Suloise Tomb Architecture

Outside two ruined tomb structures

Today I'm happy to feature a guest post from our friend Angela Black, who wrote to me with what I thought was a clever idea for establishing standard architectural/layout detail for Suloise tombs in the World of Greyhawk. A stylistic notion like this is both (a) realistic to the world as we know it, and (b) jumps-starts the design for possibly many such tombs scattered throughout the campaign world, without needing to come up with a brand new design concept for every one of them. I'll let Angela explain in her own words:


In my campaign world, loosely based on Greyhawk, the Suel fill the role of the Roman Empire in medieval Europe - they used to own all this land, but they're gone now, and they've left behind interesting ruins and whatnot. Specifically, I stipulate that the Suel buried important dead people underground in elaborate tomb complexes filled with offerings to glorify the dead, so as to have a handy excuse for "dungeons" more or less where-ever I need them to be (in addition, of course, to the usual adventure-sites, which may or may not be connected to the Suel, like abandoned fortresses and ruined temples and so on).

It occurred to me, however, that such places in the real world tend to have common, identifiable characteristics - that is, buildings intended by a particular culture to serve a particular purpose usually share common features and maybe even a standard layout. And in a flash I realized how useful it would be to have a general notion of these features when it came time to design a dungeon. What IS the "typical" Suel tomb like? I don't have to pick up a blank piece of graph paper and stare at it until drops of blood form on my forehead - I can approach the same way the (fictional) people who (fictionally) built it would have approached it, with a "pre-loaded" set of ideas about what it must include and how those spaces would be connected to each other.

So, as an exercise in world-building and as a tool for future use as a DM, I wrote up a little archaeological-style report on the "Standard Features of a Suel Tomb." I thought you might find it interesting as an example of how a DM might approach such a thing. It could serve not only as a rubric for one's own use, but also, in part or in whole, as lore to give to players whose characters might actually know this kind of stuff. Of course, it could also be fun to watch the players put together over the course of a campaign that there really are certain standard, predictable features to these things!

Outer Tomb

Devotional Sepulcher

A round room, underground but open to the air without a door, that contains a sarcophagus, which itself contains a symbolic representation of the deceased. This symbolic representation may be anything from an item of clothing to the body of one of the deceased’s servants – there is no general rule. The walls of the Devotional Sepulcher are decorated to illustrate the life of the deceased; If the tomb’s occupant was especially important, there might be one or even two Siderooms attached to the Devotional Sepulcher (see below).

The Devotional Sepulcher is where people can come to pay devotion to the deceased, so there are typically small offerings, some of which might be valuable, in this room. Generally, the traffic of admirers to the tomb will keep it from becoming a nesting place for any animals except inconsequential vermin, but if the tomb is forgotten or in a distant location, larger and more dangerous beasts may have taken up residence.


If the deceased was particularly important, additional rooms might be constructed in the outer tombs. These are always smaller, rectangular rooms opening off from the Devotional Sepulcher without doors, decorated with murals or even statuary that reference particularly notable events of the deceased’s life. A Sideroom might likewise have some offerings, as noted under the description of the Devotional Sepulcher.

Inner Tomb

The Outer Tomb exists only to satisfy the need to revere the dead – the dead themselves are hidden in the Inner Tomb. The Suel, however, felt it very important the honored dead receive their due, and the Inner Tomb must always be connected to the Outer Tomb, so that the offerings made in the Outer Tomb are properly directed.

As such, behind one of the walls in the Devotional Sepulcher (or, very rarely, one of the Siderooms) will be the Entry Hall to the Inner Tomb. This Inner Tomb is not intended to be entered by anyone, so the masonry will be solid, but that does not mean it is impassable – tomb robbers are not often deterred by a little work with a pickaxe. The masonry is invariably painted to look like the rest of the wall in the Devotional Sepulcher, but a sharp eye can determine where to start working.

In remote tombs or ones difficult to reach, the masonry wall has likely already been breached, but entry into the Inner Tomb is only the first obstacle, as shall be seen.

Entry Hall

The Suel, always very proper, always build an Entry Hall into the Inner Tomb, even though it was never made to be entered. The Entry Hall is constructed in typical Suel fashion, a wide rectangular space with a high ceiling, decorated on the floor, ceiling, and wall so as to indicate the owner of the abode – in this case, the deceased.

The Entry Hall is also the first line of defense against would-be grave robbers. This typically takes the form of a magic mouth warning intruders to turn back, as well as perhaps some kind of trap, such as a spiked pit or even a glyph of warding. If the deceased was extremely important, there may even be a permanent wall of fire blocking further progress. Undead are never intentionally placed in the Entry Hall, however.

