How did character generation work in 3rd edition?

Question

Years ago, as a teenager, I played in a 3rd edition Stormbringer campaign. It's a game based on the Elric novels of Michael Moorcock.

I remember distinctly that the book had a big table of the races/cultures of Elric's world. The GM had us roll on this table to start our characters. I lucked out and rolled the 1% chance of a Melnibonian, the same as Elric himself.

As far as I recall, everything else in character generation hinged off race/culture. For certain my character was vastly overpowered, able to easily defeat entire groups of foes by himself that the rest of the party would struggle with, and possessing a wealth of other skills that rendered them all obsolete.

After a few sessions of dominating absolutely everything, I volunteered to retire the character on condition I could still play a sorcerer. And after that we had much more fun. It was obvious, even to us as teens, that rolling for such an imbalanced aspect of the game was a daft design decision.

Anyway, getting such a lucky roll and it having such a huge impact I makes me wonder: were we really supposed to roll on that big table? Was character generation really that potentially flawed, in spite of play testing? Or did the GM miss some rules somewhere?

Answer 1

In at least the first through third editions of Stormbringer the rules really do indicate that unless the GM says otherwise a PC's nationality is randomly rolled, and this absolutely could result in one PC being a Melnibonéan sorcerer and another PC being a Nadsokor beggar. (This potential disparity between PCs is mentioned in examinations of the game here and here.)

…But is this flawed?

By contemporary standards, this may seem deeply flawed. However, many of gaming's earliest experiments—and Stormbringer is certainly among those, its 3rd edition (1987) largely unchanged from its 1st (1981)—weren't particularly concerned with balancing the PCs against each other.

For example, Villains & Vigilantes (1979) and Traveller (1977) both use, in part or in whole, random character generation that can lead to one PC being significantly more powerful than another. In V&V, this involves rolling which superpowers a PC receives. Like its TV Tropes entry says

The "official" way to roll up a character for the game was to pick a table of powers (like general powers, magic/psionic powers, magic/psionic items, and skills), and then roll percentile dice to determine what powers your character had. This could result in some strange combinations, which can also be seen in some of the characters, like Mace from Crisis at Crusader Citadel [and here]. As his name implies he wields a tricked out mace, but for some reason also has a superhuman sense of smell despite having a background that fails to suggest he's anything more than an everyday thug.

(Link added.) A particular V&V example is the superpower Heightened Endurance that can add 3d10 to a PC's Endurance score. Assuming a starting score of 11 (on a scale of 3 to 18), a lucky 3d10 roll sees a PC with an Endurance score of 40, and a massive increase in the PC's hit points. A typical PC without heightened ability scores has between 3 and 12 hit points; a PC with Heightened Endurance could have well over 100! Similarly, in classic Traveller (a character generator for which can be found here), a PC could have sufficient starting funds to skip adventuring altogether and go straight to retirement, maybe giving the title of his spaceship to the less fortunate PCs before he departs for a life of leisure.

Even Advanced Dungeons & Dragons is no stranger to this: Its Player's Handbook (1978) includes the so-totally-optional psionics rules. Determining if a PC had psionics is a bit of a trial:

Characters with one or more unmodified intelligence, wisdom or charisma ability scores of 16 or higher might have psionic ability. Whether or not this ability is possessed is then determined by a dice roll using percentile dice. Any score of 00 (100%) indicates the ability exists. For each 1 point of intelligence above 16 add 2½% to the dice roll, for each 1 point of wisdom above 16 add 1½% to the dice roll, and for each 1 point of charisma above 16 add ½% to the dice roll (drop all fractions). (110)

Then if a lucky PC possesses psionics, the player rolls to determine his PC's number of attack modes, defense modes, and minor and major disciplines. Then the player rolls again to determine which minor and major disciplines his PC possesses. In the end, a fortunate level 1 illusionist PC could end up able to employ the equivalent of a 9th-level spell, while an unlucky level 1 fighter PC could end up being a fighter.

My point is this: Stormbringer is far from unique in having mechanics that may yield a PC who is significantly more or less powerful in the abstract than another PC. And,—both because I am no apologist for this process and because I lived through this era while being infamous in my circle for my utter lack of dice luck—, I can say decisively that Yes, this kind of sucks. However, ultimately,—the theory goes—it shouldn't matter if PCs are not balanced against each other. Let me explain.

