Tag Archives: combat

Dungeon World weapons

Leave a reply

The weapons system in Dungeon World is perhaps the best (for my purposes) that I have seen in any published RPG. It uses a tag-based system to distribute effects between the various weapons. Thus, they become distinctive while offering interesting trade-offs. Best of all, it looks like it could be dropped, unmodified, over a flat d6 or class-based damage system in D&D. Dungeon World itself assumes class-based damage.

It’s not perfect, as there are some items that are just plainly suboptimal other than cost (such as elven arrows versus normal arrows), and I’m never satisfied with cost being the main differentiator because it is not really much of an obstacle in a game about treasure hunters (DW is by assumption less coin-heavy than traditional D&D though, so maybe it works there).

Below I have reproduced the DW weapon rules, since Dungeon World is creative commons licensed. Hence the legitimacy of this cut-and-paste job. Thanks to Adam and Sage for making their work available in this way. In the published book, this material is on pages 324 through 326.

For other weapon systems that try to approach the problem similarly, see:

General Equipment Tags

These are general tags that can apply to just about any piece of gear. You’ll see them on armor, weapons or general adventuring tools. Applied: It’s only useful when carefully applied to a person or to something they eat or drink.

  • Awkward: It’s unwieldy and tough to use.
  • +Bonus: It modifies your effectiveness in a specified situation. It might be “+1 forward to spout lore” or “-1 ongoing to hack and slash.”
  • n coins: How much it costs to buy, normally. If the cost includes “-Charisma” a little negotiation subtracts the haggler’s Charisma score (not modifier) from the price.
  • Dangerous: It’s easy to get in trouble with it. If you interact with it without proper precautions the GM may freely invoke the consequences of your foolish actions.
  • Ration: It’s edible, more or less.
  • Requires: It’s only useful to certain people. If you don’t meet the requirements it works poorly, if at all.
  • Slow: It takes minutes or more to use.
  • Touch: It’s used by touching it to the target’s skin.
  • Two-handed: It takes two hands to use it effectively.
  • n weight: Count the listed amount against your Load. Something with no listed weight isn’t designed to be carried. 100 coins in standard denominations is 1 weight. The same value in gems or fine art may be lighter or heavier.
  • Worn: To use it, you have to be wearing it.
  • n Uses: It can only be used n times.

Weapons

Weapons don’t kill monsters, people do. That’s why weapons in Dungeon World don’t have a listed damage. A weapon is useful primarily for its tags which describe what the weapon is useful for. A dagger is not useful because it does more or less damage than some other blade. It’s useful because it’s small and easy to strike with at close distance. A dagger in the hands of the wizard is not nearly so dangerous as one in the hands of a skilled fighter.

Weapon Tags

Weapons may have tags that are primarily there to help you describe them (like Rusty or Glowing) but these tags have a specific, mechanical effect.

  • n Ammo: It counts as ammunition for appropriate ranged weapons. The number indicated does not represent individual arrows or sling stones, but represents what you have left on hand.
  • Forceful: It can knock someone back a pace, maybe even off their feet.
  • +n Damage: It is particularly harmful to your enemies. When you deal damage, you add n to it.
  • Ignores Armor: Don’t subtract armor from the damage taken.
  • Messy: It does damage in a particularly destructive way, ripping people and things apart.
  • n Piercing: It goes right through armor. When you deal damage with n piercing, you subtract n from the enemy’s armor for that attack.
  • Precise: It rewards careful strikes. You use DEX to hack and slash with this weapon, not STR.
  • Reload: After you attack with it, it takes more than a moment to reset for another attack.
  • Stun: When you attack with it, it does stun damage instead of normal damage.
  • Thrown: Throw it at someone to hurt them. If you volley with this weapon, you can’t choose to mark off ammo on a 7–9; once you throw it, it’s gone until you can recover it.

Weapons have tags to indicate the range at which they are useful. Dungeon World doesn’t inflict penalties or grant bonuses for “optimal range” or the like, but if your weapon says Hand and an enemy is ten yards away, a player would have a hard time justifying using that weapon against him.

  • Hand: It’s useful for attacking something within your reach, no further.
  • Close: It’s useful for attacking something at arm’s reach plus a foot or two.
  • Reach: It’s useful for attacking something that’s several feet away— maybe as far as ten.
  • Near: It’s useful for attacking if you can see the whites of their eyes.
  • Far: It’s useful for attacking something in shouting distance.

Weapon List

The stats below are for typical items. There are, of course, variations. A dull long sword might be -1 damage instead while a masterwork dagger could be +1 damage. Consider the following to be stats for typical weapons of their type—a specific weapon could have different tags to represent its features.

