Is Your Journey Really Necessary?

We interrupt our scheduling programming to answer a question from Umberto Pignatelli (ciao, Umberto!), who wanted to know whether so much detailed world building is necessary to run Traveller.

The short answer is “No”. The longer answer, at least my longer answer, goes like this…

Traveller is a child of its time, namely the late 1970s and early 1980s, back when the Old School was the New School. Original D&D had been available for about three years, if you knew where to look for it, which in 1977 meant Games Workshop in Hammersmith (yes, that Games Workshop, but pre-Warhammer) or a couple of other places in equally seedy sidestreets.

In those days, the GM was expected to fill in the gaps in the rules and create his setting from scratch, himself. That was a natural result of where RPGs came from; initially they were written by, and for, tabletop wargamers, who were used to doing that and had all kinds of tricks for it, mostly spread by word of mouth as I recall. RPGs then were all sandboxes; the GM had to generate material for everywhere the players might go and everything they might do, because until the players sat down nobody – not even them – was sure what they would do next. This meant that games had to have random tables for things like encounters, which you’ll notice have largely disappeared from the current generation of RPGs.

After a year or two, games companies realised that one of the big obstacles to starting a game – and therefore, indirectly, to selling their products – was the amount of time, effort and imagination the GM had to put into generating the setting before anyone could play; at that stage, the only generally-available RPG with a setting was Empire of the Petal Throne. And thus setting books (and eventually adventure paths, which are a different answer to the same problem) were born; in the case of Traveller, the first published adventure – the Kinunir, in 1979 – had several pages of setting material in it, a pregenerated subsector with a map and some vague hints about the Imperium. Soon after, Supplement 3: The Spinward Marches, was released, and then things kind of snowballed.

The Rules As Written assume that you will generate at least one subsector for the players to adventure in, maybe 30-40 worlds. When you’re familiar with the rules, that takes about an afternoon to do, maybe longer if you want to draw a nice map, and then as much time thinking about backstory and plots as you need. (As an aside, a number of GMs went crazy and generated hundreds or thousands of star systems, but those campaigns tended to be stillborn, crushed under the weight of their own statistics.)

One of the beauties of this approach is that you can run a campaign with no GM, especially if you have a Free Trader starship, because that gives you a spine to build the rest of the game around – trading. A group of us did that for a while with a crew of scouts; the mission was to explore a subsector, and we diced up each world as we arrived, then either took it in turn to answer questions as they arose, or used reaction tests to answer them, almost a primitive version of Mythic.

But there are other ways, ways that don’t need so much prep time; it depends on the kind of game you want to run. Here are some I’ve tried:

  • You can make stuff up. Traveller’s world generation rules are actually intended for the situation where you’ve run out of ideas, and you need another planet. Until you run out of ideas, you just allocate stats to match your vision of the world, and give it a concise write-up in the form of a Universal World Profile.
  • You can use one of the published settings.
  • You can use a setting from books, movies or TV. A lot of people did this, and in fact one of the stated aims of Traveller was to allow people to explore aspects of their favourite setting that hadn’t been covered in the source material.
  • You can limit the players to one world, if necessary by having their starship break down – now they’re stuck until they have enough money to fix it. (I have fond memories of a game like this that my brother-in-law ran, set in the universe of Andre Norton’s The Beast Master, which had a definite Wild West vibe.)
  • You can make all the worlds the same, so you only need one set of stats. This works well for a universe like C J Cherryh’s Alliance/Union novels, where most star systems have only a space station as a colony as there are few habitable worlds. (This is the gaming equivalent of early seasons of Stargate: SG-1; all planets look like the same stretch of Canadian forest.)
  • You can allocate worlds as you go. In my last Dark Nebula game, I stayed a couple of worlds ahead of the exploring PCs – wherever they went, I pulled the next world off the stack and that’s where they were. Then that got written into the GM’s notes as if it had always been there, and I generated a new one to replace it in the stack. I also had a “default” secondary system in case I ever ran out – that’s what every system looked like until somebody went there.

There’s another angle to this, though; Traveller tends to attract the kind of player or GM who enjoys generating worlds, or characters, or starships.