There is always one and only one exit to the entry hall, a portal with no door opposite to the ingress. This invariably opens to a set of stairs going down, which leads to the Hallways. The stairs are a crucial feature, as they symbolize the descent into the underworld, and they are, owing to their ritual function, never trapped.


After taking the stairs down from the Entry Hall, would-be robbers enter the Hallways. There must be enough rooms in the Inner Tomb to properly honor the deceased, and these are always connected to a series of winding halls – the Hallways, which provide access to the Devotional Rooms and the Shrine.

However, the Hallways are also part of the Inner Tomb’s defenses against intruders. They are intended to be confusing, decorated with repetitive designs and wind about in no discernible pattern, criss-crossing and sometimes leading to dead-ends. They are also commonly littered with traps of the usual kind – spiked pits, poison darts from the walls, and so forth; for this reason, the Hallways always use a tiled floor, the better to conceal the trigger mechanisms for traps. The Hallways also typically have at least a few secret doors, concealed by the architecture and the patterns of the paintings on the walls, which lead to other sections of the Hallways. The Hallways sometimes incorporate stairs that go up or down, but as these have no ritual purpose, they may be trapped, unlike the stairs descending from the Entry Hall.

Obviously, after so long, some of the traps in the Hallways will be non- or only partially-functional, and some that may have been triggered by tomb robbers will have failed to reset. Likewise, if the tomb has been opened and is remote enough, wild beasts may have made lairs in the Hallways.

Devotional Rooms

In the Inner Tomb, there are always at least two Devotional Rooms – one to honor the deceased and one to honor the deceased’s family. However, there are also rooms to commemorate significant events in the deceased’s life, so for anyone important enough to have a full tomb, there will be at least a few extra Devotional Rooms.

The Devotional Room for the deceased will be treated separately and is more properly known as the True Tomb (see below). The other Devotional Rooms are always alike – large, high-ceilinged square rooms, filled with devotional treasures and featuring not only decorative murals that depict the event or the deceased’s family, as appropriate, but also a stela, in the center of the room, which details the event or describe the family. It may or may not have other exits, concealed or obvious, back to the Hallways, but it never opens onto another Devotional Room. It is possible, however, that access to other sections of the Hallways can only be gained by passing through a Devotional Room.

The treasures are simply meant as offerings to the deceased and will be of value proportional to the glory and fame of the deceased. Sometimes, there might be an item relevant to the event, whether a piece of art, a weapon, or something else. For the family’s Devotional Room, however, there are always a few items that commemorate the family, usually things that were precious to them. There may be statuary depicting the family in the family’s Devotional Room, as well, but this is not consistent. If such statuary is present, it is ornamental and serves no ritual function, and so may well be trapped or even enchanted to animate and attack intruders (though such enchantments are rare and expensive).

What is consistent, however, is that the Devotional Rooms are well-protected. The most basic method is the use of skeletons who will mindlessly attack any living creature that enters the room, but for particularly important people, a few mummies might be found, as well. Such rooms are also sometimes protected by common traps or magical deterrents like glyphs of warding.

The Shrine

Somewhere in the Inner Tomb is a shrine to the god of the dead. This shrine is almost never connected to the Hallways in an obvious way – it may be connected to the Hallways by a secret door or perhaps even connected to one of the Devotional Rooms, either in an obvious way or by secret door. The Shrine is always protected by a glyph of warding at minimum and will usually have mechanical traps and undead guardians as well. It always contains a statute of the god of the dead as well as tablets or stelae with devotional prayers affirming the supreme status of the god of the dead and so on. Treasure is not usually found in this room, but there are rare exceptions – it has been recorded on at least a few occasions that the priests of the god of the dead have stored valuable objects d’art or even magic items in The Shrine. It is unclear why: sometimes the objects are situated in the open but sometimes they are well-concealed and protected. In any case, the Shrine is not intended to be entered, even in a metaphorical way; where the Shrine is connected to the rest of the Inner Tomb with an obvious door, the door is sealed, and where connected with a secret or concealed door, that door is always trapped. The usual caveats apply, of course – if the tomb has previously been breached, and the Shrine discovered already, it’s possible any traps will have been triggered and perhaps not reset, and likewise possible that some animal or monster will have taken up residence therein.

True Tomb

The True Tomb is the genuine resting place of the deceased. It is much like other devotional rooms except larger and more grandly decorated, with murals and/or statuary depicting the deceased in a variety of glorious poses. It will have a stela standing against the wall opposite the entrance that praises the deceased, and in the center of the room will be the genuine sarcophagus. Devotional treasures of the highest value are piled on tables against the walls or in the corners. In some cases, the True Tomb will have a mezzanine level, the better to provide space for depicting the glory of the occupant, but there are never stairs to this level (evidence from unfinished tombs indicates works would use ladders to complete and stock the mezzanine level, which would then be removed along with other tools and materials).

