TRAVELLER: Out of the Box–The Casual and Improvisatory Nature of Early Traveller Play

lScreen Shot 2016-05-17 at 12.12.37 PM

Years ago, when I first read the original Traveller (Books 1-3) something confused me:

I could not figure why there were so many random tables. Not just Animal Encounters, mind you. But random encounter tables for NPCs, Starships, Patrons, and possible Legal encounters. The game seemed so busy outputting the random events and troubles for the Player Characters I had no idea where the Referee could fit in the adventure he had built.

If one total them all up:

  • Legal: Daily throw law level or less to avoid legal harassment.
  • Random Encounter: Daily throw 5+ (on 1D) for a random encounter to take place.
  • Animal: Daily throw as directed on animal encounter table for terrain and
  • World type.
  • Patron: Weekly throw 5+ (on 1D) for patron to be encountered.
  • Starship: A roll every time the PCs arrive in system, with the Throw ranging from 6+ to 9+ on 2D6)

That’s a lot of random die rolling producing a lot of random encounters!

A couple of years later I read Adventure 1–The Kinunir, and I had more questions:

There are deck plans of the ship, but no maps of the other environments mentioned in the scenarios. Also no details about how to run the adventure, what the cultures are like, and so on. The book contains four scenarios that are basically a gloss of situation for each one and little else. How could this be called and “adventure” when so much work was left for the Referee to do!

Keep in mind, I had been playing D&D, where I crafted dungeons for my friends with care. This over-randomized, under-detailed adventure philosophy sent my head spinning.


A couple of years ago I decided to re-read the original Traveller rules from the boxed set and really dig into them — which is what produced this series of posts.

(Again, the assumption of this project is that original Traveller (Books 1-3) were well designed, with pieces that all work together well. Rather than stripping things out willy-nilly that I didn’t understand or didn’t think would work, I was to put myself in the headspace of the game and see what sort of play would be produced if I used the rules as written.)

I kept getting stuck on all the Random Encounter rolls. Especially the Patron Encounters (which I’ll focus on in a moment), but really, when combined in total, these rolls had a huge implication for play.

Now, I know all of us know some variation of, “Well, I have my adventure, and then the Players encounter it, and so I never know how it will all turn out.”

I need to make something clear: I’m not talking about that.

Each one of the encounters listed above is something that can lead to many results that will impact how the evening’s session goes.

Using the Surprise Rules, one side or the other might be able to Escape or Avoid the other side completely. This might generate a few minutes of play. Or, if on side or the other fails to Escape, the encounter might escalate.

Or there might be a brief conversation that ends after five minutes of play.

Or it can lead to some tension, where the NPCs or the PCs want something from the other, and one or the other side is trying to diffuse the tension. Maybe fifteen minutes of play.

Or it might turn out to be a full-blown firefight or battle. Let’s say half an hour.

Or there might be a firefight, with consequences: Having to flee, needing to get off planet, the need for medical care, getting lost, the loss of equipment, the group getting separated. This might become the next hour to three hours of play.

Or the encounter might have even larger implications: arrest, desperate negotiations, the offer to help someone that leads to a whole new adventure, the need to do research, the need to pursue someone or something, the need do recon on a new environment, the need to get off planet, the need to go into hiding, or be called upon to do a favor or service… and frankly, the possibilities are endless. All of which could lead to hours of play, if not sessions of play.

In fact, it is possible that the fallout from one encounter, if the Referee is playing the game with the looseness that the game seems to want, might lead to weeks of play.

It is possible that the fallout from one encounter might end up being the spark that provides the spine for an entire campaign of play as one thing leads to another and the Players become interested in the details from the encounter and make choices for their Player Characters that drive them forward in ways the Referee had not anticipated at all.

So, if the Player Characters are wandering through the wilderness looking for a rare mineral (because that’s the adventure the Referee has prepared), and then (through a random encounter die roll) get caught up in the conflict between two warring tribes (because that’s what the Referee fleshes out on the spot), the next few weeks might involve the efforts of the Player Characters helping one side or the other, or working one side against the other, or trying to escape the whole mess.

Literally, the game becomes about the fallout from that one random encounter roll.