For the average Traveller GM, generating worlds isn’t a chore, it’s more like a solo mini-game within the main game.

The Universal Empire Profile

One thought that has been meandering around my head for a couple of decades now is using the Traveller Universal World Profile to describe interstellar states.

By definition, such a state would have starport class A. The size, atmosphere and hydrographics codes don’t really work, although given that most Traveller states focus on controlling the communications routes between worlds rather than the worlds themselves, you could set them to 000. It’s the last half of the UWP that intrigues me the most.


On average, a group of 36 worlds will have one Population 10 world, two Population 9, three Population 8 and so on, which will give it a total population of 12,345,654,321. That’s using the standard Traveller rule of 2d6-2 for Population; the Mongoose space opera and hard science options are interesting, but too much work to figure out, as Population then depends on a lot of other factors.

The Imperium is said to have 11,000 worlds; that’s enough for any local fluctuations to cancel each other out, giving it a total population of 3,772,283,264,750.

Or in UWP terms, Population 12. The other major states in the Traveller universe are smaller, but not by a factor of ten, so we can use that as a rough guide to the population of any star nation.

(As an aside, the average population across that group of 36 worlds would be 342,934,842. Notice that the couple of really big population worlds skew the average a lot; the most common population is 100,000, on one world in six.)


This is 2d6-7 + Population. With Population 12, your interstellar empire has a Government type ranging from 7 (if you rolled a 2) to 17 (if you rolled a 12), with the most likely outcome being 12. In the canonical Traveller universe, we have:

  • Government Type 7 (balkanised): Aslan Hierate, Vargr Extents.
  • Government Type C (charismatic oligarchy): The Third Imperium, the Zhodani Consulate. You could make a case for the Imperium being Government Type B, but the Emperor’s personal influence on the Marches is minimal – for example, the entire Fourth Frontier War happened before he had time to respond – so I think actual power lies with the nobility and the Navy.
  • Government Type D (religious dictatorship): The Two Thousand Worlds, the Solomani Confederation. I put these at D because their loyalty is more to abstract ideals than to a group of sophonts.

With only six data points, that spread could happen by chance; but maybe there is an extra dice roll modifier I haven’t taken into account.


2d6-7 + Government. Range: 2 to 17, most likely is 12. However, anything over 9 is much the same. This is the rating that makes least sense, since it suggests any interstellar empire aims for rigid totalitarian control of its citizens. I rationalise this by analogy with the Imperial Rules of War; if you disturb the Imperium, it comes down on you like a ton of bricks, but as long as you don’t wake the sleeping giant by messing with its sovereignty or economy, it’ll leave you alone – not because it wants to, but because even a stellar empire doesn’t have enough money and troops to keep an eye on everyone, all the time.

The difficulty for the freebooting adventurer is that the rules are not codified; the Imperium makes them up as it goes along, so you’re never quite sure what will draw its attention. However, as the high tax revenue, high tech level, high troop levy worlds are those with Population 9 and 10, any star empire would focus its legal muscle on those – and we find that they have generally high law levels.


There are too many variables for an analysis as simplistic as this, but somewhere in your 11,000 worlds is a Population 10 world with Size 0 and a Class A starport (probably just the one), and that has a Tech Level of 1d6+13. So the best tech level in your empire is probably in the range 14-19, and in Traveller’s Charted Space we find that the racial capital worlds typically have TL 14-15.


You can represent an interstellar empire by a UWP code something like A000CCC-E, if you’re prepared to stretch the rules a bit. If, like myself, you are wont to run random adventures without necessarily generating a bunch of worlds, you can use that as shorthand for the empire as a whole.

This exercise also reminds me how powerful and flexible the Traveller world generation and coding rules are. They have not changed significantly since 1977, unlike pretty much every other aspect of Traveller. Marc Miller definitely got it right there.

End Game

In the absence of any players, I find myself re-reading my oldest Traveller rulebooks – the 1977 Little Black Books. To my current eye, Book 1 – Characters and Combat – is the most dated of the three original LBBs.

But, the thing about character generation in the 1977 edition of Classic Traveller is this: It’s the end game.