There is always only one entrance to the True Tomb, and it is intentionally very difficult to access, usually behind several series of secret doors in the Hallways. Aside from (possibly) the Shrine, the True Tomb alone in the whole Inner Tomb has a door, and it will certainly be trapped and/or warded. Within, there will be deterrents more deadly than anywhere else in the tomb (except, again possibly, the Shrine) –a contingent of skeletons commanded by a skeleton warrior, a small group of mummies, a clay golem, or perhaps even a demon that has been summoned and bound into service as a guardian. Of course, “more deadly than anywhere else in the tomb” is relative to the status of the deceased; at least a few Suel tombs have been recorded which were, shall we say, “aspirational” on the part of the occupant, with most available funds being spent on the mere construction of the tomb, leaving little for defenses.

The genuine sarcophagus is always trapped and/or warded as well, though with good reason – not only does it contain the remains of the deceased, it also contains any significant items associated with them, such as special weapons. Or rather, it contains such items in theory – in far more cases than the Suel would have been likely to admit, highly desirable weapons and other powerful items were replaced with copies which were buried with the deceased so that the items could be secretly passed to relatives. However, even the most grasping relations would not be so bold as to replace a powerful item with a mere bit of brass; when such substitutions were made, a less powerful but still genuinely magical item was always used. It was not considered wise in Suel culture to tax too greatly the patience of the god of the dead.

Open Questions

Dan back here again -- I thought that was a really interesting piece of fantasy architectural digest. Thanks so much to Angela! And a follow-up idea that I immediately had: How hard would it be to code up an online generator for that particular "style" of dungeon (maybe with selections for small, medium, or large-sized Suel tomb)? Anyone else have an itch to make that happen?


The Big Mistake in Weapon vs. Armor Adjustments

Players of 1E AD&D duel with the most heavy-weight table in the PHB (p. 38): the "Weapon Types, General Data, and 'To Hit' Adjustment" table, which includes Armor Class Adjustments, intended to recreate the matchups of certain weapons versus certain classes of armor. 

AD&D 1E PHB Weapon TYpes Table

It's among the more complicated things in the game. On the one hand, they're not listed as variants or optional rules; and they're ingrained to the DMG example of combat -- so many 1E players do pound these tables into their games, determined to faithfully use them no matter how awkward they are. On the other hand, at least as many players of the game overlooked them, and Gary Gygax is even on record as saying the same thing. Let's document a few quotes:

I did not use psionics, generally ignored weapons vs. armor type and weapon speed. – Gary Gygax, ENWorld Q&A, 24th January, 2003

There is often player pressure to add complexities and complications to rules and systems, such additions being urged in areas that the players like and believe to be critical to enjoyment of the game. I did that for some writing in OAD&D and regretted it considerably thereafter – mainly weapons vs. armor types and psionics. – Gary Gygax, ENWorld Q&A, 24th July, 2003

In all, I included the details because of insistance of some avid palyers that were in touch with me, regretted listening to them, for the RPG is not suited to combat simulation... As I noted above, we never used the weapons vs. armor type adjustments. – Gary Gygax, ENWorld Q&A, 7th September, 2005

He elsewhere says the same thing for the Space Required value given to each weapon; that is, Gygax wants to entirely ignore everything on the weapons table taking up all of AD&D PHB p. 38. (And his son Luke Gygax reiterated the same thesis when we interviewed on him on the Wandering DMs, the day after I wrote this.) But if that's the case, it begs the question, where exactly did these numbers come from? Were they ever playtested? The questioner on ENWorld Sep-7 2005 asks this question, but Gary lightly dodges it.

As usual, the answer is given more clearly if we take a step back to the Original D&D texts. Essentially these same modifiers first show up in an earlier table in OD&D Supplement-I, Greyhawk, p. 13-14. (I'll show a recreation of that chart a bit later.) Right before this table Gary writes the source for these numbers: 

For those who wish to include weapon types in the determination of hit probabilities the following matrix drawn from the “Hand-To-Hand Combat” section of CHAINMAIL is offered. If this system is used it is suggested that the separate damage by weapon type and monster type also he employed.

Okay, so we need to take one more step back to Chainmail and to the actual origin of these numbers. Here's the earliest expression of that work to which I have access, from Chainmail 2nd Edition (p. 37):


Notice the presentation here is different; the numbers in Chainmail aren't modifiers to D&D-style attack rolls; they're final target numbers for the attack. That is, this system effectively assumes that all combatants are the same level and strength ("normal men", you might say), with the basic factor in this system being only the weapon and armor employed on each side. (They're also target numbers on a 2d6 roll, instead of D&D's d20 roll, but that's not the most salient thing.) So how do those different systems correlate? Note that Gygax is a bit cagey on this point -- he says, "the following matrix drawn from the 'Hand-To-Hand Combat' section of CHAINMAIL", but he doesn't say exactly how it was drawn. And here's where the math comes in, which allows us to pierce the veil, and see exactly what he did.

