In our gaming session last night, I made a call that was controversial, and I'm pretty sure I made the wrong call. I could really use some feedback from the GMing community.

Here's the setup:

I'm running my players through the AoR "Dead in the Water" adventure from the GM screen, and we've been going through it for four sessions now. We're getting to the endgame, and currently the group is split up, trying desperately to save the Nebulon-B frigate from getting sucked into a black hole. The frigate gets to the event horizon at event counter 10, and it was at 7 when we started the game. Our big game hunter/resident Force-user is trying to clear out the aft hangar bay by herself and having a rough time of it. Our thief/assassin and our doctor/melee fighter are trying to restore power to the engines. Our mercenary captain and scoundrel are trying to take the bridge with a squadron of soldiers. Events are getting heated.

The doctor rolled a Triumph and some on his melee attack, and I figured that the Triumph would let him know the secret about the static vane and how to ionize the droids. They knock out the droids, and now they have to get the engines started. According to the adventure, starting the engines requires a Daunting Computers check and a Hard Mechanics check. Against all odds, the thief and the doctor make the Computers check, and then it's time for the Mechanics check. They fail, but with two Advantage.

The doctor says, "So, for the two advantage, could we say that the repair was made, but there is some other part that has to be fixed as well?" It strikes me as a reasonable request, so I agree to it. In retrospect, it might not have been the greatest decision ever, but I think it works well with what is going on. I say that the component that needs to be fixed is located in a maintenance shaft. The thief has the higher coordination, so she goes in. At this point I'm thinking it's going to be an Average Mechanics check. I've spent a Destiny point on the Daunting computers check, so I'm thinking this is going to be a good place to stop for the night.

The thief fails the check. No Advantage, no Threat--just a failure.

The controversy:

The thief asks if she can try again. Here's where things start to get complicated for me. If there was some other symbols out there, I would probably have had a bit more of a guideline to follow. Still, there's just nothing, so I figure it's reasonable for the thief to try again. Nothing indicates the repair was impossible to make. She rolls again and gets a failure with two Threat. I say that she fails, and what's more she can't try again.

Then the doctor says, "Okay, I'm going to go in there and try to fix it." At this point I'm wondering if I should just say the engine can't be repaired, but I also know that if the engine isn't repaired, there is going to be a lot of NPC deaths, and the game is going to get very dark very quickly. I don't know that I can refuse, but I'm also wondering how many chances the PCs deserve. I say that he can do it for a Destiny point. I throw a couple more Destiny points at the doctor and increase the difficulty of the Mechanics roll, but he pulls it off, and with one Advantage, too.

I end the session saying that the ship's engines have been repaired, but the bridge hasn't been taken back yet, and that the heroes have all of five minutes before they fall into the event horizon, to account for the multiple attempts to start the engine.

I'm wondering if I made the right call in having the group take so many shots at restarting the engine. I was thinking what I should have done was said that particular engine couldn't be repaired and have the PCs try to get to a different engine's computer systems and attempt the same checks again. The way I did it I feel as though there were no penalties for failure, and that's not the feeling I want to give my adventures. Having said that, I don't want a major point of the adventure to get torpedoed by a bad dice roll.

I would love to hear what other GMs thought about my decision, and whether they would have done things differently.

If a skill roll is critical to the plot and MUST be completed to continue, you may wish to consider using a "Yes, but..." result that requires additional time, resources, or some other mishap to occur.

I wouldn't worry too much about any single game decision. You can always discuss it with the players afterwards and change it for next time.

I think you were more than reasonable.

Were it me, I wouldn't have allowed the thief to make the second attempt. With a flat null result, I would have said that once they got up there, the wires were beyond their ability and they realized that if they tried anything, they'd likely have made things worse.

No net change in the situation, but the character attempted it, failed, but didn't cause any lasting damage -- the only thing lost was time.

The doctor going up there and succeeding in the check would have been the next logical step. They have a different set of eyes and a broader understanding (higher Int, probably) and could make the attempt.

