” has nothing to give a bigger scene, so a bigger scene needs a bigger rule. A rule that pays out on failure has something to give at any size — which is how one four-row table ended up resolving both a stuck lock and the fall of a hold.