I've tried a few things, and I won't present all the failed investigative paths on my part, but here's the recreation that works the best. Gary first takes the average of all the target numbers in the Chainmail table; we'll call this the "base to hit", and it works out to about 7.8, or simply 8 if you round it off. Next, he just subtracts every number in the table from this base value of 8 to see what the effective modifier is in each weapon-vs-armor combination. That's it; pretty simple, really.

You can see these computed differences compared to the numbers that appear in Greyhawk below (and afterward, mostly repeated in AD&D). The numbers match very closely; this is clearly what Gary did.

Chainmail Man-to-Man Melee Chart and Conversion

Greyhawk Weapon Adjustments and Errors

More detail on exactly how closely these match: As shown in the last table above, most of the entries in Greyhawk vary from our computation by either 0 or 1 pips. The sum of all the differences in the entire table from our formula-based conversion is only 30 pips total. Hypothetically, if you use any number other than "8" as the basis (the average of all the targets in the Chainmail table), then you get much more divergent differences (far bigger sum of absolute errors).

In the number of cases where our formula differs from what appears in Greyhawk, clearly Gary was bothered by the larger penalties, and shaved them down to something less onerous. In particular, it looks like he was especially troubled by the line for "Spear"; with our formula it mostly has penalties all down the line, up to a hefty "−4". Gary joins many of us amateur historians in recognizing the spear as a dominant force on the real-world battlefield, and in response, dials down the penalties by about half. That's the only row in our "Absolute Error" table where the adjustments are as much as 2 points off our calculated formula; and we empathize with that. 

But wait. 

If the Spear made sense in Chainmail (presuming it did), then why is the translated version giving us such a headache in Greyhawk, anyway?

Can you see the gaping flaw in the conversion method?

It's huge, and it's obvious, but I didn't see it for decades, and I've never seen anyone else point it out.

I'll give you a minute to think about it.







It fails to recognize the natural protective value of the armor included in the Chainmail table.

Even if your "base to hit" in Chainmail was 8 versus Leather, say, it shouldn't also be 8 for Chain and 8 for Plate across the same row. It should be adjusted by some amount for each step of increased base armor type protection, even before the weapon effect gets involved. (In D&D, we expect a natural 1-point extra difficulty in hitting per step of armor; it's not explicit in Chainmail, but maybe 1/2-point per step in the 2d6 mechanic would be fair.) 

And as a result, you can't just be subtracting from 8, you should be subtracting from 8, 9, 10, 11, or something else, depending on the armor type at the top of the table. Whoops!

Let me be more specific by highlighting a single row and thinking about the story that it tells. Here's the Mace line in Chainmail. It is in fact pretty much just 8's all the way across. The story this tells is, "Screw your armor! Whatever armor you wear, I just ignore it. Maces reduce all worn armor to the same as 'none'". 

Mace 8 8 8 9 8 8 7 8

And here's the row for the Mace in Greyhawk . It is pretty much 0's all the way across. The story this tells is, "The mace is helpless against all armor! Whatever armor you wear, you get the full protective benefit against the mace. Maces have zero capacity to help you punch through heavier armor."

Mace 0 +1 0 0 0 0 0 0

(Remember that you need to flip one of the rows left-to-right to synch up with the reversed ordering of armor types in those two books; I picked this case for convenience in that regard.)

And those are precisely opposite stories, right? Maces reduce all armor to null-value in Chainmail, but give no benefit whatsoever in O/AD&D. And which of those two stories is more in synch with our real-world historical understanding? Chainmail, yea; O/AD&D, nay.

You can repeat this inspection on every row throughout the tables. Wherever Chainmail indicated an advantage over heavy armor, in O/AD&D this turns into no advantage (e.g., the Mace). Where Chainmail indicated no advantage, this turned into a hefty penalty in O/AD&D (e.g., the Spear).

Here's another way of putting this: The factors that go into the Chainmail combat table are weapon type, and armor class. The factors that go into the Dungeons & Dragons attack matrix are character level, and armor class. So when you blindly slam those numbers together, the protection from the base armor class has been double-counted, and every converted weapon in the game looks incrementally worse against heavier armor. 