Fresh out of character generation, the typical PC is in his late 30s or early 40s. He’s the same age as Conan was when he seized the throne of Aquilonia. He’s done the spacefaring equivalent of all that dungeon-crawling crap, served his time in the trenches, and now he is ready to concoct "daring schemes for the acquisition of wealth and power", as Book 3 put it.

Comparing the weapons skills statistically to the other game we all played at the time, Original D&D, we see against an unarmoured target at optimum range, the hit probability for expertise level 1-2 (which covers most characters) is roughly equivalent to a D&D fighter of 4th to 6th level. Expertise level 3 is about the same as a 7th to 9th level fighter, level 4 to 10th-12th level, and level 5 to 13th-15th level.

9th level for an OD&D fighter (Traveller expertise level 3) is when he builds his castle and starts playing the Game of Thrones. In this regard, CT is more like contemporary FATE-based games like Spirit of the Century and Diaspora, in which the PC begins the game as good as he will ever be as far as skills go, and improves in other ways – power, wealth, influence.


Conan is about 5th level ("he had already taken wounds which would have killed any four normal men"), so in CT terms would have Sword-1. Given his Strength of 12 (at least), he’d get another +1. Against an unarmoured opponent at optimum range, he hits on a 2; that’s every single time. He’d hit a stock NPC with Sword-1 and Mesh ("the mailed chief of Akif") on a 6, 72% of the time; untrained NPCs (-5 to hit and +3 to be hit) are as grass before his blade, he can carry on hitting them every time, indefinitely, even when fatigued.

A character with Dexterity 8 and a telescopic sight already has +3 to hit at very long range (over 500 metres). He only needs to roll an 8 to hit an unarmoured target, so expertise level 3 or better guarantees a hit every time – there are no automatic failures in CT. If the target is in Cloth armour, that drops to hitting on a 4+, or a mere 92% of the time. Note that with 3D damage and the first shot rule, if he hits the average NPC, they are incapacitated, no ifs, ands or buts. There’s your world class sniper, right there, and interestingly he corresponds to the 7th level D&D fighter I’d use as a template for Olympic-level archers in that game.


So, in the 1977 flavour of Classic Traveller, expertise-1 is Conan. Expertise-3 is more like Jason Bourne or the Batman.

I wish I’d understood that at the time, and been able to convey it to my players through exciting descriptions. Oh well.

What CT ‘77 Got Right

Holidays, babies and exam revision have removed my regular group from play for a few weeks, so today you get another rant instead of the usual Monday evening game report.

I’ve already mentioned that I prefer the subsector generation rules in the 1977 edition of Classic Traveller to anything written since. The same is true for several other areas:


These grew increasingly complex as the rules developed over time; that’s especially true of ship encounters. I have never felt the need to move on from the originals.


The concept of generating critters by ecological niche was brilliant. The rules created beasts which worked the same mechanically whatever they looked like, with appearance and habits assigned by the referee. If he (or she) didn’t feel like doing that, the game lost much of its atmosphere but you could still play.

I still remember the PCs hunting 30-ton pouncers in AFVs. Man, that was nuts.


You have the basic stock designs, and a very simple system for building variants. Initially I spent a lot of time designing ships and house-ruling new ship systems, which was fun, but eventually I came around to the viewpoint one of my players put forward: The ship is only there to transport the PCs to the next scenario. Starship design grew increasingly complex for the next few editions of the rules, and for me at least, it stopped being fun.

As the number of stock designs increased, the ship encounter tables grew more complex too. See comments above.


There aren’t any. Nor do you need them.