Note, I’m not saying that the game only depends on Random Encounters. I am saying (look at the bullet point list above) that Random Encounters of various stripes can slam into any situation the Referee has planned and send it off course. And if one uses the rules as written, that’s likely to happen.

You either think this an awesome way to play or you think it’s too outlandish to deal with. But if you get into the headspace of late ‘70s RPG design and play, it becomes a perfectly viable and exciting way to play.


Now, we have to take a step back in time, to 1977, when GDW originally published Traveller. We have to do this to get some perspective and leave behind some assumptions about RPG play we have.

First up, in 1977 there are no “stories” of any kind published and sold to use in RPGs. Such things simply do not exist. What do exist are environments, like dungeons, or megacities that are big dungeons with lots of people. To go even further, there are no metaplots. There can’t be, if there are no stories.

Second, in 1977 there are no published settings for RPGs. (In 1977 both Gary Gygax and Marc Miller assume players will want to create their own settings. At this point there is no Greyhawk Folio (and will not be until 1980). And as Miller has stated several times, GDW would not start thinking about a setting for Traveller until it started working on adventures for the game, the first of which is published in 1979, two years after the publication of Traveller.)

Third, while pre-plotted, meta-plot driven game products will soon be the rage (DragonlanceThe Traveller Adventure, and countless others across most game lines), at this point RPGs are inspired by the episodic adventure tales of the pulp Fantasy and Science-Fiction era. No one is thinking in terms of a singular, long arc of a story, but in terms of a series of encounters and tales spun by the actions of vagabond adventurers making their way across adventure rich environments. It is the tales of Conan, Dumarest, Fafhrd and the Gray Mouser, van Rijn, and other characters that inspire early RPG design and play… not long, focused epic like The Lord of the Rings.

And fourth, all this rolls up into one big point: In 1977 both original Dungeons & Dragons and original Traveller thought they were complete games. Both games assume that the players will take the framework provided in the set of rules and go off and have hours of pleasure rolling up encounters, buildings worlds, setting up situations for the Players to puzzle through and adventure through. There are no settings, no published modules, no metaplots… and at the beginning no one assumes there a market for publishing such things will exist.

Given all this, is it a surprise that Marc Miller builds a game designed specifically to help you generate material and adventures on the fly at home?

And how did he decide to do this? With random dice rolls acting as “prods to the imagination.”


When reread the original Traveller rules a coupe of years ago, the point I kept going back to was the Random Patron Encounter rules, again and again…

Screen Shot 2016-05-30 at 11.05.04 PM.png


The key to adventure in Traveller is the patron. When a band of adventurers meets an appropriate patron, they have a person who can give them direction in their activities, and who can reward them for success. The patron is the single most important NPC there can be.

A patron will, if he decides to hire a band of adventurers, specify a task or deed to be performed, and then finance reasonable expenses for the pursuit of that task. Some tasks may be ordinary in nature, such as hired guards or escorts; other tasks may be for the location and procurement of items of great value. Generally, a patron’s agreement with a band of adventurers will specify that the patron will receive the item he is seeking while all other goods or items acquired will belong to the adventurers.

In a single week, a band of adventurers may elect to devote their time to encountering a patron. They may frequent bars, taverns, clubs, perhaps the Travellers’ Aid Building, or any other likely places. One throw is allowed for the entire band: a result of 5 or 6 on one die indicates a likely patron has been found. Two dice are then thrown consecutively, and the patron table is consulted to determine the general character of the potential patron. If necessary, the patron’s personal characteristics are generated at this point. The band then meets with the patron, and an interview takes place. Throw two dice on the reaction table to determine if the patron concludes that the band will be suitable (generally, a throw of 7+ on the reaction table is sufficient). The patron then discloses his task, and the adventurers may accept or reject the offer of employment.

Once the patron and the adventurers have met, the responsibility falls on the referee to determine the nature of the task the patron desires, the details of the situation (perhaps a map or some amount of information), and the limits of the patron’s resources in the pursuit of the task.