(Side observations: What's the best weapon vs. plate in the D&D Greyhawk table? The military pick, largely because it wasn't included in Chainmail, so wasn't tainted by the double-counting error, and Gary could pick fresh values that actually made sense for it. Another: Swords & Spells Appendix A has a "% Chance to Hit" by weapon type for normal men table, which largely recreates the format of the Chainmail table -- but the value curves are totally incompatible, because Gary included the D&D modifiers, thereby inheriting the double-counted-armor problem)

Yikes. I'm pretty sure this is the biggest numerical error I've ever seen in the legacy of D&D, and I do think it seems to have escaped everyone's notice for lo these 45 years and counting. 

So I'd say that any 1E players who are still engaged in this gnashing-of-teeth exercise with these tables would be wise to put it to bed, because the whole effect of those tables in O/AD&D was fundamentally broken all along. It doesn't even begin to serve the goal that they're allegedly for. I'm guessing that they were never playtested at Gary's table -- again, he was adamant that he never used them, and was essentially disinterested in the whole project -- but once they got printed and published, everyone took it on faith that they were fit for the purpose. But they very much weren't.

Mea culpa on my own part for not ever noticing this before. I was amazed when this finally dawned on me a short while ago.

To your knowledge, did anyone ever point out this arithmetic-modeling mistake previously? Can you think of any reason to use these tables as-written today? Or would you be amenable to a series of corrected, rationalized tables for this purpose?

Get a spreadsheet for this comparison here.


Friday Figures: Fighter Multiattacks

Do hero-types get multiple attacks vs. 1 HD creatures? 16 Yes, 1 No.

Here's another poll I ran on the ODD74 boards a while back. This dovetails with our article Monday on running mass fights in classic D&D -- we highlighted there how critical for balance purposes the original rule was in which fighters got a number of attacks equal to their level against normal (circa 1 HD) opponents. 

That's not a rule that appears anywhere in the Holmes Basic/B-X/BECMI line, or editions of D&D from 3E on. It's not even explicitly stated for fighters in the OD&D books themselves. It's not a rule that I really like design-wise. So at the time I was myself rather skeptical, and asked the community if they really played by that rule. (On the other hand, it is explicit in Chainmail Fantasy and 1E AD&D, so I wouldn't bother to ask this question in a forum for those rulesets.)

Many of these poll-presentations are meant to spotlight the fact that many pretty basic mechanics in Original D&D have wildly varying interpretations -- that there is no one true OD&D game (even for essential things like initiative, the action sequence, morale, etc.). But this is not one of those things. Somewhat to my surprise at the time, the response here was very lopsided: Almost everyone agreed that was a core rule. We got 16 "yes" votes, only 1 "no", and 0 "other". So I spent a bit of time re-evaluating the evidence, and why I was so far out in left field on the issue.

Briefly, here's where the evidence led me to think maybe "no":

  • The multiattacks rule is explicit in the Chainmail Fantasy rules (prior to OD&D); it's the essential way in which heroes & superheroes operate.
  • The rule is also explicit in the AD&D 1E PHB (after OD&D); it's given as a key ability for the fighter class and its sub-classes.
  • Some people point to a paragraph in OD&D Vol-2 (the Monsters book), p. 5, that says everything gets a number of attacks according to their hit dice vs. normal men. I don't find that to be persuasive, because (a) it's in the monster book and not explicit for PC fighters, and (b) no one I know has ever run D&D combat that way for monsters. 

But on closer inspection, there's two places where Gary wrote about OD&D that imply the intended answer was "yes":

  • The Strategic Review #2 has an OD&D FAQ, with an example of combat in which a 4th-level fighter gets 4 attacks per round against orcs (strangely doing so unarmed, but moving on...)
  • Swords & Spells p. 1, paragraph 2, notes that a solo 12th level fighter will do 1.2 times the damage given normally for a 1:10 scale figure, which works out to be the same thing. 

So in total: The fighter-multiattacks rule is written into Gary's rules before OD&D, after OD&D, in the FAQ for OD&D, and in the mass-combat rules for OD&D. So in total: yeah, it's pretty clearly a consistent intent there (and another example of his tendency to "implicitly assume rules from the prior edition").

After that is when I realized that in addition, all the monsters in OD&D that appear in numbers of hundreds are precisely the ones that fighters get multiattacks against (excepting gnolls, but then they too were 1 HD in the pre-D&D draft).