You did a good job, given the situation. The only thing I would have changed is to disallow the second roll by the thief.

As to the repercussions of the events, if they failed to get the engines online, that's pretty much it for the ship. Time to head for the escape pods. You might have given them another out with a more dangerous job near the actual engine reactor itself, but there would have to be some consequences -- maybe an easier check, but automatic crit +20 each round due to radiation exposure or somesuch. In that case, I'd allow re-rolls of the check with increasing amounts of setbacks due to pressure and fatigue.

As Fang said, I wouldn't sweat too much, one or three decisions isn't going to kill your game - particularly these kind. However I'm happy to give suggestions of what I would have done, because your right that a few of the pieces feel a bit off in reflection.

1) Before even getting to this point, you need to decide as a GM if this engine is plot-critical. If the story is going to get to dark if they don't succeed, then you should be rolling for "how long it takes" or "how much extra time it buys you" rather than "does it work". If they're under a timeline and it's not plot critical, don't let the players get too stuck on it and spend forever to fail.

2) The doctor's suggestion is more along what I think of as succes with threat than failure with advantage. If you fail, you shouldn't accomplish the goal of the roll (fixing the engine). You should, however, be in position for something else. Say they got the engines running for a second before they shut down again, buying them an extra turn. Or "you don't know how to fix it, but you realize that quickly, so your turn hasn't been spent yet."

3) If your going to go into secondary repair levels, by that point I feel you need to be in the "Yes, But.." territory. The thief is going to fix the engine, it's just how long it takes him, or how much environmental damage he takes doing it. (say, losing 2 wounds + 1 per failure due to radiation/sparks/etc) I feel like after spending this much count-down time on it without threat/despair, taking the engines away entirely is a major let down.

I personally hate re-checks. If it's a pass/fail, its dead when you fail. If it's something you can keep trying until you succeed, your rolling for time/level of success, unless you despair, than you broke it.

I really dislike when players keep throwing attempts at a roll. It screams meta-gaming to me and if you're going to let them roll until they succeed then basically just say they succeed from the beginning. As discussed in many other threads, anything that is critical to the plot shouldn't hinge on a die roll. If that's the way the adventure was set up, then poo to FFG for doing that.

Some things you could've done instead:

• Failure means failure. Time to get off the doomed ship. A bit harsh and dark as you mentioned.
• Failure means the PCs fail, but some other agency fires the engines (saving the ship but frying the engines permanently perhaps).
• Failure means no safe way to bring the engines online. Give the thief a 5-second (real-time) choice. "The circuits are fused, but you can bypass them by bridging the connection, unfortunately someone's going to have to hold the bridge in place." That gives the doc something to do in trying to save the multi-crit result you can toss onto the thief. Heroic sacrifice and all that.
• Failure means the thief can't do it. "You have no idea how to fix what's wrong. The circuits are so complex they seem more like nerves and blood vessels than what you're used to working with." (an obvious hint to let the doc try).
• Failure means nothing to do. The engines are online, the monitoring circuits that would show that are the ones that are fried.

To be fair, I might've fallen into the same trap. It's a staple of D&D style adventures, and one I'm working very hard to eliminate from my GMing style. The plot always moves forward, rolls are to show how the PCs interact with it. If the ship must be saved then it will be, but the PCs might not be the prime saviors.

Ah, the good ol' "roll until they suceed" trap! Many a GM has fallen into that!

I might have gone "Yes, you can try again. However keep in mind that the clock is running and you will be cutting your window of escape VERY close. And if you blow that one, you wont have time for for a third attempt"

If you didn't want to go that way, you could always have gone with "How long it takes" method. Success means that you fix the engine in the allotted time. Failure means that it takes longer. Failure and threat means that it takes significantly longer and failure with despair means that the countdown clock is a 00:02 when you turn things around and now there's all kinds of additional complications you have to deal with.