The reason I kept going back to them was this:

  • Let’s say I’m the Referee of a weekly game
  • Let’s say I’m using the rules as written
  • Every week of game time, the Player Characters can look for a Patron
  • On a 5+ Throw on a 1D6 they find such a Patron (33% chance per week… pretty good!)
  • I then make two D6 rolls on the Patron Encounter Matrix to randomly determine who the Patron is… thirty six possible results!
  • I then, uh… make up an NPC, have the NPC make up an offer for a job…
  • The PCs can accept or refuse the job…
  • If the NPCs accept the job… we begin playing that adventure right then and there and I, uh… just make it up?
  • If the PCs refuse the job (for whatever reason, and in my view the Players must have the right to refuse the job, or why bother rolling randomly?) then another week passes and we do it again until they accept a job… and, again, I make up the adventure on the spot?

This is a very different experience than going out and buying a pre-made module of some sort. It’s even very different than the Referee having a light sketch of a scenario and changing it as the Players deal with it.

This is starting everything from scratch, during play, and then revving up from 0 to 60 right at the table.

Now, I want to make something clear. I have no fear of improvising play at the table. That’s a whole different discussion, but whether its Pendragon or Sorcerer or Primetime Adventures and countless other games, Refereeing on the fly is something I’m comfortable with. The trick is, I didn’t expect to find it in an old roleplaying game—because I had conflated the original rules of Traveller with the later Traveller products (the Ancients sequence running through the Adventure Book line, the railroad plot of The Traveller Adventure, and the detailed and overwhelming setting of the Third Imperium) which focuses on lots and lots of detail and need to keep the PCs steering in the “right” direction. (Even so far as to constantly cripple their ship, as happens in several Double Adventures, to make sure the adventure can happen.)

Nor am I saying the Referee can’t plan scenarios and situations. Having ancient ships floating in space, and compounds to raid, and wilderness environments to explore is part and parcel of the Traveller experience and there is no reason to assume the Referee can’t detail these things to his heart’s content. The point is that the rules, as written, encourage a kind of play where these things can be complicated or derailed by new random elements. One either embraces this, rejects it, or makes peace somewhere in the middle.

That all said, the implied setting of Traveller play is a sprawling mess of worlds with countless possibilities. Confronted with the thought of having so many options, my mind kind of short circuited. I mean, if the Players can stumble across a random employer and I’m on the spot to come up with an adventure to run right then and there, there is on question that had to be answered:

“Was Traveller written with the notion that the Referee would simply improvise and evening’s session of play on the fly?”

The answer, I decided, was yes.

This answer confounded me, since the publishers of RPG material seemed to have made it clear that the point of prepping an RPG scenario was to run the Players through a planned scenario of some sort, often elaborate and on rails to make sure the Players reached certain climaxes. (See: The Traveller Adventure.)

And then I came across an article written by Marc Miller about how he ran a session of Traveller for his friend’s one evening.


Inside issue #40 of The Space Gamer, on pages 4 and 5, you will find an article by Marc Miller recounting a session of Traveller he ran for his group in 1981.

He begins with the text of Adventure 4-Leviathan…

Vior 0805-X500401-1 Non-industrial G

In appearance this world is uninhabited (and uninhabitable), however a detailed survey will indicate geological anomalies in one particular location. Investigation will reveal several airlock chambers giving access to an extensive underground habitat, where the Viorans lead a troglodytic existence. In one of the worst cases of regression yet found, chronic overcrowding has produced a primitive semi-cannibalistic society. Healthy crew members will be greeted primarily as a source of food.

The Viorans produce nothing of value; however, one reason for their overcrowding is longevity. After much detective work, assuming the researchers have not been eaten in the meantime, this longevity may be traced to the inclusion of a particularly repellent rodent scavenger in their diet, with remarkable anagathic properties. The planet itself has a number of extractable deposits of various minerals.

He then improvises the entire evening of play.

I came across the article when I was searching for clues about how people played Traveller in its early years. In particular, I became fascinated by all the tables in the game to produce Random Results. These tables, combined with the fact that the Referee of Traveller is somehow responsible for building way too many worlds in detail if he’s going to spend any time playing, boggled my mind. The published material from GDW about The Third Imperium promised some sort of concrete, certain setting. And yet, if one reads the rules contained in Books 1-3, the game seems to be about creating something fluid and on the fly.

And sure enough, when I read Miller’s description of how he Refereed the session, there it was. An entire evening of RPG play with just a few clues at hand, each clue sparking his imagination to create a series of mysteries and encounters for the Players.

