... now with 35% more arrogance!

Monday, July 8, 2013

Elegant Rolls II: Failure

To continue on with the discussion about replacing the D&D roll-over mechanic with something more elegant, resembling the Basic Roleplaying roll-under mechanic: the easy, obvious way is to just use a d20 attribute roll, as in some late 1e writings, almost like the Dungeoneer's Survival Guide and Wilderness Survival Guide. Just roll a d20, higher is better, but you can't go over your ability score, and an exact match is a critical success, but a 20 is a fumble (the "blackjack mechanic", as someone referred to it in the comments on Zak's post.) But it's worth thinking about what a failed roll should mean in these circumstances.

  1. As you travel past a barrow, the GM rolls a d20 and compares it to your Int or Wisdom, whichever is higher; on success, the GM tells you that you notice a partially-concealed entrance on the eastern side. If the roll exceeds Int or Wisdom, you don't see it. Does that mean you can never find it? Do you get to roll again, and if so, when? Do you always have to roll?
  2. Later, you've captured a goblin and want to tie it up. You roll less than Int (or less than the infamous Use Rope skill.) Presumably, success means the goblin is successfully tied. But what does a failed roll mean? Does it mean you forget how to tie knots? Or you really screw up the knot badly? Does the goblin roll?

Some games explicitly tell us that skill rolls work like pass/fail checks for basic tasks. Even when not explicitly described this way, it winds up being used that way at some tables. Players wind up rolling for everything, and getting frustrated when they routinely fail at simple tasks. Other people have suggested that rolls should only be used in stressful situations. But I'm going to go one step beyond that: if you know how to do something, you always succeed unless opposed by someone; what you are rolling for is whether or not you finish in a short span of time.

  1. One person searching the east end of a barrow for an entire hour (or six people searching six sections for ten minutes) may find the entrance automatically; what you are rolling for is spotting the entrance immediately. Thus, there is only one roll.
  2. Tying up a goblin is easy; what you are rolling for is to see if the goblin can get loose. If you fail, the best you can do is delay the goblin for one minute; if the goblin rolls and succeeds, not even a full minute. If you succeed, the goblin has to beat your score to get free in one minute

The general guideline should be: figure out the minimum time something could be done (in combat, this is generally one minute.) If the roll fails, and you aren't interrupted, it takes you ten times as long. 
  • Picking a lock? Success means one round, failure means ten rounds. 
  • Building a quickie lean-to? Success means ten minutes, failure means 100 minutes. 
  • Building shelter that's a little more weatherproof or concealed? Success means one hour, failure means ten hours.
The only time a character should actually fail in a grand manner (can't tie the knot, jams the lock, shelter collapses on the character) is when the roll is a fumble. In this case, a roll of 20.

Combat has some other considerations, including what to do about class levels. But I'll deal with that in a separate post.


  1. I really like the idea that skills take something that would normally eat resources (time or whatever) and reduce the loss. It brings the skill system into line with the original idea of general competence by adding something instead of acting as a gatekeeper mechanic.

    I think I was the first one to use "blackjack mechanic" in Zac's comment section there, but I am pretty sure that I heard the Pendragon system described by that term somewhere else.

    1. Yeah, I probably should have explicitly mentioned the concept of "gatekeeper". Gatekeeper features for magical or extraordinary effects are OK, but for ordinary effects, like tying a knot, they kind of suck.

  2. In the case of tying up the goblin, you could use success/failure in place of the goblin rolling to escape. Success means your knots are too skillful for it to untie, while failure means the goblin can and will escape at the first good opportunity. This would cut down on the number of dice being rolled and makes the situation more player-centric (i.e. the player's good roll is not stymied by the DM's even better roll).