Of The Black Company

Setting Up The Black Company RPG: What Actually Works

I spent way too long trying to get The Black Company roleplaying system running on my first go. The published materials assume you already know how the Crooked Man's house rules work under the hood, which is fair since Glen Cook wrote the source material decades ago before any official RPG existed. But if you're starting from zero, here's what I figured out through trial and error. The core issue is that The Black Company as a tabletop system isn't one clean product. There's the original Black Company RPG from Chaosium back in 1995, then there's the later Black Company d20 System from Mongoose Publishing around 2006, and people on forums keep confusing the two. They use completely different resolution mechanics. If you download the wrong one expecting the other, you'll be frustrated fast.

Getting Your Hands on The Black Company

For the original 1995 edition, you're looking at out-of-print territory. The PDF circulates on various RPG file-sharing sites, but I'd recommend checking DriveThruRPG or the Open Game License archives first. The Mongoose d20 version is more likely to show up on legal retail sites. Either way, make sure the edition matches what your group expects before anyone buys in. Here's the thing nobody mentions: the game runs fine without either official sourcebook if you read Cook's novel trilogy first. The Crimson Guard, The Brothers of the Skull, and Return to the Pass are basically the campaign modules. I learned this when my group couldn't afford the PDFs and we ended up running a six-session campaign using nothing but the novels as reference. It worked surprisingly well.

How the Resolution Mechanics Actually Feel

The d20 version uses standard D&D 3.x style rolls, which means most checks are straightforward attack rolls against AC or skill checks against DC. The original Chaosium edition uses percentile dice with opposed rolls, which feels more medieval and gritty. Both work, but they create very different pacing. With the percentile system, a typical skill check might read like this: the agent tries to pick a lock, rolls under their Locksmith skill of 45, and the guard's Perception of 60 is rolled opposition-style. The tension comes from the fact that failure doesn't just mean "nothing happens" — it means something worse. In my experience running this system, about one in six attempts at a routine task triggers a compounding failure that the GM can escalate. That's intentional and it's what makes the game feel like Cook's novels. The d20 version handles this differently. Critical failures happen on natural ones, and the narrative escalation is more mechanical. You get the failure condition spelled out in the rules rather than left to GM discretion. Some people prefer this clarity. I find it takes some of the dread out of rolling, which is simultaneously better for player comfort and worse for atmosphere.

Character Creation: What the Books Don't Emphasize

Building a Credible Agent takes longer than building a regular soldier. The published rules give you templates, but the real work is in deciding what kind of moral compromise your character has already made. In practice, every successful campaign I've run had characters whose backgrounds implied specific loyalties and specific betrayals before session one. The spells and magics section is where most groups hit problems. The Black Company setting treats sorcery as dangerous and corrupting, but the rulebooks sometimes imply it's just another power option. Don't make that mistake at your table. Magic in this setting costs something — usually relationships, sanity, or physical deterioration. I once ran a wizard character who could cast decent hexes but kept losing memories of his own family. The players accepted it because the book supported that reading, even though the stats didn't explicitly track memory loss.

Running a Session: Common Pitfalls

The biggest mistake I see new GMs make is treating the Setting Company as a standard mercenary contract simulator. It isn't. The political maneuvering between the Lady, the Wizard, and various external powers matters more than any sword fight. If your players spend every session clearing dungeons, you're playing a different game than what the book supports. Another issue: the reputation and information systems interact in ways that aren't obvious from the rules alone. In the novels, knowing too much gets people killed. The game mechanics support this through the Knowledge and Gossip skills, but new GMs tend to let players accumulate information freely. I started tracking a "Dangerous Knowledge" counter alongside regular skill checks. When a player rolled high enough to learn a secret, I'd mark it on a hidden sheet and let them know their character now had a target on their back. It usually took three to four sessions before the consequences caught up, which felt right.

Adapting Scenes from the Novels

When I ran the Scylla campaign, I pulled the structure directly from Cook's second novel. The party starts contracted, discovers they're being used as pawns, and has to decide whether to continue serving or find another path. The published adventure paths don't cover this arc well, so I built my own using the core mechanics. It took about two hours of prep work to translate the novel's plot beats into encounter tables and skill challenge sequences, but the players said it felt more like the book than any official module ever had. The main technical challenge is balancing the magic system. The Lady's power level in the novels is essentially unmatched, but the RPG needs NPCs that are beatable. I solved this by making high-level magic extremely rare and costly in-game. Only one character in a group of five ever attempted serious spellcasting, and they required specific ritual components that took real time to gather. This kept the power fantasy contained without breaking the setting's internal logic. If you want to run The Black Company setting without using either official rulebook, the OSRIC or Basic Fantasy systems both handle the low-magic, gritty tone adequately. I've used Basic Fantasy for a single one-shot and it took about twenty minutes to convert a novel scene into playable encounters. The percentile roll-under mechanic is close enough that players didn't notice the difference.