I really recommend you read the article.

After reading the article, I knew I wanted to do a post about it.

But I got to this post, however, I came across a thread over at the Steve Jackson Games forum, in which Bill Cameron sums up all the points I would want to make:

More playing than prepping – Mr. Miller describes the session as being an almost a spur of the moment event. It’s a holiday weekend, IIRC, and he puts together a group from friends and family in order to provide an evening’s entertainment. He doesn’t spend hundreds of hours putting together some seamless and uber-detailed setting, he just grabs an encounter out of a published adventure instead. His players don’t spend hours putting together min-max munchkins either, they just grab characters from the published campaign. Everyone is there to play and have fun. The idea of “winning”, “leveling up”, or other metagame metrics for “success” isn’t even mentioned. All they want to do is play.

More descriptions than spreadsheets – Rather than providing reams of documentation, Mr. Miller instead “verbally illustrates” the surface of the planet, the primitive “airlock” system, the underground caverns, the cannibals living there, and much more. He doesn’t “design” the caverns using some construction supplement in order to figure out everything down to the number of rivets holding the handles to the night soil buckets, instead he describes what the PCs are seeing. He doesn’t have a huge list stats/skills for every potential NPC either, he makes those up when and if they’re needed. He doesn’t bog down the session or his prep for the session with unnecessary trivia.

More role-ing than rolling – Granted, CT didn’t have a task system yet and, apart from a few specific tasks/DMs scattered throughout the few books, there isn’t much guidance regarding when to roll the dice. That being said, the session Mr. Miller recounts doesn’t seem to have much rolling at all. He describes the situation, the players describe their actions, and the game moves on. The dice aren’t picked up every time someone needs to inhale, talk to somebody, or look at something.

More talking than shooting – The session does feature some combat, but it occurs near the end when, after interacting with the natives for some time, the players finally realize that the natives are cannibals and see the PCs as a new source of protein. What the players don’t do is waltz into the caverns with FGMPs, battledress, and itchy trigger fingers ready to slag anything that moves. Even when the encounter with the natives slowly deteriorates, the players prefer to Jaw, jaw, jaw rather than War, war, war. The guns – and the dice – only come out when the players need to secure their retreat to the surface and then, rather than burn the caverns to the ground, they only use enough force to escape.


Did you notice:

  • Miller didn’t have an adventure planned out… at all?
  • That he used the UWP of Vior and two paragraphs of description to spin an entire evening’s play on the fly?
  • Vior was described as a) having a population; b) having no atmosphere; c)having airlocks; and d) having a Tech Index of 1…  and yet the contradictions some might find in these details did not bunch Miller up into a panic of “This makes no sense,” but instead offered a chance for him to invent a clever and intriguing environment for the characters to puzzle through?
  • Miller did not concern himself with a fully realized background built from countless books printed by GDW (countless books had not been printed yet), but grabbed details from a short story he had read to flesh out the environment?

Keeping all of the above details in mind, lots of the details about the design and structure of original Traveller fall into place.

The Main World generation system is described as “a prod to the imagination” for the Referee. (That is, it was a tool for Referee, not a method for the Scout Service to classify the planets within the fictional environments. This strange twist in the definition of UWPs would come later.)

Now look at the Leviathan article and see this prod put into action!

I offer that a large chunk of Traveller is designed for exactly the same purpose:

The Patron Encounter Tables, the daily Throw to avoid legal harassment, the daily throw for a Random Encounter with NPCs, the daily throw for an Animal Encounter, the Starship encounters. The creation of Main Worlds. The creation of the Animal Tables. The rolling for Cargo and Passengers.

All of these things are there to prompt the Referee to come up with ideas on the spot, to present the Players with obstacles and opportunities on the spot, the have the Players make choices and take actions via their PCs on the spot.

Overall, one is left with the realization that original Traveller was never meant to be beholden to a huge, sprawling setting with every detailed worked out, from economics to politics. Instead, one finds a setting full of the same sort of haphazard, random adventures of stories form the 50s, 60s, and 70s science-fiction and fantasy short stories and short novels that Miller read and that inspired him to write the game. In this kind of storytelling, characters travel from one place to another, encounter, explore, and interact with strange and exotic environments, meet new characters along the way… and the “story” is the details left in the wake of the characters’ actions and decisions as these encounters take place.

