You can download the rules here (PDF), or from a link at the bottom of this page. 

DESIGNER’S NOTES & COMMENTARY

The impetus for developing these rules came when I began to run the adventure Queen of Lies (by Monte Cook, published by Fiery Dragon Productions) in my D&D campaign.  In this adventure, the PCs can lead a force of low-level soldiers on a campaign against subterranean foes, which in addition to the normal “level-appropriate challenge” fare also sometimes include large groups of weak opponents (fondly referred to herein as “chum”).  This can pose some problems for the DM.  Fighting battles against chum isn’t particularly interesting for anyone: the PCs aren’t really at risk, the outcome is never in doubt, and it can be very time-consuming and tedious to resolve (the burden of maneuvering and rolling for every individual enemy becomes particularly onerous when there are 50 or more of them).  It’s possible to just hand-wave the battles with low-challenge foes (“Okay, so you spend 5 rounds mopping up the twenty orcs, and let’s say you use 2 cure light wounds spells healing back to full hit points”) but that isn’t very satisfying or realistic in a few specific cases:

First, sometimes the low-challenge foes are accompanied by NPCs or monsters sufficiently powerful to threaten the PCs.  In this case the chum, while still mainly ineffectual against the party, can at least serve as distractions or shields for their more powerful allies.  They might provide flanking bonuses, or just impede movement (making it harder for fighter-types to close with spellcasters, for example).  In this situation, the chum function like terrain for the PCs and NPCs to contend with.

Second (especially relevant in Queen of Lies) is the situation in which the PCs have hirelings, cohorts, or allies for which the chum are a significant threat.  In a military-themed campaign, the PCs might actually be officers commanding a company of troops, leading them against other forces.  Here one objective is to preserve the friendly forces as much as possible while still defeating the enemy.  How to keep things “interesting” for the allies of the PCs without bogging the game down too much?  (And how to efficiently represent large numbers of PC allies?)

And third, sometimes the low-challenge foes are in sufficient numbers that they do pose a significant threat to the PCs.  If your PCs are somewhere around 5^th to 7^th level, they probably only have access to a few area-effect spells such as lighting bolt and fireball, their equipment probably doesn’t render them “unhittable” except on rolls of 20, and they don’t have so many hit points (and so much ready access to healing magic) that a few sword or arrow hits can be laughed off.  In this realm, a mob of say 40 or 50 vanilla kobolds or goblins could pose a significant challenge, and one that is probably well-suited to resolution using these rules.

My point of departure for the development of these rules was Cry Havoc (by Skip Williams, published by Malhavoc Press).  Cry Havoc has unit combat as its focus, but I found many of its abstractions and assumptions unsuitable for the tactical-scale play that I wanted.  In Cry Havoc the ground scale changes to 1” = 50’ (instead of 5’), and each figure represents 10 individuals.  The time scale also changes, so each round represents one minute instead of 6 seconds.  As a result of this scale change it becomes different to resolve traditional d20/D&D-style combat in parallel with the larger clashes of units and armies.  This isn’t intended as criticism of Cry Havoc, which is focused on army combat; but for my game I wanted to keep the focus and scale on individual PCs, merely streamlining play involving lots of chum.  Cry Havoc doesn’t deliver quite what I was looking for in this case.

While I liked the general approach taken in Cry Havoc of treating a unit as a single entity, and of averaging the individual statistics of the unit’s members to arrive at the unit’s statistics, after some analysis and thought I didn’t much like the combat resolution tables themselves.  While within the abstraction of 1 minute turns they might serve well enough, in the 6-second round of D&D in my opinion they are insufficient to capture the expected result of a large number of attack rolls.  It was relatively easy to apply basic statistics (the binomial distribution) to the d20 resolution mechanic to come up with tables that more accurately reflect the expected number of successes for a given number of rolls.  As an added bonus, these tables can be used to resolve large numbers of other rolls/checks, such as saving throws and skill checks, as discussed in the rules.

