Having a discussion with a friend about our views of GMing, I realized that we take very different approaches to what our worlds are.

My view is that when I GM, the world exists as-is before the players start playing. What I mean by that is facts of the past can't change based on what my players do.

My friend has what you might call a more player-focused plot. He rewards any sort of intelligent player action by moving the plot along as long as it can be done without seeming like that.

Here's an example. Let's take a game where there's been a murder. And let's say I decide that the person was killed by a lot of sleeping pills being put into their food. But the players, seeing no sign of injury, immediately start testing for well-known poisons.

My approach would be to say that all of the tests come up negative, but of course if any of the tests would suggest the presence of sleeping pills, I'd say so.

My friend would confirm the presence of some poison that they test for.

His way has the advantage of helping the players feel smart and moving the plot along without slowing down the action unnecessarily. But I really prefer my way, because I feel like I'm setting up a puzzle for my players. However I have definitely noticed that this has its issues because the problem of setting up a solvable-but-not-too-simple-puzzle is not exactly trivial. I usually refuse to give any more hints that I wouldn't expect to be present on the scene, and once this even resulted in players basically saying to each other "Well, I don't know what to do."

What's your style? Is it something in between?

  • 1
    \$\begingroup\$ yes, i prefer something in betwen, both as a player and a gm. i'd rather give them something and reward them for thinking outside of the box, but in some situations i'll stick with what i had in mind and make them work to get there (maybe witht he occasional hint here and there, depending on if they make their insight roles [dnd4]) \$\endgroup\$ – DForck42 Aug 29 '11 at 20:19
  • 3
    \$\begingroup\$ VTC - not well suited to Q&A format - It's an excellent discussion question, but there can be no one "right answer"... \$\endgroup\$ – aramis Aug 29 '11 at 20:49
  • \$\begingroup\$ You are nitpicking. If the players say, "I test for poisons," that implies checking the food/body for any sign of contaminants. I would not expect a player to say, "I test for sleeping pills, poisons, pollutants, tapeworms, or any other sorts of contaminants that might be in the food." Set a skill DC (Heal, or K:Nature?) and let their roll determine whether or not they identify the unusual nature of the poison. Bear in mind that you're not trying to "beat" the players. Introducing clues which the PCs are unable to find doesn't add value to the story. \$\endgroup\$ – RMorrisey Aug 29 '11 at 22:17
  • \$\begingroup\$ @Rmorrisey Your point is taken, but this was an arbitrary example that you've shown could have been better constructed. I don't GM games of murder -- I GM conspiracies, and players have theories about what's going on. \$\endgroup\$ – Jeremy Aug 29 '11 at 22:26
  • 2
    \$\begingroup\$ @Jeremy I think there's a good question in here, but "which way do you prefer fight" isn't a good fit for RPG.SE. It is an interesting dichotomy and really encourage finding a SE-legit question to ask on it though. Asking for the pros and cons of each approach would be legit, or asking "how do I avoid the problems of my method." \$\endgroup\$ – mxyzplk Aug 30 '11 at 1:48

There is no singular "right answer" to this question.

I myself can't answer it singly, even, as it varies by game system and group.

The extreme "player side" end

Running Blood and Honor, by John Wick, I can't even CONTROL the answer to the player's questions... as the task mechanic is high roller gets to decide yes or no, and most players get to say one or more things...

Emepli Gratis:
Fred: I want to test for poisons
GM: The poisoner has 7 dice, 3 set aside for risk
Fred: I've got 6; I'll set 1 aside
Mary: I've got 4: I'll set 1 aside, too.
Rolls F=19 M=16 GM=18
GM: Fred has privledge. I keep 2 dice, mary 1, fred 1. Fred?
Everyone takes those numbers of dice to hand
Fred: Yes, there is poison.
GM: (sets aside 1 die) There also is evidence of sleeping pills
Mary: (sets aside her only die in hand) But neither at toxic levels.
Fred: (sets aside his only die in hand) She died of an allergic reaction
GM: (Thinking fast, setting aside last die in hand) She was allergic to the wheat-starch used in the sleeping pills!

The swings both ways example

Mouse Guard, by Luke Crane and David Petersen, has a two-phase session structure. In phase 1, it's all GM-centered fixed succeed-or-consequences, players pick an option given by the GM. In phase 2, it's GM can only tell them to roll the dice for an action, when their narration indicates such, and set the difficulties based upon their narration.

In the GM phase, I might say, "You spot an eagle overhead. Do you hide in the grass and hope not to be noticed, call it's attention and try to look too dangerous to eat, or make the run for the trees?"

In the Player phase, I can't say anything until the player narrates something requiring a die-roll.

Further Discussion

Systems vary on how they approach the issue; the issue of the designer's approach to the preestablished reality of the setting is one that determines, in many ways, the nature of the rules.

Some other ones where players have narrative control: Fiasco, Burning Wheel, Mouse Guard, Burning Empires

Some presuppose a firmly set universe which the only things the players control at all are the PC's: D&D, Call of Cthulhu, Traveller; these are pretty explicit in their GMing advice.

Most can be run somewhere in between, including the ones that presuppose a fixed universe, but very few actually provide much rules support for player input. The only time such a fixed universe is actually enforced is in solitaire modules, and that also limits even player agency by restricting to a handful of choices.


I think it depends on what problem you, as a GM, are trying to solve.

1. My players aren't being innovative enough

In this case, reward them for taking initiative by having their ideas work, even if you hadn't planned it. Doing this creates a sense that the game world is malleable to the players, something they can interact with and explore and affect. You're essentially making the puzzle up as you go along.

2. My players feel that puzzles are too easy

When your players decide that puzzles are too easy when any answer works, increase the difficulty by planning the clues in advance. This creates the sense that there's a real puzzle with a real answer. This has a greater sense of achievement. You should invent the puzzle ahead of time, in order to make something solid and interesting.

If one of these approaches fails, you can always try a little of the other: occasionally say no when they're used to any approach solving the puzzle, and occasionally say yes when they can't find your clues but come up with a good idea anyway.


Not the answer you're looking for? Browse other questions tagged or ask your own question.