This is all something to think about as you prep your Traveller play:

How much world building and setting building do you really have to do first? How lightly can you sketch environments, and then move onto the next, knowing you can fill them out later as needed during play? In what ways are the Random Tables not a “distraction” to the story you want to tell, but an aid to building situation, conflict, opportunity, and opposition? What is the focus of play: The fully planned setting full of details the Players might never get to? Or enough setting details in front of the Players that they can interact with, touch, and explore?

There are many ways of playing RPGs and creating settings. I write this post as a reminder of a certain style of play that was lost as RPG consumers got more and more caught up in studying official settings published by game companies, and a certain kind of play that worked very well 40 years ago. A kind of play the rules for original Traveller were written to support and handle well.


13 thoughts on “TRAVELLER: Out of the Box–The Casual and Improvisatory Nature of Early Traveller Play

  1. Out of curiosity, I just looked up Vior on and find the UWP has become: D500401-7. I don’t have my copy of Leviathan to hand, but that isn’t the UWP given in Miller’s article so I assume Leviathan has the X500401-1 version.

    Now I don’t know how or when the UWP changed, but I would guess someone came along and adjusted the UWP to make it more “sensible”. Tech level 1 with no atmosphere? Doesn’t make sense – change it!

    Except in the process they made the world much less interesting than the place Miller dreamed up.

    That seems a shame, and a good example of the downside of treating the game as a reality simulator, rather than as a facilitator of SF adventure gaming.

  2. Another enlightening article. I have all the Traveller Books and Supplements and probably the Adventures, too, but I find I enjoy using just the original boxed set of three Books from 1977. ’81 version has some changes I prefer not to use. And currently, as I’m reading volume 20 “Web of Sand” in the Dumarest series, I am really envisioning a fantastic Traveller game…need to recruit some players interested in this style of gaming. The older I get, the more I appreciate the originals of Traveller and D&D as spurs to the imagination and game play when compared to latter day versions.

  3. I think this “emergent play” process is easier to use now than it was back when it was invented- because I have ready access to computerized random generators. I don’t know of a Web version of the Random Patron table, but I know in principle I could write one. I could randomly generate a Traveller patron with a button press if I needed their characteristics. I do analogous things in my D&D game all the time, and play never grinds to a halt while I roll dice and collate the results. It’s awesome.

  4. Pingback: Misha Burnett on How Classic RPGs Were Really Played | Jeffro's Space Gaming Blog

  5. Pingback: TRAVELLER: Out of the Box–What We Mean When We Say “Encounter” | Tales to Astound!

  6. Pingback: SENSOR SWEEP: Creepy Elves, Prattian Magic, Dangerous Demonization, and Soaked-In Misogyny –

  7. Pingback: The Sandbox vs the Railroad… or Spontaneous Story Creation vs. Pre-Plotted Story | Tales to Astound!

  8. According to the article Mr. Miller “was caught up short, not having prepared too much for the night’s action”, so I doubt that “making things up on the fly” is the default truly old-school definitively sandbox way of refereeing Traveller. If you read the thread at Steve Jackson games carefully, as well as another thread linked to it, you can see that points made by Bill Cameron are also not some kind of ultimate truth about the way Traveller was played in late 70’s – just his own subjective view on the matter.

    I think extensive preparation and refereeing by the rules method as well as pure gm fiat and storytelling improvisation existed from the beginning, so one is not more old-school than the other – it is just the matter of individual preference. I made things up on the fly in late 90’s and I don’t see anything ‘old-school’ in it – it was a rather natural shift from trying to run official modules and railroading players for years before that. The randomness of the rules gets in the way, so you decide those are a mere “inspiration” to be ignored at will, while game sessions turn into collaborative storytelling without any dice being rolled. After the initial feeling of liberation wear off it turns out all this impro method (or rather absence of any method) is not that different from railroading players down some linear plot. What they have in common is absence of _meaningful_ player choice.

    According to Mongoose Traveller Rulebook:

    “The Referee should only call for checks:
    • When the characters are in danger.
    • When the task is especially difficult or hazardous.
    • When the characters are under the pressure of time.
    • When success or failure is especially important or interesting.”

