How do you create a campaign that gives the players absolute freedom but still leaves the GM in control?

Ask the gamemasters

Recently, Campaign Mastery was asked,

Is it possible to create a truly player-choice driven campaign (within reason)? I’m trying to create a campaign that’s anywhere from 75 – 90% roleplaying (there will be dice, but not battle so much). I’ve been developing the campaign for some time now and have a list of interconnected rumors which, if followed in the right order, will lead the PCs to bigger and bigger discoveries. There are even totally different directions in which to take the campaign if another set of rumors is followed….

The problem is that, even though I’ve not had any of the rumors be false, if connected in the PCs’ minds in the wrong manner, the rumors could lead them to believe the wrong thing. (In trying to start this particular campaign, I had just that very thing happen.)

This can be problematic when trying to have the players wrap their minds around the plot. Also, to keep things interesting (since I’m using rumors as a starting point from the game), should I give out red herrings as well? Any advice would be appreciated.

Ask the GMs - Mike

Mike’s Answer:

The good news is that it IS possible. In fact, it sounds like you are well on your way to achieving your objectives. However, it also sounds like you are well on your way to running a plot train (with limited switching points) and are concerned about it running off the rails. This would make it no more character-driven than a colouring book; but again, you seem to have escaped this particular trap fairly adroitly, letting the characters believe whatever conclusions the players have reached, regardless of whether or not that is what is actually taking place.

You should never force the players to wrap their heads around the plot – in a truly character-driven campaign, they will make mistakes and wrong assumptions, and reach faulty conclusions. But they will also learn from these.

So here’s my recipe for making your campaign truly character-diven, regardless of the amount of combat that you want to take place, based on the process that I am using in my Shards Of Divinity campaign:

Prerequisite: Active Players with Motivated Characters

The first thing to note is that none of this will work if your players are the types who need their scenarios served up to them on a plate. And the second thing to note is that it won’t work unless the PCs come pre-loaded with back stories that motivate them to be interested. A little personal stake in the outcome makes a big difference.

The key to achieving the first is something that’s eluded me for decades, so I can’t help you there, but it sounds like you’ve got that aspect of things cooking along quite nicely anyway (I’m moderately jealous, but wouldn’t trade in the players I’ve got anyway).

The key to achieving the second is to ensure that much of the character backgrounds are open-ended, unresolved, or otherwise unexplained. Even if the character thinks he knows why the Warlord slew his parents, and that the Warlord was captured and hung for his attempted revolution, and has written as much in his background, you can still place a hidden shadowy figure behind the known villain that the character knows nothing about – in which case, the revolution may have gone underground, but the wheels are still in motion, and rumours of distant events deriving from the revolutionary activities should resonate with this element of the character’s background.

Establish the big picture

The first step to achieving a truly character-driven campaign is to determine the big picture – who are the various vested interests, and who are the opposition (in general terms). Note that there will be many of these, and most of them will be side-issues to the real plot. Power brings power-struggles with it, almost by definition. It doesn’t matter what someone has, someone else will want it. Every significant power bloc should have a defined opposing power bloc, and the drive of the campaign should emerge from the tension, the move-and-countermove, between these groups. And don’t neglect the potential of groups that are being blackmailed into activities they would not normally countenance, or of groups targeting the wrong enemies, etc!

These opposing groups, and the current state of play between them, should be part of your campaign briefing materials, the springboards from which the players generate their character backgrounds.

But it is always possible to reverse-engineer the information from existing character backgrounds. In fact, I would go further than that, and state that the GM should re-engineer his campaign background after the PCs have been generated, just to make sure that everything fits.

One of the best techniques is to treat the character background generation process as a series of mini one-on-one roleplaying sessions. Take an hour with each player to work through the past history of the character and how the different factions and campaign history have influenced them. This can either be done in place of a day’s play, or can be done away from the table.

Generate rumours

There’s usually a vast gulf between what is actually happening and what the general public thinks is happening. Generate rumours that describe the latter, and pre-load your characters with them. These are usually the mere tip of the iceberg in comparison to the real story, but these breadcrumbs are starting points. Some of these will never become relevant to the campaign, some should be out-and-out wrong, and some should hold a nugget of truth. Some should directly conflict with the rumours given to other players. Some should be wildly improbable. It’s up to the character to assess how credible he finds them.

