Idraluna Archives

Holme Alone

I recently ran across a neat thread from 2008 on the Dragonsfoot forums where Geoffrey McKinney of Carcosa & Mike's Dungeon/World fame pitched the idea of treating the Holmes Basic D&D set as a complete game (henceforth referred to as "Holme Alone").

For the unfamiliar, the Holmes Basic D&D set was released in 1977 and is kind of the black sheep of TSR Dungeons & Dragons (certainly overshadowed in the OSR by the Tom Moldvay's basic set, but not without passionately devoted fans). John Eric Holmes was a neurologist(!) & fantasy author who apparently reached out to Gary Gygax offering to edit the (let's face it) convoluted mess that is the LBBs & early supplements:

"...I persuaded Gygax that the original D&D rules needed revision and that I was the person to rewrite them. He readily conceded that there was a need for a beginners' book and "if you want to try it, go ahead". I went through the original three rule books and the first two supplements, Blackmoor and Greyhawk, of which Greyhawk is the greatest help. Trying to use the original words of the two game creators as much as possible, I edited a slim (48 page) handbook for beginners in role playing, published by TSR in 1977 as Dungeons and Dragons and usually marketed as "the basic set".”

The resulting document closely resembles OD&D but with generally more coherent rules,1 and only covers character levels 1-3. As I understand it, it was intended as an introduction to AD&D, rather than as a precursor to the separate Basic D&D product line beginning with Moldvay in 1981.

Basic Rules_Page_27
(Illustration by Tom Wham)

McKinney's pitch in the Dragonsfoot thread is that if one were to ignore the exhortations to adopt AD&D rules after 3rd level, Holmes Basic is surprisingly complete and workable as a standalone D&D edition, all in a mere 48 pages. I encourage going through the thread, but the main points are:

And as a lover of systems with some assembly required, I'm intrigued by how the lacunae in Holmes-only are orthogonal to those in OD&D. The LBBs provide scaffolding for advancement past level 10 and beyond, but they lack a complete combat system, details for many monsters, etc. Holmes, on the other hand, has more complete and consistent rules, but just stops after level 3.

Diegetic Addons

Although the core of McKinney's proposal is that Holmes works remarkably well sans house rules, I'm drawn to the idea of grafting diegetic advancement house rules on to the Holme Alone chassis. By 'diegetic advancement', I have in mind games like Cairn, Wolves Upon The Coast, or Goblin Guts v2, wherein the ways characters advance are closely tied to specific actions in the fiction, rather than arbitrary XP totals.2

My specific take is that a 3-level Holmes-inspired system paired with light house rules & good sandbox design melds the advantages of both conventional D&D leveling & diegetic games. I like diegetic advancement because I like intense engagement with the imagined world of the game, but I find that it requires a lot of prep (hand-placing spells in the scenario, inventing strange powers, fighting styles, etc.), is tough to balance (I'm not enough of a purist to discount balance entirely), and can divide parties if different character types have divergent advancement requirements. But by blending a truncated level system with diegetic advancement, players get two levels of 'regular' gold/kills-for-xp advancement to familiarize themselves the setting, and after that they have to fully engage with it to become more powerful.

McKinney's idea of using magic items to grant access to higher-level spells is already diegetic advancement, but one could also write up rituals for selected 3rd-6th spells a la Wolves Upon The Coast. I lack the time & inclination to draft setting-specific requirements for petty cantraps (and do I really want to devote precious session time to role-playing the preparation of a Light spell?), but writing a curated list of high-level spells into the setting is less of a lift.

This also meshes well with the fact that Holmes Basic, following OD&D, hands out attack matrix & saving throw improvements every three levels (i.e. not at all!). Goblin Guts v2 has players record "reports of peril" when they nearly die, granting +4 to future saving throws against the threat, or +1 HP.

Fighters can always improve their hit rate with magic weapons, but one could also sprinkle selections from the myriad 'here's how to improve fighters' house rules around as learnable 'fighting styles'. (E.g. fray dice, maneuvers, cleave, WuTC weapon rules).