Ragged Bow
near, 15 coins, 2 weight
Fine Bow
near, far, 60 coins, 2 weight
Hunter’s Bow
near, far, 100 coins, 1 weight
Crossbow
near, +1 damage, reload, 35 coins, 3 weight
Bundle of Arrows
3 ammo, 1 coin, 1 weight
Elven Arrows
4 ammo, 20 coins, 1 weight
Club, Shillelagh
close, 1 coin, 2 weight
Staff
close, two-handed, 1 coin, 1 weight
Dagger, Shiv, Knife
hand, 2 coins, 1 weight
Throwing Dagger
thrown, near, 1 coin, 0 weight
Short Sword, Axe, Warhammer, Mace
close, 8 coins, 1 weight
Spear
reach, thrown, near, 5 coins, 1 weight
Long Sword, Battle Axe, Flail
close, +1 damage, 15 coins, 2 weight
Halberd
reach, +1 damage, two-handed, 9 coins, 2 weight
Rapier
close, precise, 25 coins, 1 weight
Dueling Rapier
close, 1 piercing, precise, 50 coins, 2 weight

Escape or Initiative

14 Replies

In my previous necrology post about Satyavati, Hedgehobbit brought up a rule from The Underworld & Wilderness Adventures regarding when it was appropriate to call for an initiative roll. The situation involved a PC approaching something carefully that they knew to be dangerous, presumably ready to turn and flee at any moment. Under what circumstances is it appropriate for a threat to engage a PC in combat? Hedgehobbit suggested that the rule was if within 20 feet, roll initiative, otherwise compare movement rates to see if the PC can outrun the monster.

The actual text (from page 20 of TU&WA) is:

There is no chance for avoiding if the monster has surprised the adventurers and is within 20 feet, unless the monster itself has been surprised. If the adventurers choose to flee, the monster will continue to pursue in a straight line as long as there is not more than 90 feet between the two.
…
Distance will open or close dependent upon the relative speeds of the two parties, men according to their encumbrance and monsters according to the speed given on the Monster Table in Volume II. In order to move faster characters may elect to discard items such as treasure, weapons, shields, etc. in order to lighten encumbrance. 

Though initiative is not mentioned explicitly (in fact, as far as I know, initiative is not mentioned at all in the 3 LBBs), Hedgehobbit notes that “no chance for avoiding” does seem to imply that combat rules, whatever they are, become active if those requirements (surprise, within 20 feet) are satisfied. Now, in Satyavati’s case, no surprise dice were rolled (though I suppose that a perfectly still statue suddenly moving could maaaaaaybe be considered surprising) so strictly speaking this rule would not be relevant.

The rule “distance will open or close” is difficult to manage without a grid or doing extra math. I actually prefer Hedgehobbit’s version as originally stated. If a PC is within 20′ of something, the combat rules are potentially in effect. If farther than 20′ away, the pursuit rules are used (movement rates are compared directly).

This would have resulted in the outcome produced by the session, is easy to remember, allows players to reason about risk without introducing certainty, and seems intuitively reasonable.

The 20′ threat range could also potentially be derived from monster movement rates (movement divided by 4, round down maybe) so that faster monsters would be more able to force combat. It might also be reasonable to factor reach into the equation, so that an ogre (with a movement of 9) would have a 30′ threat range (9 / 4 = 2.25 + 10′ for reach = 3.25, round down to 30′ effective threat range). This also makes ranged attacks and missile weapons more dangerous, which is as it should be.

Combat & movement

3 Replies
Image from Dark Classics

Jack recently wrote about differentiating weapons. I have also in the past sought to make weapon differences meaningful over and above damage dice, with varying levels of success. Jack’s proposal has some Third Edition assumptions (such as critical ranges) that I don’t use, but I find one of the properties he gave daggers particularly interesting. In his system, dagger wielders may use movement actions to attack. Presumably, this is to represent quick close attacks and perhaps grappling.

5E is also experimenting with using movement as a resource that can be “spent” in combat. For example, five feet of movement can be used to stand up from prone, allowing a character to stand up, move (slightly less than normal) and attack all in the same round. Now, this particular rule might be too fiddly (and might be difficult to make work without using a grid). If doing combat only using a shared imaginary space, is there really much difference between 30 feet worth of movement and 25 feet worth of movement? Probably not.

I am, in general, not enamored of the action economy approach to combat. It tends to slow play down and make the decision process more complex without adding corresponding depth. For more on how this worked in 4E, and the proposed simplification for 5E, check out this blog post. Fifth Edition is also experimenting with other ways to spend combat time resources which look intriguing, such as spell concentration being required to maintain continuous effects (which should help control the problem of appropriately enchanted wizards being potentially better at any conceivable task, a problem that I gather can be relatively acute in Third Edition and Pathfinder, though I have never played high level games in either of those systems). For more on concentration in 5E, check out the second half of this Legends & Lore article.

