Clarify failure for me

This came up in another thread, so I wanted split off the discussion.

It was my understanding that tests have three possible outcomes:

  1. You succeed; you achieve your intent.
  2. You fail; you do not achieve your intent, and the GM introduces a twist.
  3. You fail; you achieve your intent, but with a Condition proscribed by the GM.

Folk in the other thread are telling me that #2 is actually:

  1. You fail; the GM introduces a twist, and you achieve your intent after dealing with said twist.

I realize that p. 91 says:

Alternately, if the twist is successfully dealt with, the patrol moves back on track for their mission. They dust off their paws and say, “Now that that’s over with, we can get back to business.” Move the story forward as if the patrol had overcome the initial obstacle that caused the twist.

…but I didn’t think that meant that the player literally gets their original intent from the test that prompted the twist. That wold mean that players always get their intent, it’s just a matter of the cost. And that just doesn’t sound right to me.

Am I wrong?

I don’t have the rulebook (yet) … but from what I’ve read on these forums, here’s my understanding.

Sample scenario: The players are looking for a secret door.
The obstacle and test: searching for the door.

Possible outcome 1: Search check succeeds. The players find the door.

Possible outcome 2: Search check fails. But the GM can’t just say “oh well, you needed to find that door to get into the dungeon. Too bad. See you next week.” So … he puts in a “twist”: An enemy team comes up – maybe they reveal the door, or after overcoming the enemy team, the players find a map of where the door is.

Possible outcome 3: Search check fails… but that’s because they didn’t find the trap. The trap goes off … in other words, a condition is applied to the players.

And really, this set of outcomes is not unique to Mouse Guard. A similar set can be used in D&D, and just about any RPG.

There is, of course, a 4th outcome:

Possible outcome 4: Search check fails … the players also fail the followup twist, and perhaps the twist after that. They fail so badly, that they fail the entire mission. Friendly GMs would say “your patrol is left in the wilderness for dead, and you are rescued back to town by the follow-up patrol that Gwen sends after you.” Cutthroat GMs would say, “you guys died. Hmmm, we still have two more hours of gaming to go … wanna make new characters?”

So, I would have to say that there is a possibility that the players really, really, fail … and the game can come to an abrupt end. But for the most part, they simply have to “pay” the costs of failure … and the game goes on.

I don’t think so. I don’t have my book in front of me, but I believe the only things that can keep the mice from succeeding are losing at a conflict or having unfinished business at the end of the Players’ Turn.

I went to grab my MG book on the self and grabbed the comic instead. That was confusing.

Looking at page 91. . . If a failed rolled leads to a twist, the players must deal with the twist. After the twist is over, the GM either puts up another obstacle or allows the players to complete their mission. (Arrive at the town, etc.)

On the other hand, conditions explicitly say “He gets what he wants but you get to give him a condition.”

I would say you’re correct. I’m glad you asked, before hand I would have answered the other way. That also makes playing from twist to twist make a whole lot more sense!


While I appreciate the answers thus far, I think a basic part of this rule is being overlooked. It is possible that the original goal is never completed. The twists can keep spinning the story further and further away until such a point is reached in which resumption is nonsensical.

You can and should go from twist to twist to twist to twist, etc. Twists are new situations, once presented the players might not even want to go back to their original goal.

That’s on page 91, second column first paragraph.


Hmmm … so it’s like a recursive function …

Function PlayersPerformTest(checkSkill)
… check succeeds … players get their intended goal; return achievedGoal
… check fails …
… GM imposes a condition, and players get their intended goal; return achievedGoal
… GM does a twist; return (call PlayersPerformTest(newTwist))
… players fail; return failedGoal
… players spin out of control; return (call NewStory())

Luke - Thanks! I just had an aha moment and failure in MG clicked. Playing from twist to twist sounds really really fun. I might have to put the BW one shot I’m planing on hold and break out MG again.


Essentially, the mission failure conditions are “Give Up” or “lose a conflict”, and the twists taking away so far that going back is a form of Give Up…

That being said, since almost any obstacle can be made into a conflict, the GM has a lot of power to put failure on the line, and in some seriously surprising ways.

Any A intent vs B intent roll can be either a quick-roll contest or dragged out into a conflict.