To generate these, I usually start with a true statement describing an event and break it into short, declarative sentences. I then wash each sentence through a “Chinese Whispers” process – rephrasing and restating the message, changing factual information slightly, occasionally inverting or distorting the meaning. I’ll generally roll a d6+2 for the number of times each sentence should be ‘processed’ – what comes out is a bunch of rumours describing the incident.

Most of these changes will be in the direction of heightened drama, even of melodrama. Exaggeration, hyperbole, and rumours go hand in hand! Sometimes I will like the ‘amped up’ version so much that I’ll make it the literal truth and start the process over!

Generate propaganda

Every power bloc will generate propaganda of some kind. This might be to play up or play down their activities and motivations, or to assign blame for something to their enemies (however big a stretch it is). Each PC, by virtue of who they are and who they have had contact with, will be ‘exposed’ to this propaganda, which can also be stated in the form of rumours; their relationship with the source will dictate the level of credibility.

If a rumour will not be believed by the character because their opinion is coloured by their existing relationship with the source, that rumour should be rephrased as a ‘conspiracy theory’ element before it is given to the character; instead of a rumour that “X did Y,” tell the character that “The [prejudice target] is trying to make people think that X did Y,” where the ‘prejudice target’ is someone that the character already thinks the worst of. I will often run the results through the “Chinese Whispers” process a couple of times before doing so. Exaggeration and Hyperbole, always!

This is usually impossible to decide until the character histories are known, and hence what rumoured actions the character will find credible. And the resulting rumours should be interspersed with those from the previous section, more or less at random, so that players can’t block-assign credibility assessments.

Determine next actions

Once you know who the power blocs are, and how they operate, you can also decide what each faction’s next step is going to be, even if the answer is “lie low until the heat dies down”. Have any recent events in the campaign background presented an opportunity that is too good to resist? Have a faction take action accordingly. Have any recent events left a faction vulnerable? Alliances may be formed or broken, leopards may change their spots, etc. Every faction should always be doing something to bring advantage to themselves and disadvantage (or discomfort) to their enemies – even if they are nominally on the same side!

Generate rumours of the results

You don’t have to decide on the outcome just yet – these are events that are taking place concurrently with the PCs adventures, and in which they may take a passing interest or even an active involvement. You should always scale these broad actions down to determine what the local events are that the PCs will see, first-hand, based on where they are and where they intend to go next.

The other thing that needs to be done is to note when these actions will become apparent to the faction’s opposition, and how long they will actually take to complete.

Generate rumours of impending enemy responses

Every action has an unequal and opposing reaction, to misstate one of the most famous laws of physics. This misstatement is literally true when it comes to the maneuverings of factions – their enemies will always take an opposing position, i.e. will react in some manner, and the strength of this reaction will always be greater or less than the original action, depending on the level of threat perceived in the original action. The smarter the heads of each faction, the more appropriate their response will be. Some of these reactions will be covert, some will be overt, and many will have a secondary reaction designed to act as a ‘cover story’ or misinformation to distract the enemy while the real move is being made.

Of course, no reaction can take place until the target faction knows there is something to react to, and by the time an action is complete, it’s usually too late to undo the effects. The numbers assigned to these two time-frames in the previous section dictate the window of opportunity for a counter-operation by the target faction, and the later into that window a reaction comes, the more likely it is that it will need to be some overt response.

Determine actual enemy responses

It’s my preference to come up with half-a-dozen possible reactions in the form of rumours, ranging from the improbable to the near-certain, wash these through the “Chinese Whispers” technique a few times, then pick the actual reaction that will take place.

None of these reactions will take place in a vacuum – all the moves being made by other factions, against other targets, will change the environment just a little. Other, presently uninvolved, factions may also get involved because they perceive an opportunity or a threat.

Each reaction then becomes a new action in this never-ending game of chess.