At first I felt they were necessary to explain who was handing out thirty million Credit starships to retired scouts of good character, but consider this: A population 8 world with an average per capita GDP of Cr 50,000 (about the same as the contemporary USA), which spends 2% of its GDP on its armed forces (unusually low in the modern world), has a total defence budget of about a hundred billion Credits per annum. If half of that went on starships, and was spent so that an even amount of money (somewhere around 9 billion) was spent on each of the six types of standard starships, that one planet would could buy roughly 250 scout ships per year, and assuming upkeep is 10% of purchase price, the total scout fleet could be over 2,500 Type S for that one planet alone. If you say that one person per annum qualifies for the constructive possession of a ship, and that person just happens to be in the party, it seems plausible. Perhaps originally the President’s daughter wanted a ship, he signed it into law to get her one, and the bureaucracy never got around to repealing that law…

What about all those scout and naval bases? Well, if the planet has the technology and population to operate them – say TL 10+ and Population 7+ for the sake of argument – then they belong to local forces; otherwise, they belong to the nearest planet with those capabilities which already has bases (if it doesn’t have a base itself, it probably doesn’t project power abroad either). It’s easy to envision some sort of subsector-wide agreement for scout services allowing them to refuel at each other’s bases; that actually amplifies the argument for some scouts being spies, as you now need to spy on each other’s scout bases as well.


There were only the vaguest of guidelines for this, but there were also enough random tables that it could be done, hanging a campaign off the spine of interstellar commerce – keep dicing up characters until you get one with a ship, use the others as crew and other NPCs, then take your ship and crew around the subsector trading and dodging pirates.


Of course, it wasn’t perfect; but then, the expectation in the 1970s was that you would use the rules as a starting point, and tinker with them.

There was no point-buy option for character creation. Players, including me, often had a specific character concept which the dice disagreed with. If I want my dreams crushed by random events outside my control, I don’t need to play a game for that, thank you, the real world is more than adequate. To be fair, in 1977 no other RPG really had that option, and you could always house-rule it in.

Combat was clunky; actually, it was OK except for the separate range and armour die modifiers. That was in line with the way D&D theoretically worked at the time, although I never played with a group that used the armour modifiers; but RPGs now have moved towards using range as a modifier on "to hit" rolls and armour absorbing damage. That is much better in my opinion, but again, no RPG in 1977 really did that. Traveller didn’t really catch on until Mongoose Traveller came out in 2008, although obviously the GURPS and Hero versions had that option.

Ship combat should have used range bands like personal combat. Starter Traveller adopted this idea after a few years. That bugs me less now, as space combat doesn’t appear very often in my games.


I really wish I had worked all this stuff out in the ‘70s, you know; my games would have been very different, and cooler. Still, we all had fun, so close enough.

Subsector 1977

No game this weekend, and of all the things I could have done, for some reason I got the urge to generate a Traveller subsector. For me, solo gaming in the 1970s was generating setting information; dungeons, subsectors, NPCs, starships, encounters. Here is a completely random subsector using the original 1977 edition of Classic Traveller, rendered in Hexographer; the point of this exercise is to show how subsector maps have changed over the decades.



Under the 1977 rules, a subsector would look more like a Stars Without Number sector; a bunch of black dots connected by lines. However, even then I was writing information on the map; so I’ve done that, and used later conventions for bases, gas giants etc. How is CT ‘77 different from more modern implementations of Traveller?

Jump Routes

These were created using dice rolls and a table showing the odds of a route depending on the starport types at each end, and the distance between them. It was a royal pain dicing for all the possibilities, which is probably why this rule disappeared almost immediately; but I like the outcome better than the contemporary approach, not least because it lends itself to solo sandbox play; you can generate a viable map using just the starport classes, and fill in the rest of the Universal World Profile as you explore the subsector.

Players could travel to worlds not on these commercial routes, but couldn’t buy passage; they had to use their own ship, or charter one. However, you could make a long Jump over a series of shorter ones; so for example, a Jump-3 ship could move directly from 0101 to 0401 without stopping at the intervening worlds in 0201 or 0302. That makes less sense than the rest of the rules on jump routes, but was intended to make subsector maps more legible.

I always used the jump routes to identify interstellar empires, although that is not in the rules; I reasoned that any group of worlds connected by commercial traffic would be part of the same political unit – completely unrealistic, but it worked for me, and I’ve used it in this map too. In fact, the 1977 rules made no assumptions about multi-world states, other than a brief explanation under noble titles that “The title emperor/empress is used by the ruler of an empire of several worlds.” Empires were small in those days, and presumably determined by the chance presence of government type 6 (“a colony or conquered area”), although I immediately made the leap to a galactic empire of thousands of star systems; judging by the changes in later versions, so did everyone else.

