What makes failure interesting in a roleplaying game? That’s a question we’ve been discussing here at Room 209 Gaming as we work on our crafting and social systems.
Imagine a group of adventurers who have spent many days – both in game and in real life – working together to acquire a rare magical metal to craft a weapon of great power. One of the Players has the vocation of Weaponer, which allows him to use this metal to make any type of weapon he wants.
This Player – let’s call him Bob – finally has a chance to use this wonderful material to make a weapon the entire group has been working toward for weeks. All dice rolls in Infinite Earths are made openly against an announced Difficulty. That means there’s no fudging of the results by either the Guide or the Players.
And Bob rolls a natural 1 on his d20 for his Craft check.
What should happen next? That was the scenario we considered as we designed the crafting system.
There are four basic types of failure in most roleplaying games: lack of success, random results, punishment and rewarded failure.
- Lack of success means that a Player attempted to achieve a task, failed to do so and did not suffer any consequences other than his loss of time and effort.
- Random results refer to systems that have Players roll on a chart to determine what happens when they do not succeed. This can lead to success of a different kind (you succeed while also suffering some sort of setback) or extremely unfortunate side effects (you stab yourself instead of your opponent).
- Punishment refers to botches or other systems that guarantee that not only do you fail to accomplish the task, but you also suffer a painful setback that makes your goal even harder to reach. Critical failures that can maim or even kill characters fall into this category.
- Rewarded failure means you failed the task, but you gained something from the failure. This can be XP, information, a new storytelling opportunity or some other structure that takes the sting out of not succeeding.
The more random results are used, the more likely it is that one die roll can dramatically change the course of the story. That’s why combat rules that include a chance of instant death with every attack roll are not a good fit for a game that intends to keep the same characters alive from beginning to end.
Conversely, a lack of penalties for failure can encourage Players to keep trying an absurd amount of times until they finally do succeed. And rewarding failure can risk a slightly different effect, where it might in some situations actually be more beneficial to fail than it is to succeed.
Punishment failures on skill checks tend to create odd situations where the Players trust NPCs more than themselves (if the goods sold in stores are tested first), or the Players inhabit a world where every single crafted object has a percentage chance of being afflicted with a curse or some hidden defect that will manifest at the worst possible time.
That’s why our Enchanter vocation – the role that creates magical trinkets and enchants items – cannot accidentally create cursed items. Enchanters must explicitly choose to curse an item instead of empowering it (and they can take a Talent that allows them to transform enchantments into curses). That means when a merchant sells them a cursed sword, someone cursed that sword deliberately, and the merchant might be in on it or might be an unwitting scapegoat. A cursed weapon is the signpoint to an interesting story, not a speed bump that is bound to happen a certain percentage of the time.
Consider Bob the Weaponer. If he and the other Players knew that all their hard work could be undone by one random die roll, it’s much less likely they would even put in the effort. Their attempt would have been focused on finding an already forged weapon instead, because that would be a better investment of their time.
That brings us back to the basic question: Should a single die roll be able to completely undo hours of work by the Players? Our answer is no.
In combat, an attack roll of a 1 always fails regardless of the target’s Defense score. That failure doesn’t negate hours of work; instead it forces the Player to consider how to deal with an enemy who is still standing. That’s how failure makes things interesting in a roleplaying game – by providing obstacles that the Players can deal with as they choose. The only additional risk is the potential for additional incoming damage because the enemy is still there; there is no “fumble penalty” of some extra, unwanted effect being layered on.
Likewise, if a weapon crafting skill check of natural 1 means the metal is ruined, all that time was wasted. A random result might mean the rare metal weapon loses the very properties that made it worthwhile. Both options mean that there is a very real possibility that the Players will be forced into a dead end from a narrative perspective.
If a natural 1 means that Bob must try the crafting attempt again, that means the Players must decide if they can afford to spend that time on crafting again. That’s a choice, not a roadblock. The Players might decide to hire additional helpers or rent a better workshop to make the second roll easier. That means more people will be involved, which could have story implications.
This continues in other parts of the Infinite Earths ruleset. Failure on an attack roll should force Players to adapt their tactics, not buy a casket. If an enemy cannot be defeated head-on, then clever players will lure their opponents into an ambush. Cavalry riders can choose to hold their attacks until they have room to maneuver their horses. Story goals are explicitly ability-independent, so a failed Knowledge check or Speech roll doesn’t block players from moving on with their adventure; they just need to find another way to approach their goal.
The traditional solution for this is “hide the dice the GM rolls, so the GM can fudge the results” or “don’t tell the players the DCs so you can fudge their results.” There are some systems wherein fudging the dice is not only a time-honored tradition, but is encapsulated directly into the ruleset as the way challenges are supposed to be handled. This strikes us as flawed, because in such a system ultimately the die roll doesn’t matter; the results are the results through GM Fiat.
We want die rolls to have meaning and consequences, but we don’t want one bad die roll to undo hours of planning and roleplay. Failure is interesting when it is a setback or a detour, not a roadblock. How players deal with and overcome failure is what tales of adventure are all about.