A few other interesting tidbits on this topic:

  • The one place in OD&D where this rule is arguably given (the Vol-2 passage), says, "Attack/Defense capabilities versus normal men are simply a matter of allowing one roll as a man-type for every hit die...". There are a small number of players who attest that they apply the "roll as a man-type" strictly, so that the multiattacks themselves are all rolled as if the attacker was 1st level. I've never done this, but: it is compatible with the Chainmail mechanic, and it would solve my problem with the discontinuity/double-dipping effect of the rule.
  • Some commenters on Monday were worried about how much play might slow down because of all the rolling for the multiattacks -- and I share that concern. Interestingly, the ODD74 forum has reports that at a later date Gygax -- and maybe Arneson, too -- boiled the whole thing down to a single die-roll in this way: if your fighter is level N, then roll a dN, and that's how many normal men you put down. (E.g.: At 8th-level, roll a d8 for total casualties this round.) I think that's fascinating. I still don't want to see your damn d7, though, so don't start with me on that.
  • Even though I posted that poll a year ago in February (and the poll itself is long closed), interestingly, the conversation is still going on even as I write this!

There are several other interesting takes or modifications to this rule. If you have an account, see here for the discussion thread on the ODD74 forums

Do you normally use the fighter multiattacks rule? Do you import it to a ruleset that normally doesn't have it on the books (Holmes, B/X, BECMI, 3E+, etc.)? What other novel modifications do you make to it?


Running Mass Fights in Classic D&D

A common troubling point of DM'ing appears when you run the O/AD&D game, especially in the wilderness, by the book: it's pretty common to encounter bands of malicious bandits, brigands, or humanoids numbering in the hundreds.

OD&D Monster Reference Table

What's the best way to handle these cases, and what is likely to happen? I've actually had pretty good luck in these cases, so I want to share it here with you.


First, I'll point out that this is something I've been wrestling with for a very long time. In particular, I ran up against this in preparing to run the famous World of Greyhawk adventure, Isle of the Ape. The first encounter has the PC party ambushed by over 250 native warriors, with a minimum barbarian level of 3rd, about 30 name-level leaders or above, and 5 trained giant apes. How to adjudicate that? I had the opportunity to ask Gary Gygax about this on the old ENWorld Q&A thread (21st June, 2005):

DELTA: How did you ever deal with the initial combat with some -300 barbarians of varying high levels in AD&D? (Personally, I had to jury-rig a set of mass-combat rules to handle it.)

GARY: Actually, IIRR the PCs from my campaign popped in, surprised the natives, and offed the main leaders quickly, so the mass fled from them. Then the lads ventured past the wall, got a look at things, and got out of there. I don't recall how they managed it, but they left, returned to attack the shaman and his guards, and eventually some of them faced Oonga...

Now, a few things you'll note here. One is that Gary's rough instinct is that defeating a large warrior band comes down to defeating the leaders. The other is that I refer to "a set of mass-combat rules" to handle the situation, that later became my publication of OED Book of War (see sidebar).  


Honestly, this is something I tend to overlook in my analyses, until someone reminds me, but this piece of game rules is very prominent in O/AD&D, and you shouldn't entirely overlook it. If PCs encounter an overwhelmingly large band of enemies, then it's quite likely they can choose to evade and escape completely without any combat. 

In particular, the OD&D evasion chart is specifically biased to make escape easier, the more monsters in the enemy party. E.g., for a PC party of 4-9 people, the chance to escape the upper-end of monster numbers is set at 70%. (Increased in woods, if the enemy is surprised, etc.) Although remember if the PCs are surprised this implies that they're inescapably surrounded by the monsters (Vol-3, p. 17).

A couple notes on that: The exact rules text of the outdoor evasion rules is mangled in both OD&D and AD&D in a way that several details are unclear. Slower monsters can't ever catch the PCs, but by the letter of the rule, the evasion process continues, possibly in an infinite loop forever (see other blog articles for this examination). You'll need to make some reasonable interpretations for yourself to fill that in.

Moreover, the evasion rules are not a panacea to PCs confronting the immense danger of the O/AD&D wilderness. Even in the beneficial case noted above, there's about a 1-in-3 chance that PCs are spotted, cannot evade, and have a 50% chance to be caught by a faster monster army. (With this chance repeated over and over, depending on your interpretation of the rules there.) Conclusions we can make are that the by-the-book O/AD&D wilderness rules are meant only for very high level characters (see stats here) -- or that it's super important to keep a high move rate in the wilderness, to avoid all the galumphing bandit and humanoid hordes on foot.

I'll point out that you'd better explicitly tell your players that this is an option, because the Evasion charts don't appear in the player-facing rules of any early edition (OD&D Vol-3, AD&D DMG). 

Note also that in most versions of the rules, elves and halflings have some special ability to hide from enemies, especially outdoors (see: Chainmail, OD&D Vol-2, etc.). Consider letting those PCs check that chance before turning to the Evasion table for effective members in the escaping party (and thus possibly increasing the chance to avoid: for 1-3 PCs the chance can be as high as 90% base).

If all evasion fails, and you have low-level PCs entrapped by an army of humanoids, consider hand-waving the combat, declaring them captured, and switching the campaign to other PCs until the original group can be rescued, ransomed back, etc.