I like the idea of letting thieves pay a master thief for skill training. If any class should have a strong incentive to be greedy, it's them.

Finally (having hexcrawls on the brain), I like the idea of clerics earning higher-level spells by making pilgrimages (reason to travel) and performing sacrifices (money sink).

Implications for Sandbox Design

From the above, one would want to incorporate magic items (duh), fighting styles, master thieves, and shrines/pilgrimage sites into this hypothetical Holme Alone sandbox. Pretty straightforward; what I like is that these things acquire relevance less immediately than in a campaign with only diegetic advancement -- one can hit the ground running with just a dungeon & the advancement rules in the book, adding in opportunities for high-level training & spells later.

The other intriguing wrinkle is that in McKninney's proposal, NPCs are capped at level 3. Here's his take on converting from published AD&D settings:

What is the enterprising DM to do who wants 3rd-level to be the highest level attainable, and wants to use a commercially available setting (such as Greyhawk, the Wilderlands, etc.)? What should he do about those 15th-level NPCs running around?

Well, the DM would convert any sub-class to its base class. An illusionist, for example, would become a magic-user. Then the DM would convert levels as follows:

1st-7th level NPCs in the product would become 1st level. 8th-14th level NPCs in the product would become 2nd level. 15th+ level NPCs in the product would become 3rd level.

Hence, a 17th-level paladin would be converted into a lawful good 3rd-level fighter.

This is a pretty wild reduction in power, especially since monsters will presumably be unchanged -- I don't know if I'd use the same cutoffs, or bother trying to convert such a level-inflated module.

Edit 9/14/25: My preferred alternative would be to roll all the NPC's HD and then take the three highest results. This would roughly preserve the intended relative power, but with diminishing returns.

But, I love the simplicity & groundedness of restricting NPCs to three levels only. I'm currently writing a fairly large OD&D sandbox for the Antarctic Adventure Jam and am finding it tricky to navigate the way OD&D treats level as both social status and personal combat prowess. For all the janky charm of making every Lord a 10th level Fighter, the constrained power levels in Holmes-only seem easier to work with from a fiction-forward, 'new simulationist' approach to setting design.3 If an NPC is powerful, it's more interesting to break that down in terms of their relationships, arcane knowledge, wealth, etc. than to slap a level label on them and move on.4

Etc.

Holme Alone seems quite close to type of D&D I most want to run & play: grounded, without superheroic characters, mechanically close to OD&D, concisely presented, and able to mix diegetic & level-based advancement. It's certainly nothing groundbreaking, but I'm less interested in novelty than in refinement.

I'm tempted to build my Antarctica Jam entry around this idea (the Blueholme Prentice rules are free), firstly because it's a framework for writing some diegetic MU & Cleric spells into the sandbox without worrying about compiling a full roster, and secondly to cut down on the rampant level inflation in all the OD&D castles I generated. I'm going to sit with it for a while but may take a stab at formally writing up these house rules in a future post.

Edit: A couple additional relevant posts that came up discussing this on the OSR discord:

  1. A notable exception: RAW, all weapons do 1d6 damage, but daggers can attack twice/round, one-handed weapons once/round, and two-handed weapons once/two rounds. Being accustomed to OD&D's mechanically identical weapons, it seems easy enough to simply keep the 1d6 damage and ignore the attack speeds.↩

  2. "Diegetic Advancement" is handled differently in all three systems; Cairn is probably the the most purist about tying advancement specifically to things that happen in-fiction, thought the scar system has always felt a bit arbitrary to me. Maybe "diegetic" is misleading; it's all really a spectrum of abstraction and elision (even gold-for-xp is technically tied to in-fiction activity).↩

  3. The thread has many amusingly grognard-core arguments about whether the three levels in Holmes can represent Conan, Gandalf, etc. I personally don't care that much, but I do agree with McKinney that actually, yes, three levels are adequate.↩

  4. Of course, this is a D&D specific problem -- worth acknowledging that if you're designing a setting for, say, Cairn, you're likely doing this already.↩

#game-design #musings #odnd #slush-pile