But otherwise, I probably wouldn't have allowed retries until success. "Sorry, you are not familiar with this type of engine. If you had more time and a dry-dock facility, probably - but now, running full blast into a black hole? Sorry."

If a skill roll is critical to the plot and MUST be completed to continue, you may wish to consider using a "Yes, but..." result that requires additional time, resources, or some other mishap to occur.

I wouldn't worry too much about any single game decision. You can always discuss it with the players afterwards and change it for next time.

Actually, I'm with the OP. Sometimes you have to go with what happens and think of the greater good. They actually failed four times because starting the engines and the component in the maintenance shaft are really the same test - can you get the ship running. If you fail at something four times and the GM continues to let you keep trying, it weakens the game itself. It's a bit like the remark in Doctor Who - you can cause the odd paradox and the Universe will by and large get by. But you can't keep punching holes in it or things start to wear thin.

I don't know the module but if it set a Daunting and a Hard roll as requirements to achieve something, then it must have planned for failure - that's a lot of difficulty there. I think as the OP believes, it was the wrong call. But having been there myself and also made the wrong call on occasion - I know why they did it. It's a lot easier to say it when it's abstract.

My rule with interpreting the results is that Failure means failure. You can fail at something and get wonderful compensations or side benefits if you get a lot of advantages, but it still means what you attempted didn't work.

To the OP, I would say learn and move on. Maybe, if it is necessary, clarify to the group that you wont normally allow that level of second, third and fourth chances - just so they don't expect it next time.

While you form a dice pool let the players know what the pool is about, that way they will take your guidance and run with the results and help your narrative. If the roll to repair the engines failed then they will beleive that repairing the engine is beyond them, asking for a reroll is pretty "metagamey" because they can see the dice roll and are playing into that a little. If the roll was to time how long it takes to fix the engine then each roll specifies an amount of time and there is a count down in effect. Even here you have to follow through as the GM, if you have 15 minutes until the ship falls into the sun, and each roll consumes 5 minutes then what happens if the players fail three rolls? This is not the climatic ending you where looking for I bet!

The failure to fix the engines is fantastic, you get to build up tension and twist it on the players. If the doctor makes the theif try again, the thief fails no matter what, if that first roll told him he wasn't going to repair the engine then no further rolls will change that. You or the players have to come up with a narrative that gives them a different solution.

"That reroll consumed more time you do not have you can hear the ship dying now, the bulk heads creak and groan. You can hear alarms and the hiss of air venting to space. Flip a Destiny Point, then, quick make a Piloting Space test using your intellect, a "knowledge spaceship" test as it were. Failure here will end up having the players lose a minute or two as they get lost looking for the shuttle bay. Say the pilot fluffs that roll. Quick, can't you remember where the shuttle bay is? Yikes the artificial gravity goes off and you all start floating. Pushing off the bulkheads you follow the pilot. He suceeds this time, you enter the docking bay with seconds remaining, the pilot opens the docking bay doors as the rest of you enter the shuttle. You leave the ship but escaping is still going to be close, as your pilot works the controlls the gravity pulls you all towards the sun/black hole. Better flip another destiny point here guys, this is a hard piloting check. What if this fails? You don't want to kill the players so look for something that changes the threat. You escape the sun/black hole, but your engine dies, you are now drifting and have another problem to deal with."

If you need time to think up a new threat take a break, end the session 15 minutes early, or ask for help from the players and see what ideas they have.

Perhaps you could be very melladramatic and have the players suffer 3 strain each turn until they all pass out and then suffer wounds for a turn. End the session there are leave the players hanging on the beleif they are all dead. At the start of the next session they wake up in a hospiltal, cell or enpty ship and a completely new problem from which they must escape.

I am in agreement with the general sentiment here. The only instances where I would usually allow a "retry" is when I've got an established time limit (in terms of rounds) and each check the PCs make gets them closer to the end of that time limit. Plus that, Advantage & Threat grant Boosts & Setbacks to retries

More importantly, the narrative must evolve, so that if the players want to try another "Computers check," the player characters have to be actually doing something different.