Update ‘The Big Picture’ regularly

These ‘boxing matches’ may be economic, or political, or social, or related to intelligence-gathering, or propagandist, in nature. A major side-benefit is that the campaign background is never static; it is always evolving, and introducing new plotlines that the PCs can get involved in. This obviates any need for plot trains at the same time as it brings the world to life for the players.

The downside is that as each action matures, the campaign background needs to be updated. For this reason, I like to stagger the outcomes of actions, making some brief and opportunist, and others subtle, preplanned, and long-term. The nature of the faction clearly needs to be a factor in these determinations. As a rule of thumb, I like no more than two factions to have acted or reacted in any given post-session update; this keeps the campaign admin at a manageable level.

Another advantage of this approach is that the material that can be given to new players, or new characters, is always pretty much ready-to-go. And if you take a holiday it’s relatively easy to get back into the swing of things by reading through your updated background.

Listen to your players’s ideas

Your players will put 2+2 together and come out with 7 on a regular basis. Sometimes, their ideas will be better than yours; sometimes, you will have a grand plan that their ideas don’t fit. You should always take the player’s theories and consider them carefully to decide whether or not they are better than your own. And if they are, you smile, give them a metaphoric pat on the back, and never admit the truth! They will not only think of themselves as brilliant (and pride goeth before a fall), but because you agree with them, they will see you as brilliantly Machiavellian to boot.

Make an INT check for the speaker

One approach is to secretly make an INT roll for the character whose player announces the theory. Succeed, and their theory is correct; fail and they’ve got it wrong. All game systems have some sort of mechanism that addresses this need, or can be adapted to it. The result is that no matter how brilliant the player might be, it’s the character whose brilliance or lack thereof that is reflected in the campaign events. Of course, if you really like the idea, you can ignore or even fudge this roll!

Even if there are some plot holes or logic errors to be filled in, this can still be a powerful weapon in making the campaign credible.

Red Herrings

These should be part and parcel of any set of rumours. In fact, many of the results of the steps described above won’t be characterised as anything else by most DMs. The alternative is for the rumours to give a roadmap to the overall plot, once a few side-roads created by mistakes in logic (or simple misjudgements by the players) are eliminated. Nothing kills game-play quicker than being able to predict with certainty every step of the rest of the campaign; it kills the suspense.

I have seen suggestions that red herrings should outweigh true rumours by 5-to-1 or more. I don’t agree with that; much of this quota will be met by rumours that are related to the truth, but heavily distorted, and I don’t consider those to be true “red herrings” as they still advance the plot.

A genuine Red Herring is a deliberately-false rumour that is inserted into the mix for the express purposes of deception. The problem is that if there are too many of these, players will never be able to sort the wheat from the chaff and, as you put it, “get their heads around the plot”. Worse still, the players may come to feel that the DM is deliberately trying to deceive them. That’s why the majority of false rumours are not red herrings per se, just propaganda from one faction or another; that then becomes something that illuminates the nature of that faction, and something that is “their” fault, not something that is “you vs the players”.

I think that a ratio of one-in-five or even one-in-ten is about right, representing people who invent wild stories out of thin air for the pleasure of hearing themselves speak (and to make themselves, however briefly, the centre of attention).

Pink Salmon

I am especially fond of rumours that can be easily dismissed as Red Herrings – but aren’t. A few pieces of “Pink Salmon” – to extend the metaphore – makes everything more credible. It’s always possible that some of the wild stories invented by these local wits and spread far and wide by the gossip of passing travellers actually hit the nail on the head.

Be selective, though; you don’t want to introduce a “the more improbable it sounds, the more likely it is to be true” dynamic into your rumours, because that also kills the mystery and suspense of the campaign. As a rule of thumb, 1-in-5 or 1-in-10 “Red Herrings” should actually be Pink Salmon.

White Salmon

PT Barnum could sell just about anything. The story is that he once came into possession of a truckload of grey salmon, which are neither especially popular nor tasty. He relabelled them “White Salmon,” (there’s no such thing) “Guaranteed never to have been pink”. And sold the lot at premium prices.

