Repeated Attempts by Different Characters

DM David has an interesting new entry on judgment calls the Referee makes when players want to try the same action over again after the dice already said the outcome was failure.

You should read that smart discussion to make full sense of this discussion, but the issue is one that every experienced Referee will recognize: under what circumstances do you let players re-roll to succeed after their character fails through a dice roll? How do you keep rerolls from depleting excitement and raising a degree of boredom?

I'm with DM David on this. First, don't make players roll for something when failure would cause the game to stop in its tracks. If the object of the quest is behind a portal that they can't open because of a failed die roll, that's not fun.

Second, the dice rolls mean little if your solution to their failure is merely to penalize them in time spent. (DM David cites a D&D third edition rule, which I had not heard of, that failure can mean it takes 20 times as long.) I can see how that would be boring, although perhaps it would be less of a downer if the Referee had already said, "You are sure you can pick this lock, but the roll will determine how long it takes." That matters when every minute is critical in the scenario, but what's the point of a roll otherwise if the outcome is guaranteed success?

Third, and most importantly (and again with DM David), I usually interpret the individual player's roll as determining whether any character can succeed. If the player's roll fails, that determines something about the game world: whatever it was is just not going to happen by ordinary efforts.

It's not going to happen, that is, without a change in approach. If the players come up with a new way to deal with the challenge, I may let one of them try again.

But does characters' taking turns count as a change of approach? 

There's one tricky kind of repeated attempt that I've seen players requesting as a change in approach: taking turns attempting the same task in view of varying character capabilities.

This example will explain.

The best lock-picker in the party fails to pick a lock. The players now want the second-best lock-picker to try, because the lower ability score suggests there's still a small chance. If the Referee says "no, you already had your best shot, it's not opening" then next time they will have the less-skilled lock-picker try first, to justify having the better lock-picker attempt a follow-up "change of approach." Shouldn't there be a chance that the expert succeeds when the novice fails? In this situation the players are are playing the game, not telling the tale, in wanting to increase their chances with multiple dice rolls. But no party in a dangerous situation would realistically put the amateur first (unless they were deliberately training the amateur). This is an example of characters relying on game mechanics (a.k.a. metagaming), which, for many, hinders immersion, and deprives the story of sense.

Would you count this tactic as a "change in approach," granting another attempt? I wouldn't.

One response to this that I have used is to tell the players that the lock-picker who fails encounters a jammed lock, or perhaps jammed the lock in the initial attempt, making it worse, like my own amateur home-repair attempts that create new problems. "You should have let me do it!" hisses the most expert lock-picker's player in response. Next time, they send in the expert first and the amateur holds the torch; if the pro fails, I say, "It's clear that there's no way you can resolve this aging lock without different tools, maybe a lot more light, maybe a lot of time--unless there's something stuck inside, or broken, which could be the problem here... Timekeeper, tick off five minutes on the torch. Folks, what do you do now?" If they insist on trying again, I don't let them roll. More time ticks away and I tell them it's not opening. No reroll.

Instead of letting them use variation in their character's abilities as an excuse for a change in approach, they should try a real change of approach: try a crowbar (making a lot of noise), contemplate a spell, figure out another way, or move on.

I do make one exception. In my home rules, characters have a Luck stat that diminishes whenever they Test their Luck. Sometimes I say, "Well, you'd have to be really lucky to pick that lock after the expert failed! But if you want, Test your Luck, and give it a shot!" The player, who is (probably unconsciously) metagaming in this request, is confronted now by a metagame stat that is a limited resource. If the player wants to spend that resource, which is also the basis for life-preserving saving throws, I don't mind giving one more chance. Confronting a player's metagaming with a metagame system seems to work. If they miss the Luck test, or, if lucky, nevertheless miss the lock-picking attempt based on that lower ability, that was their Luck point they spent. Too bad! And if they pick the lock, well that was sheer luck!

Of course, for problems other than locks, other versions of the "decisive failure" solution are possible. The point is that sometimes the answer is just "no." That's what the dice roll was for.

Then there's the "I try too!" phenomenon. I've seen this happen many times since I began GMing in the old days, and still today: A player fails at something like a test to see whether a character has some background knowledge (lore) or notices something (perception). The roll indicates failure. Immediately every other player jumps in: "Well, can I try?" "What about me?" "Me too, maybe I know!" Sometimes they even just start rolling hopefully and shouting results. If I say no, you can't, players feel shut down. "You mean there's not even a chance I'd know that?" If I say yes, we have a bit of a circus of dice rolls after the initial dice already seemed to say "no!" Depending on what was at stake, I have sometimes ruled, "If the educated wizard doesn't know this, none of you knows it. I don't care if you got a critical success." This applies story justice to the characters' varying specialties, keeping the dumb fighter from upstaging the scholar in book learning.

Or, with an initial failed perception test, I might say, "None of the rest of you was in a position to notice whatever it was" (and I try to be sneakier in making perception tests covertly next time).

Other times, I just let them all roll. It depends on what it was for.

Come to think of it, this happened a lot when I ran Call of Cthulhu, where plenty of rolls are used to reveal clues either in the environment or in simulated background knowledge.

Sometimes we Referees paint ourselves into a corner by requiring rolls to determine certain things that don't seem quite so random on reflection. For a quantitative matter like the force needed for snapping window bars out of brick masonry, or tipping a boulder out of the way, I try to avoid making players roll, unless it's to randomize the previously undetermined durability of the bars or the position of the boulder. Instead I decide a Strength total (or equivalent) needed to accomplish that task. I can't allow a weaker character to accomplish a feat of might that a very strong one already failed at, not without a very different approach to the problem.

The same applies to background knowledge. These days I don't ask for rolls to determine background knowledge of a specialized variety. In my home rules, for example, either you have the needed rank of the Ancient Languages trait to read that particular ancient inscription, or you don't. You either know the vocabulary and the grammar or you don't. (I would let a player with an insufficient rank in Ancient Languages roll to decipher it if the character brought reference works along and took the time to work it out.)

Collaborative efforts pose another challenge, but can be used to ameliorate the chaos of "I try too!" I'm sure I'm not the only one to have had situations where I required a dice roll to see if a player could lift a small dungeon portcullis (or similar task). Upon failure, another player chimes in and says, "Oh, my character gets down there and joins in the effort!" My mistake was allowing this to be decided by a single player's dice roll in the first place. Do you give players a re-roll when they help each other out after initial failure? In a way, that's not much different from the lock-picking situation I described above.

Most RPG rule sets don't have effective rules for collaborative efforts. They focus on the heroics of individuals in the group. It is the individual player who rolls the dice. Setting a fixed strength total requirement is one solution to this kind of problem. The drawback is that the excitement of rolling dice is missing when the Referee decides by fiat a fixed amount of required effort. The dice create excitement as uncertainty in the setting is resolved at random. Player dice rolls somehow also simulate effort in the player's experience. Referee fiat takes that away. Yet maybe the setting seems more real if quantities are fixed in advance.

In the end, navigating player requests for re-rolls is a matter of judgment calls. Increasingly, I firmly say no when the dice say no the first time, but there are two main variables for which no single rule will account: the specific nature of the task and the degree of variation in the players' new approaches to challenges in which they failed. That's why we have a Referee, after all: to allow characters to attempt anything when rules can't account for everything. Someone has to make the call. The issue then becomes, what keeps challenges fun? Allowing the players to wear down a rule system by repeated attempts to overcome the same challenge in the same way is not a satisfying, fun solution. Telling the players definitively "no" when the dice say no may make it more fun for them in the end.