"Okay, I broke the datalink ( Threat ), so I'm gonna have to jury-rig a solution here with these spare bits of wiring ( extra maneuver, add a Setback for improvised tools ) to re-establish a connection..."

-or-

"The info I'm getting from the computer ( Advantage ) is telling me I'm at the wrong terminal. I have to go and access this other admin-level terminal before I can do what I need to do ( maneuver or two to get there, Boost die for the extra info/using the right tools for the right job )."

The way I generally do a check like that depends on whether they're in structured time or not.

In structured time, they can make the Mechanics check as many times as they need to in order to succeed, because there are probably people shooting at them or enemies trying to stop them in some way. So each time they fail incurs a little more danger.

In non-structured time, I'd probably do a "Yes, but" type of thing, as suggested. Maybe a failure means they get the thing working but it's still dangerous in some way or there are more enemies or it works just long enough to get them away from danger or there's some other non-ideal side effect.

If a skill roll is critical to the plot and MUST be completed to continue, you may wish to consider using a "Yes, but..." result that requires additional time, resources, or some other mishap to occur.

I wouldn't worry too much about any single game decision. You can always discuss it with the players afterwards and change it for next time.

I think this is the one thing that I missed. There wasn't a "Yes, but..." as a result of the thief's failure. Again, a large part of it was just that it was a failure--no Advantage, no Threat. I'm not trying to blame the roll for how I called it, but there wasn't any cue I could take as to whether what had happened was good, bad or otherwise. My thought was looking at it like an auto mechanic might try to get a car to start up again. Failure with advantage would mean that the engine turns over once or twice then stalls, letting the mechanic know he's on the right track. Failure with threat means that the car doesn't start and something smells like it's burning. Just failure means the engine won't turn over and that's it. Where do you go from there?

Having said that, this isn't being an auto mechanic--this is Star Wars. Again, I think the big thing would have been to acknowledge that the failure took time, and go from there.

I think you were more than reasonable.

Were it me, I wouldn't have allowed the thief to make the second attempt. With a flat null result, I would have said that once they got up there, the wires were beyond their ability and they realized that if they tried anything, they'd likely have made things worse.

No net change in the situation, but the character attempted it, failed, but didn't cause any lasting damage -- the only thing lost was time.

The doctor going up there and succeeding in the check would have been the next logical step. They have a different set of eyes and a broader understanding (higher Int, probably) and could make the attempt.

You did a good job, given the situation. The only thing I would have changed is to disallow the second roll by the thief.

As to the repercussions of the events, if they failed to get the engines online, that's pretty much it for the ship. Time to head for the escape pods. You might have given them another out with a more dangerous job near the actual engine reactor itself, but there would have to be some consequences -- maybe an easier check, but automatic crit +20 each round due to radiation exposure or somesuch. In that case, I'd allow re-rolls of the check with increasing amounts of setbacks due to pressure and fatigue.

I really dislike when players keep throwing attempts at a roll. It screams meta-gaming to me and if you're going to let them roll until they succeed then basically just say they succeed from the beginning. As discussed in many other threads, anything that is critical to the plot shouldn't hinge on a die roll. If that's the way the adventure was set up, then poo to FFG for doing that.

Some things you could've done instead:

• Failure means failure. Time to get off the doomed ship. A bit harsh and dark as you mentioned.
• Failure means the PCs fail, but some other agency fires the engines (saving the ship but frying the engines permanently perhaps).
• Failure means no safe way to bring the engines online. Give the thief a 5-second (real-time) choice. "The circuits are fused, but you can bypass them by bridging the connection, unfortunately someone's going to have to hold the bridge in place." That gives the doc something to do in trying to save the multi-crit result you can toss onto the thief. Heroic sacrifice and all that.
• Failure means the thief can't do it. "You have no idea how to fix what's wrong. The circuits are so complex they seem more like nerves and blood vessels than what you're used to working with." (an obvious hint to let the doc try).
• Failure means nothing to do. The engines are online, the monitoring circuits that would show that are the ones that are fried.