Ranged Combat

Assuming that the monsters are encountered at long distance, and the PCs decide to engage, then some initial long-distance missile fire may occur. (If evasion is attempted and fails completely, the OD&D text says monsters "will catch" the party, while AD&D says there is a "confrontation" -- I interpret this as being in melee distance, but you might differ.) 

There's an interesting observation on this point in one of my favorite Dragon articles (issue #20, Nov. 1978), by Lyle Fitzgerald, "It's a Good Day to Die (Death Statistics of D&D Players)". In his data, Goblin races at the top of the list, accounting for 10.1% of the 600 deaths documented by player-characters in Lyle's campaign. He writes:

‘Goblin types’, while not being strong individually (although they may have the occasional troll or ogre with them), are usually found in large groups, eager to destroy, and can be encountered practically anywhere or anytime according to the monster encounter tables. Typically, you will come across up to 300 or 400 orcs or goblins, who will proceed to pepper your group with a huge cloud of arrows, wiping out all the low level players and hirelings. Very few higher level players ever get done in by these creatures, however.

Now, I might possibly counter-argue this a bit in that where evidence exists, Gygax specifies a fairly small proportion of any humanoid band as carrying missile weapons -- usually around 20%. In the D&D Monster & Treasure Assortment, the percent ticks upward with dungeon level, but tops out at 20% for kobolds, goblins, and dwarves; 25% for orcs and hobgoblins; 30% for gnomes and gnolls, 40% for bandits (there may be typos here), and 50% for elves (as given in OD&D Vol-2). In the Monster Manual the ratios are 15% for gnolls; 20% for kobolds, goblins, orcs, and hobgoblins; 25% for dwarves and elves; 30% for bandits; and 40% for halflings. 

I should probably write more on that analysis later, but, in summary, I use 20% as the standard proportion of missile weapons among any humanoid group. Looking back to Fitzgerald, the average goblin encounter of around 200 has only 40 missile weapons, and the maximum of 400 has 80 -- e.g., specified as slings in the Monster Manual (and undefined in M&TA).

Then you run into the issue of reasonable long-ranged accuracy, for which I've probably written tens of thousands of words on this blog. At maximum bow-range, hitting an army is certain, but hitting an individual man is impossible -- so a small PC band has a distinct advantage here. I'd let PC missile shots target a random figure in an army at effectively no range penalty; while some significant extra penalty should be applied to the army's shots on PCs.

Looking to Gygax's Swords & Spells rules (a work which is otherwise flawed in significant ways), there are fairly well-considered penalties in this situation for missile fire. If the "target is one rank deep or in order" there is a -30% adjustment to casualties; if the "target is single creature, very large" then -50%; if "target is single creature, about man-sized" then -90%. Perhaps for brevity we might say a PC band of up to 10 people gets the 50% reduction, and we simply disregard half of the goblin shots (so maybe have 20 incoming sling-stones on average). 

At this point what I do is split up the arrows equally to each character or unmounted beast, take a handful of d20's, and go around the table rolling for each in batches. If there are 8 figures in the current example, then that's 2 or 3 shots per character. If the PCs have decent armor, and are being shot by 1-hit-die humanoids, then it's quite likely that only a 20 hits them. I would definitely skip critical hits from Nat-20's in cases like these (further abbreviating the process). I suppose you could have the players themselves roll these attacks and tell you the results, assuming you trust them, to make it even quicker. 

Characters Leading a Large Force

Now we get to engaging in close combat. There are a number of reasons why the PCs might be attached or leading a large armed force of their own. In my games, I've seen this occur when: (a) Players hire bands of mercenaries at the outset, (b) characters have previously defeated a force of bandits and pressed them into service, (c) encounter dice indicate simultaneous merchant bands & bandits locked in combat as PCs arrive, (d) a published adventure sets things up this way, etc.

In this case it does work pretty well to hand-wave the normal types battling the background, and have the PCs face off against the enemy leaders in melee. That's what all those copious notes on enemy leader-types in the Monster Manual are presumably for, right? (I mean, maybe?) And it also somewhat nicely echoes the idea of combat-of-champions in certain older cultures. 

Usually the PCs tend to have an advantage in number of leveled characters in this case -- even if the top enemy character is higher level, the action economy swings things in the PCs favor here.

For the background combat, I mentally apply the Book of War core rules -- every 3 rounds of D&D man-to-man fighting, roll a d6 for each 10 normal men battling; results of 4, 5, or 6 (depending on armor) each indicate 10 enemies down. Easy and statistically accurate. 