    Adventure without any checks is a dull experience I guess, as nothing interesting and important happens. Players don’t make any meaningful choice, the one that leads to ‘winning’ or ‘loosing’, without which the process cannot be considered a game at all. Maybe making up stories, retelling stories from some sci-fi book, but not a game. That’s why Patron mechanic is considered incomplete for out-of-the-box use, because it is. Patrons are merely frameworks for creating complete situations out of them, with detailed npc stats, loot, location maps, encounters, probabilities in time – whatever system code is enough to run it as a ‘game’ and not just some made up story.

    • Thank you so much for taking the time to reply.

      I’m not sure how to say this, so I’ll come out and say it: I don’t think there is much we disagree about.

      When I quote Bill Cameron pointing out there was “role-ing than rolling” I’m not — at all — saying there should be no rolls. Whatever concerns you have that I am advocating for “GM makes up a ‘story’ and the Players just follow along” is the exact opposite of how I Referee and the exact opposite of how I advocated Refereeing in every post I have made on the blog.

      In fact, the quote you make from Mongoose Traveller

      • When the characters are in danger.
      • When the task is especially difficult or hazardous.
      • When the characters are under the pressure of time.
      • When success or failure is especially important or interesting.”

      Is exactly how I think Classic Traveller should be run as well. Exactly.

      Please note that the operative word for me in Cameron’s quote is “more.” There are plenty of Traveller players who have their characters make rolls for everything: to take off into orbit, to put on a vacc suit, drive an ATV over typical terrain. This issue becomes more pronounced with later editions as more and more skills are added to the game and there are more and more rolls that can be made.

      But I believe, as does the text of Mongoose Traveller, that the rolls should only be made in significant moments arrive. The point of that aspect of the post it to make it clear that as the PCs are clearly competent professionals then one doesn’t need to roll all the time. A lot of play simple is conversation, and dice are used at moment of uncertainty and criss. This issue is even more pronounced while playing Classic Traveller, since there are fewer skills listed.

      But in these interesting moments of crisis and we roll the dice and and find out which way events are going… not because the Referee knows where things are going or where they should be going, but specifically because he does not.

      In the same way I am not claiming that no one ever prepped a game for Classic Traveller in the 1970s. But I am saying that the way some people have when approaching Traveller — fixating on building out countless subsectors (if not sectors), detailing dozens of worlds, cramming countless sourcebooks into their head in order to be “ready” to run the game — is a boondoggle and that a lot less prep is required. As demonstrated by Miller’s game. That was my point.

      (I need to add that the only part of what Cameron wrote in that post is the part I cared about. He went on to say other things that I’m not that interested in and don’t agree with.)

      I am an absolute advocated of Players making meaningful choices. It is what I play the game for, as both a Player and a Referee. I wrote an essay in White Wolf magazine back in the early 90s about how Modules and pre-planned stories are the opposite of what I want from RPGs. I stand by that today.

      On this blog I have written about the value of Random Encounters and Random Tables for generating ideas and content — even on the fly — for the very reason that they prevent “game sessions turning into collaborative storytelling without any dice being rolled.”

      All in all, I can see how you read the post to mean what you seemed to read it as meaning. But I can assure you that (as far as I can tell) you and I are on the same page about lots of things. I’m not saying we share all points of view, or play the same way. But we are more similar than I think you might think.

      Ultimately I advocated for neither railroads (which I hate) nor sitting around making up stories without rolling any dice.

      I do believe the Referee can come up with situations, tactical choices, and meaningful choices on the fly or with prior prep.

      Where I draw a distinction from you, I think, is that the Referee needs to do all the prep before play begins. For me, whether the Referee is creating these situations and choices for the Players one week before play, one hour before play, or thirty seconds before needed during play is moot: The Referee still has to come up with it and offer it to the Players. The time frame before it being offered is not significant in my view.

  9. Pingback: An Improvised Classic Traveller Convention Game | Tales to Astound!

  10. Pingback: More On Pointcrawls — and Thoughts on Pointcrawls for Traveller | Tales to Astound!

  11. Pingback: Implied setting – Technoskald's Forge

  12. Pingback: Run Meaningful Random Encounters: – Dungeons and Dragons

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s