To be fair, I might've fallen into the same trap. It's a staple of D&D style adventures, and one I'm working very hard to eliminate from my GMing style. The plot always moves forward, rolls are to show how the PCs interact with it. If the ship must be saved then it will be, but the PCs might not be the prime saviors.

I should point out that "Dead in the Water" does allow for the PCs failing--failing at several goals, actually. The crew can suffocate to death, the frigate can fall into a black hole, the Rebellion can suffer a serious security breach, and several other problematic things.

I think part of the problem (and something I'll need to rectify in my GM style) is how I envisioned the story playing out. I kind of wanted the story to play out a bit like "Die Hard" in space, with the PCs steadily facing more and more pressure, but eventually getting to the point where they manage to pull through in the proverbial nick of time. That's definitely a heroic narrative, and very Star Wars as well. For instance, it's like how we all talk about Luke destroying the Death Star just as it's about to fire on Yavin 4. The assumption is that it was a very difficult roll to pull off, and Luke with the help of the Force (what help depending on the system) makes it. I don't know about anyone else, but this session has brought me face to face with the idea we expect Luke to make the shot! There's not exactly any discussion of what would have happened if he had, say, rolled a Despair or a failure.

Getting back to the PCs, once they fail and a lot of Rebel lives are lost, what has the adventure morphed into? I don't want to get too meta, but at the time my thought was, "Well, the PCs have been doing quite well so far. Do I really want the entire adventure to turn on one die roll? Again, it's like Luke Skywalker firing the proton torpedoes. For this adventure, at least, I think I should have been more okay with the PCs not succeeding.

If a skill roll is critical to the plot and MUST be completed to continue, you may wish to consider using a "Yes, but..." result that requires additional time, resources, or some other mishap to occur.

I wouldn't worry too much about any single game decision. You can always discuss it with the players afterwards and change it for next time.

Actually, I'm with the OP. Sometimes you have to go with what happens and think of the greater good. They actually failed four times because starting the engines and the component in the maintenance shaft are really the same test - can you get the ship running. If you fail at something four times and the GM continues to let you keep trying, it weakens the game itself. It's a bit like the remark in Doctor Who - you can cause the odd paradox and the Universe will by and large get by. But you can't keep punching holes in it or things start to wear thin.

I don't know the module but if it set a Daunting and a Hard roll as requirements to achieve something, then it must have planned for failure - that's a lot of difficulty there. I think as the OP believes, it was the wrong call. But having been there myself and also made the wrong call on occasion - I know why they did it. It's a lot easier to say it when it's abstract.

My rule with interpreting the results is that Failure means failure. You can fail at something and get wonderful compensations or side benefits if you get a lot of advantages, but it still means what you attempted didn't work.

To the OP, I would say learn and move on. Maybe, if it is necessary, clarify to the group that you wont normally allow that level of second, third and fourth chances - just so they don't expect it next time.

I think I might just do that...although it was a one-time thing. Also, you are right--I did end up letting them do it four times in a row. Maybe one of the real problems might have been going back to the failure with two advantages. The mechanics check failed, but...although honestly I'm kind of drawing a blank after the "but..." on that.

In general, I think I'm pretty good about rerolls. In general, I'd say that with rerolls, if there's a Threat generated, I'm more inclined to put something in the PCs way, or prevent them from making the reroll altogether. Advantages can be used to make things easier for the next time.

Keeping the narrative going, though, might just the the most important thing I've gotten from this--every check should advance the narrative in some way.

I forget what happens in this particular adventure.

But I guess if someone did fail badly can make it so it can repaired but only with a new part for the system.

Like if they were trying to put a computer together and put the CPU in wrong and bent the pins.

If they are going to try new attempts they would perhaps need very creative ideas, like they replace the ships hyperdrive with the one from the PCs personal ship.

I doubt they could think of something creative enough for that particular mission though