It's potentially unbalanced, but it's also fair

So long as no one's using loaded dice or bribing the GM to get special treatment or whatever, creating a character randomly is completely fair: every player has the same chance to have an awesome PC. While it seems wrong that you are playing a scurvy-ridden penniless Nadsokor beggar while she is playing a noble Melnibonean who can summon the gods of chaos, you and she could've just as easily rolled differently, and your PC would've been the sorcerer and hers the beggar… or you could've been both sorcerers or both beggars!

Once PCs hit the table, though, a role-playing game of this sort especially is cooperative, and it's cooperative on a much larger scale than the character sheet in front of the player may indicate. A player isn't bringing only his or her character to the party. A player is bringing to the party a whole 'nother brain that adds to the group a further layer of experience and knowledge. Also, the player's bringing to the table—in the form of his PC—a whole 'nother set of actions that couldn't be taken were the PC absent. Those—let's face it—priceless additions to the gaming group are present no matter how terrible the PC! It's up to the player to put his own brain and his PC's actions to good use no matter how execrable the PC.

Also, keep in mind that the real challenge isn't even on your side of the screen. Your PC is randomly generated: you had no choice! Had the GM wanted PCs balanced against each other, he wouldn't've had players make them randomly! That makes it the GM's responsibility to build a campaign that involves equally all the PCs no matter how absurdly different. Your responsibility as a player is to trust that the GM is doing precisely that and to use your knowledge, experience, actions, and character sheet to play your PC to the best of your ability even if that character sheet alone sucks.

That's one of the great appeals of random character generation: being able to tell stories afterward about how your awful, worthless PC survived and thrived to become the Vagrant Monarch of Nadsokor… or how, after failing his first pick pockets roll, he died to one blow of a city guardsman's club.

---

Note: While further examples abound, my favorite OSR version of random character generation is The Mutant Epoch (2012)—a sort of super-advanced Gamma World (circa 1983's 2nd edition) that pulls no punches in its random character generation—expect the crazy. Similarly, no discussion of unbalanced PCs can be had without mentioning Rifts as it is famous for ignoring the idea balancing PCs against each other, this interview with the game's creator even explaining why. A Rifts party can consist of Dr. Strange, Professor X, Smaug, and a hobo with a shotgun and a pocket full of candy. However, by contrast, Rifts typically leaves it to the players to make unbalanced choices rather than having entirely characters generated randomly. (Also see this question and this question.)

Answer 2

I played the game too.

You didn't necessarily have to roll on the table. It was optional. Your GM might have seen that, or might not have.

Yes, the game really could be seen as 'flawed' when viewed in relation to modern game standards. It was from the first wave of RPGs after D&D. The hobby was still young, rule mechanics were all over the place, and balance wasn't as important of a consideration.

I think that amount of randomness was intended as the kind of 'hard mode', where a player could see the imbalance as a personal challenge to be overcome by wits and skill. Very, very few players I've ever known could pull that off, so there's a reason that kind of character generation is rare these days.

Still, even having played dozens of RPGs in that era, the range of variation in that generation system was shocking - you had an equal chance of getting an Uber-powered Melinibonean or a blind beggar from Nadsokor. But, in it's own way, it was a faithful representation of the books; Elric adventured with a wide variety of characters, every one of them was weaker than him, and they all died horribly.

Answer 3

TL;DR: Rolling for nationality was probably an optional rule.

I don't have a copy of the 3rd Edition rules, but I do have copies of other editions.

In both the 1st and 4th Edition rules, rolling for nationality is an optional rule.

To determine your character's nationality randomly, consult table [2.2.1] Character Nationality. If your campaign calls on you to play a Melnibonean, or a barbarian from the Weeping Waste, or anything else, then make your character a member of the appropriate nationality without consulting the table.

pg. 23 (in both books)

The table [2.2.1] Character Nationality is also the same for both editions.

Since the rules for rolling for nationality are the same in 1st and 4th Editions and they are optional rules, it seems likely that the rules existed in the same form in 3rd Edition. My guess is having people roll for their nationality was seen as a way of 'balancing' the otherwise unbalanced nationalities.