World Statistics

There are some combinations of world statistics that were legal in 1977, but not possible to generate randomly in later rules sets. Notice the world in hex 0302, for example, which has Population 0 (then meaning 0-9 inhabitants) but a feudal technocracy for its government, and quite a high law level; later editions would assume that Population 0 meant government, law, and tech levels of 0.


There are fewer types of bases; naval (five-pointed star), scout (triangle), and Travellers’ Aid Society (not shown on the map, because it is implied by the starport type). Quite enough for me.

Trade Classifications

There are only six trade classifications; agricultural, non-agricultural, industrial, non-industrial, rich and poor. Later editions have added many more, which I find unnecessary. In particular, the various versions of “T-Prime” or “Garden” worlds essentially duplicate the Rich classification.

Travel Zones

These didn’t appear until the early issues of the Journal of the Travellers’ Aid Society, and weren’t in the rules originally.


What’s most striking about this exercise is how little Traveller’s world generation rules have changed since 1977. I could use this map to run a game under any incarnation of Traveller, from Classic to Mongoose, without doing anything else to it. I suppose I might need to do a little conversion for GURPS Traveller, but that’s about it.

Maybe I’m just a gaming dinosaur, but the first generation of RPGs nailed quite a few things that we’ve drifted away from over the years with more elegant and complex rules. In fact, I wish the Far Future CT CD had the 1977 rulebook on it as well, as there are some bits I find superior to all later attempts.

Watch for future posts on other Stuff Classic Traveller Got Right.

The Third Terran Imperium

“Earth itself became a garden planet, bearing only one city worth noticing, the sleepy capitol of a galaxy. Pittsburgh valley bloomed, and rich honeymooners went there to frolic. Old bureaucrats went to Earth to die. Nobody else went there at all.” – James Blish; Earthman, Come Home.

Here’s an idea for a variant Classic Traveller campaign.

Use Supplement 3 (The Spinward Marches), but declare Porozlo to be Earth – it has the right physical characteristics – and spiritual home of the Third Terran Imperium. The real seat of power is Mars (Rhylanor – again it has the right stats), which for astrogation purposes is one jump away, so represented as being in a different hex, even though it is in the same star system as Earth.

The Rhylanor subsector doesn’t look much like local space, so one also has to declare either a change in scale (perhaps one hex is 5 or even 10 parsecs), or say that the map of hyperspace is different from the map of real space.

The rest of the Spinward Marches and most other support materials can be used as is, but the Imperium is likely to be much smaller, perhaps only 3-4 sectors in size, and younger. The reduced age can be handled by dropping the first two Imperia and any reference to the Vilani, making the age of the Imperium (let’s say 1105 years) as far back as human interstellar travel goes.

Traveller Zombies

“Can you imagine what this novel might be like without the violent zombie mayhem?” – Seth Grahame-Smith, Pride & Prejudice & Zombies.

Isn’t every game better with zombies?

(Don’t give me grief about plausibility and hard SF. Go read Larry Niven’s “Night on Mispec Moor”. Back already? Good.)

Like most things, zombies are easy to add to an Old School RPG. Start with 100kg eaters. They come in packs of 2d. They will always attack unless the PCs surprise them. In combat, they will always attempt to close range, so that they can use their hands to drag their victims to the ground and eat them, but cannot run, so only close one range band per turn. If the PCs shoot them, each 6 rolled on a hit or damage die means another zombie shambles to the attack. If you want to run them as NPCs rather than animal encounters, zombies ignore the first wound rule. Any hit from a zombie which causes damage transmits a disease which inflicts 2d wounds per day until the victim dies; about a day later, he rises from the dead as a new zombie. (There is rumoured to be a cure, of course, but the PCs have to find it and apply it. Good luck with that.)

These are the classic Romero zombies. If you want the more modern view of sprinters rather than shamblers, allow them to run, otherwise don’t.

Animal Weight Hits Armour Wounds & Weapons Att/Flee/ Speed
7 Zombies 100 kg 18 / 7 Jack 4 Hands A1 F9 S1

There we go; job done.