Treating units like independent figures was an early, easy decision.  Everyone is already familiar with maneuvering figures on the battlefield.  Coming up with unit size was also fairly straightforward.  Using Medium-size figures to make up the units, I looked at how many figures would fit in a 10’x10’ or 15’x15’ space, and added one figure to arrive at the nice, “round” figures of 5 and 10.  Given that the exact location of the individuals in a unit isn’t being tracked, this is I think a reasonable approximation.  If you like, you can assume that figures in a unit occasionally “spill out” beyond the edges of their space.  The intent is that each figure within a unit is acting more-or-less in accordance with the standard d20 combat rules.  You can argue that it shouldn’t be possible for all 10 figures in a squad to make a full attack against a single Medium-sized figure in a round, and that may be true, but certainly more than 3 or 4 figures could attack, as follows: the front rank attacks and shifts away, making room for another rank to shift forward and then attack.  (We allow the figures within a unit to notionally “overstack” temporarily as a concession to playability.)  The remaining figures might have to use move actions to get from the “rear” of the unit space to a space adjacent to the target, and this might also require temporarily leaving the unit’s notional space.  This would prevent those figures from using full attack actions.  But it’s my expectation that most figures aggregated into units probably won’t have multiple attacks, and even when this isn’t the case secondary attacks are generally weak enough relative to primary attacks that it’s not likely to greatly change the unit’s effectiveness.  (You can also argue that, while these figures are more effective than they should be through getting a full attack when this might not really be allowable, they are also losing out by not being able to grant other figures in their unit flanking bonuses.)  My feeling is that it all comes out in the wash.  If you care that much about those details, you shouldn’t be aggregating these figures into units anyway.  (Also note that you could allow individual figures to “split off” from units at the start of the unit’s turn to resolve any such special cases, if desired.)

As noted above, the combat resolution tables are derived from the binomial distribution, with n=10 (for the squad table) and n=5 (for the team table).  As discussed in the rules, the Target Number is the “natural” number you need to roll on a d20 to succeed (at an attack, a saving throw, or whatever), which can easily be converted to an overall chance of success.  The derivation of the Target Number from the Armor Class/Difficulty Class and any attack/save bonuses should be straightforward.  The modification to Target Number due to unit casualties suffered is an approximation.  I didn’t want to have to include one table for every possible number of rolls, and inspection of the two tables provided told me that adding one to the Target Number for every “missing” roll gives approximately the correct result.  As the casualty adjustment gets larger, this approximation gets worse (one reason for having a table for teams, rather than just treating a team as a squad with 5 extra casualties) but I could live with it being limited to a maximum adjustment of 4 (for a squad having taken 4 casualties).

The Unit Damage Modifier was inspired by a similar modifier in Cry Havoc.  The idea here is that to simplify damage resolution and tracking in unit combat, hit points and damage are aggregated into groups of 5 hit points called Damage Factors.  So instead of 10-14 hit points, an individual instead has 2 Damage Factors.  The combat tables are intended to determine how many “1 Damage Factor” hits are scored against a given Target Number.  But if any given hit only deals an average of 2.5 points of damage, that means you need to score 2 hits to equal 1 Damage Factor.  That’s where the Unit Damage Modifier comes in.  It applies an adjustment to the Target Number to calculate the approximate number of Damage Factor equivalents inflicted by a hit.  A modifier of +2 is about equal to a 33% increase in the number of hits; +4 equates to a roughly 50% increase, and +6 increases hits by 100%.  So if an attack deals average damage of 7.5 (1d8+3, say), then only 2 actual hits are needed to inflict an average of 15 damage (3 Damage Factors).  An adjustment of +4 to the attack bonus (equivalent to subtracting 4 from the Target Number) increases the number of hits scored by about 50%, so a roll that would normally inflict 2 successes instead inflicts 3.  This modifier isn’t used for attacks against individuals because damage vs. individuals is rolled normally, so we want to know the actual number of hits achieved with no adjustment for average damage.

The notion of a dispersed unit formation was added because the tightly-clustered units modeled here are ideally suited for being wiped out by area effect spells like fireball and flame strike.  In a world where such destructive magic is, if not common, certainly at least a known threat, dispersal can be crucial to survival and would almost certainly be incorporated into training and doctrine.  The penalty to attack bonus models the idea that all of the figures in the dispersed unit cannot converge to attack a single target without giving up their dispersed formation.

The morale rules are fairly simplistic, again with the goal of keeping play moving while still maintaining the idea that units of chum might decide to call it a day when confronted with massive casualties or overwhelming threats.  The list of conditions mandating morale checks and the list of check modifiers can be expanded at the DM’s desire.  The DC for the morale checks is purposely kept low, so that even weak units (e.g. those composed of level 1 warriors) are likely to maintain morale (a battle in which most of the troops on both sides run away after first contact isn’t very interesting).  Some of the morale check modifiers are included specifically to allow PCs and NPCs to play significant roles in bolstering unit morale.  A campaign with a dedicated military flavor might want to expand on these Morale Rules, or investigate the more detailed treatment (including additional skills and feats) presented in Cry Havoc.