Characters on Their Own

Or, it's of course at least as likely that you have a small force of PCs on their own in the wilderness. Let's say they opt out or fail at Evasion, survive the Ranged Fire phase, close with the enemy to do combat, and they're high-enough level that hand-waving their automatic defeat is unreasonable.

First of all, if you're running the O/AD&D rules, you'd better honor the fact that fighters get a number of melee attacks equal to their level against normal-type creatures (roughly 1 hit die or less, but varying a bit by edition). So your Superhero (8th-level) fighter is likely going to hew through 8 goblins every round. My feeling now is that this is a paramount point of balance in those editions -- and also likely the single biggest rules difference between O/AD&D and the Basic/BX/BECMI line, where it was discarded. The types of humanoids that appear in numbers of hundreds are precisely the same types that fighters get their mass-attacks against. If you don't use that rule, then arguably you need to dial down the humanoid numbers by an order of magnitude (as was done in Basic/BX, etc.). 

(Hypothetically, I'm not fond of that rule in a design sense, because I don't like hard discontinuities like the normal/fantastic split in early D&D, it slows the game with lots of dice-rolling, etc. And it didn't help me with Isle of the Ape because everyone involved is above 1st level, hence my asking about that specific case to EGG. But it is a core part of the original rules and you should consider it carefully.)

Secondly, terrain is important. When PCs in my games march around the classic Outdoor Survival map, it's quite likely that they're following a path through woods or mountains (which voids chances to get lost). If that's the case, then the main body of humanoids can easily be bottled up, constrained, and dealt with by PCs facing a fairly limited front line. Probably the humanoid army is in a column formation and not well organized to engage in a fight. (I have small d6 charts of other layouts by terrain type that I use, possibly giving advantage to PCs.)

Third, don't forget about morale. It may not be very well defined in OD&D, or commonly used in AD&D, but it's a critical expectation on the part of the system designers. Recall that the kernel of what Gygax wrote in response to my question was that the PCs "offed the main leaders quickly, so the mass fled from them". Make sure you have a mechanic to your taste to handle this (whether from Chainmail, the OD&D reaction chart, the AD&D DMG system, B/X rules, or my BOW morale system of rolling 2d6 + HD and scoring 9 or more). You could possibly apply it to the whole armed opposition, or one unit at a time (foot contingent, archers, horsemen, etc.).

And if you get to this point, you get the joy of narratively describing a fight of epic heroes vs. hordes of enemies in the way that delights your heart the most. The fighters may wade in with their mass-attacks directly, or come up with other creative responses. I've had PCs with super-strength (from a spell, gauntlets of ogre power, etc.) go up hills and roll down boulders, throw trees to block a path of charge, rip off castle gates, etc., and that's all wonderful. 

Magic-users, of course, get to unleash their best area-attack spells. My interpretation of areas in OD&D is that spells like fireball or lightning bolt can only eliminate up to 10 enemy characters at a time -- but by-the-book it could be 40 characters or even 360 if you let the area expand feet-to-yards outdoors (which Gygax clamped down on later, and in Swords & Spells wrote another ad-hoc rule to limit total damage of such spells: example in the book has a 10th level wizard killing around 40 orcs with a fireball maximum, scaling downward for lower levels). A death spell or a conjured air elemental in whirlwind form could be even more devastating. 

But other options could be using illusions (wand or phantasmal force) to create fake allied units and confuse the opposing army -- maybe wasting a round of mass missile-fire (in fact, that was the origin and etymology of the phantasmal forces spell in Chainmail). Or perhaps targeting mind-control powers like charm, hold, or feeblemind spells against the enemy leaders might be more effective. I pretty much always allow the enemy leaders to be identified and targeted in this way (and likewise assume enemy leaders can target PCs the same way, including fighters in swirling melee) -- more than once an engagement with an enemy force has ended without mass casualties, due to the commander being ensorcelled and told to just walk away. There are lots of options here.


For a mid- to high-level party in classic D&D, there are many possibilities when confronting a mass brigade of armed opponents. They can evade the encounter, bring their own armed troop, probably withstand any mass ranged attacks, and then bring super-powerful attacks to bear in fighter whirlwind-melee and wizard area and mind-bending magics. That's not even considering the option to parley, which I've ignored here (OD&D says monsters automatically attack except in unusual cases, which I've assumed by default). My best experiences have been running things purely narratively without a board or map; this is an occasion when using the "rule of cool", bouncing off player suggestions, and giving at least some chance for effectiveness, can solve a lot of problems for you. 

How has running mass combats in D&D worked for you in the past? Any tips or tricks I'm missing here?

This blog post was chosen as the Wandering DMs Patrons' Pick for February 2021. If you want to get in on future polls for things you'd like to see here, join the Wandering DMs Patreon. We thank you for the support!