Perception and salesmanship counts for a lot. Rumours are no good if you don’t sell them to the players through roleplay. Any tavern scene should involve a chat with the local gossip to get the latest unofficial news and a sense of local affairs. These should be two-way conversations, in which the players get to pass on rumours they’ve heard as well as receive new ones. In essence, you have to label your rumours as “White Salmon” and sell them to the players.

It might seem that it’s enough to convince the characters of the validity of rumour #112, but it’s not. The players may have their characters act as though deceived, but it’s not very satisfying when the players know better. So make no mistake, this is an essential application of metagaming that can have a major influence on the entertainment value of your game – you might be pitching your rumours at the characters, but it’s the players you have to convince.

Maintain an atmosphere of uncertainty

This is so important that, even though I’ve mentioned the maintenance of suspense a couple of times, it bears repeating again. Some GMs suggest that every time the players figure out what is really going on, the universe should be torn down and replaced with something even stranger (shades of The Hitch-hiker’s Guide To The Galaxy!).

Personally, I wouldn’t go that far; you risk a complete loss of credibility and of the players losing all confidence in their ability to actively decide their own fates. In effect, you can convince the players they are on a plot train even when they aren’t, if you go too far – I speak from personal experience.

This requires a judgement call or two on the part of the GM. Will the player’s knowledge of the truth impact their immediate activities? Then they’ve got it right, and should be congratulated. Will their knowledge of the truth have a negative impact on the entertainment value of future scenarios, or will it enhance it by offering a glimpse behind the curtains? If the first, then they’ve got it wrong and the GM needs to introduce a new explanation for what’s happened; if the second, then they have it right.

If both these judgement calls leave the question in a gray area in between, then the rule of thumb I use is that the players are probably right, but something prevents them confirming it – for now. It becomes a deeply-held suspicion of the characters, nothing more.

It is also important to note that juicy rumours don’t stop just because the PCs have proved or disproved them, and resolved the source upon which they were based; rumours should linger, and occasionally raise doubts about whether or not the problem really WAS resolved!

Imminence brings revelation and certainty

The more involved in any given faction-fight the characters get, the more reliable should be the information they receive. The closer to fruition a faction’s plan becomes, the more blatantly obvious it should become – though this is always a relative value.

Consider the land invasion of the allies in World War II: at first, it was just speculation when and where it would occur; then it became inevitable that sooner or later it would occur, but where and when were still no more than guesswork; then preparations for doing so got underway and deception and misdirection were employed to make sure that the enemy didn’t know where and when it would take place; and then finally it actually happened, at which point it quickly became certain where and when it had taken place.

Example of the “Chinese rumours” technique:

1. Bishop Luabird spent $100 on lady’s underwear.

2. A Bishop in good standing spent $100 on exotic lingerie.

3. A disreputable bishop in the south regularly buys exotic lingerie.

4. A disreputable bishop in the south wears exotic lingerie.

5. A disreputable bishop in the south wears the skin of a devil as underclothes.

6. A bishop in the west wears the skin of a devil.

7. A disreputable bishop in the northeast is controlled by the skin of a devil that he wears under his clothes.

Now feed the last three to the PCs! Notice that even if the original rumour is true, there is no explanation of why the Bishop needed the underclothes!


The key to running a completely character-driven campaign is to work out what the opposition is doing, generate rumours accordingly (including exaggerations and red herrings), and then let the players interpret these as they will. Rumours will often persist long after the cause of the rumour has been resolved – which should leave your PCs always wondering whether or not they have really solved “The Riddle Of The Headless Horse Thief” or whatever. Then let the players decide what their characters deem interesting enough, or important enough, or even simply credible enough, to get involved in. And when they make that decision, the first thing they are going to have to do is find out what’s really going on – which means roleplay.

Ask the Game Masters - Johnn

Johnn’s answer:

Great process, Mike! I especially like the rumour washing. That’s a nifty little engine GMs should master because they can get so much value from constant use of it.

I don’t have anything to add to Mike’s process. So instead, I’ve distilled this post into a series of steps and created a short PDF game master checklist everyone can download and use in conjunction with this post.

pdf-iconGrab the checklist. Then use it along with referencing this post to help guide you along.

Ask The GMs is a service being offered by Campaign Mastery. More info >

Related Posts with Thumbnails
Print Friendly