Make it from Lockhaven through the snow in winter, with winter’s intent to Injure the PC’s… can be handled as a single roll of Pathfinder vs Winter (8)… or as an extended conflict, with Attack being Pathfinder, defend being survivalist, Feint being Loremouse or Pathfinder, and Maneuver being survivalist or Loremouse. Winter simply rolls nature for each… Party Dispo is base Health, plus roll of Pathfinder. Winter is Nature + Roll of Nature.

If taken as an extended conflict, it’s quite possible to lose, and accept ‘we don’t make it to lockhaven, and instead hole up with a squirrel for a month’ instead of injury…

I would emphasize that while any action that results in a Twist-Fail is definitely failed, the challenge that created it is rendered irrelevant, at least as far as this mission goes:


Mice failed in a roll to find food to calm the nerves of a village? That might be a Twist involving conflict with surly mice. But after that conflict, the story doesn’t go back to finding food for mice, the GM has to present a new challenge.

Thanks, Luke.

But what I’m needing clarity on is the player’s intent for that roll, not their goal or the mission goal. I know the mission continues, and that’s how I read p. 91. The twist diverts the mission, not the intent of that failed roll. The intent is never achieved.

Example from my game: Sloan’s player wants to use the Glazier skill to help repair the windows in Honeywind’s bakery, in the hopes of impressing Thom. Sloan’s player rolls and does not beat the obstacle.

My reading: Sloan fails to repair the windows and (twist) the window frames are now completely ruined by her efforts, so Honeywind’s baker is even worse off than before.

The other reading: Sloan does eventually repair the windows, but (twist) something about the process puts her, I dunno, in debt to Lester, her enemy.

This is what I’m asking about. The “other” reading says that every roll results in an achieved intent; it’s just a matter of the cost to achieve it. My reading says that a Twist means the roll’s intent is not achieved.

What I don’t know is which reading is correct.

You’re overthinking this, Buzz.
What does the book say?


That doesn’t sound like a twist. That sounds like a failed roll.

The other reading: Sloan does eventually repair the windows, but (twist) something about the process
… costs something to Sloan.

It doesn’t have to be a twist. You can apply a condition: “Sloan eventually repairs it, but he gets really tired.” … or if the roll is really bad … “… but he gets really tired AND angry because othermice were laughing at his clumsy repair job”

if you want a twist, it would have to go something like this:

“As Sloan is finishing up repairing the windows, Buster comes up and starts arguing about how the job should have been done.” This sets up either a persuasion check or conflict as a twist. If Sloan wins the debate, then the mice are convinced that whatever “defect” was pointed out is really a wonderful work of artistic craftsmanship.

As far as I can tell, on a biffed roll you either fail with a twist, or you succeed with a condition.

And what’s a twist?

A Twist is a new plot complication.

E.g., not only do you not find the grain peddler, he’s been eaten by a snake that’s prowling the area you are in.

Right. Mechanically, a twist is a new obstacle. An obstacle is overcome using one or more tests.

Right. The twist leads to another check or conflict.

And in the case that the comic book ended up with: Kenzie, Saxon, and Lieam FOUND the grain peddler … it just happens that he’s dead.

I appreciate your input, but let me work this out with Buzz.


Right! But they still failed to achieve the original test’s intent, correct? Successive obstacles may eventually lead them to the grain peddler, but that initial one does not, right?

I mean, they find the peddler’s body because one of the players has the presence of mind to say, “Hey, I’ll bet the peddler was eaten by that big snake Lieam just killed. We should cut it open to find out!” If no one had thought of that, the players would not find the peddler’s body, because the original Scout test was failed.

More e.g.: in the Sloan example I give above, Sloan does not successfully repair the windows. The Twist might eventually lead to the windows getting fixed, but it’s not going to be because of her biffed roll. It’s going to be because of some other obstacle that was overcome.

Where are you getting this “intent” talk? Seriously, I just reread all the relevant parts of MG that I could find. There’s no mention of “intent.”

Players are presented with an obstacle, they describe how they tackle it. Or, in the players’ turn, the players describe what tests they want to spend their checks on. That’s it.

I don’t understand your first example. But the problem in your second example is that the twist does not directly lead to a new obstacle, a new problem that must be overcome. Twists should be immediate, not a vague thing that happens in the future.*

It’s not about fixing the windows! It never was. It’s about how hard life is for the guard and how they stand up to the pressure.


*Yes, there are exceptions to this rule.