Back to the topic at hand though, I still like the general idea of a tradeoff between mobility and other weapon properties. However, multiple attacks have the potential to be both cumbersome (extra die rolls) and unbalanced (that is, clearly superior to other weapons in damage dealing potential), so this needs to be handled carefully. Further, without a grid, it seems difficult or impossible to keep geometry and tactical placement relevant. Again, this makes me wish for a simplified and non-quantified representation of combat beyond conversational description. Something that would perhaps be gained by using miniatures in a loose, almost informal manner.

What kind of OD&D implementation based on movement might work for the dagger? Before a dagger wielder can get any kind of benefit from close fighting, they must first get inside an enemy’s guard. It seems reasonable to model that as a successful to-hit roll. They must also successfully bypass any kind of “hold at bay” active from pole-arms. Once the dagger has hit, the attacker is considered up in it and future attacks do “two dice, take highest” damage. Note that this also applies post-backstab for thieves. As long as the attacker chooses to maintain this disposition, no significant movement is possible, as they are focusing on carving up the target.

This is similar to a grapple (though there is no grabbing going on). Pole weapons are almost impossible to bring to bear against a dagger wielder up close, and all weapons other than a dagger or short sword attack at -1. The target may disengage by spending an action and making a successful dexterity check. Fighters may attempt a disengage maneuver along with a standard attack, but all other classes must spend all their efforts just to get the sharp thing away from them. Whether or not dagger work can be used effectively against non-humanoid enemies should be determined situationally (bear: sure, purple worm: not so much).

Combat dice

6 Replies
Semi-relevant image from Wikipedia

Prior to second edition D&D, the attack progression of characters is jumpy. In OD&D, for example, fighters don’t improve at hitting things until 4th level, and progression is even slower for other classes. This system, in the original conception (where all classes use the same chart, but get better at different rates) can be thought of as attack ranks.

Rather than an attack bonus, DCC RPG has fighters use an attack die (plus a small additional bonus at high levels). In effect, fighters determine the attack bonus randomly round by round. This only applies to fighters and dwarves though; all other classes use standard d20-style fixed attack bonuses. The results of the die used to determine attack bonus (called the “deed” die) also feed back into the DCC stunt system: a deed result of 3 or more results in a successful stunt.

Recently, I also thought about promoting attack ranks to first order game constructs that could be used in ways other than just attack bonuses. Specifically, attack ranks could be used defensively. This would allowing fighters to “spend” attack ranks on a round by round basis to improve their own AC or defend companions. This idea can be expanded to cover other areas of combat.

Rather than using a better attack matrix (or improved attack bonus), characters have a number of combat dice equal to the combat rank. In my adjusted attack rank system for OD&D, fighters start off at attack rank 2 and top out at attack rank 6 (at high level), while clerics (for example) start at 1 and top out at 4. Using those numbers, fighters would start out with 2 combat dice, which could be used for various things.

I see four main ways to use combat dice: attack bonus, personal AC bonus, defending companions, and extra damage (though not all classes would have access to all options). Basically, this allows players to change focus between offense and defense round by round, but since the number of dice is relatively small (between 1 and 6, inclusive) there are fewer choices to consider (and thus reasoning will probably be more diegetic and less about mathematical optimization).

Combat dice options by class:

  • Fighter: attack, armor, defend, damage
  • Cleric: attack, defend, armor
  • Thief: attack, damage
  • Magic-user: none
Shifting around of a relatively small number of dice between competing priorities seems like a better way to model things like fighting defensively than a flat attack penalty and AC bonus. It also makes the fighter scale up better with level compared to other classes without recourse to powers.

Thus, combat dice become general resources that all classes use and they replace the other attack progression system. In some sense, the tenor of combat for a specific class becomes the combination of how fast combat dice are acquired and what they can be used for.

Shield saves

16 Replies
Image from Wikipedia

I would like to experiment with an “active defense” option for shield use. Here is the proposal.

Shields provide the following benefits.

  • +1 AC
  • 1 shield parry per round
  • +4 to saves versus appropriate area effects
All of the benefits only apply when a shield user has freedom of movement during combat (so not against traps).
At the beginning of a combat round, characters must decide which enemy they will primarily direct their shield against. The shield parry is a reaction, and may be used against a successful hit by that enemy. Fighters use their most favorable saving throw, other classes use their least favorable saving throw. Note also the ability of axes to destroy shields.
Use of a shield save requires a full action for characters without armor skill (e.g., zero level humans and magic-users). In other words, a “full defense” type of action will allow the use of a shield saving throw even for a non-combatant class character.
Appropriate area attacks would include dragon breath and fireballs, but not, for example, cloudkill.
Hopefully, this will not prove cumbersome (an attack in addition to a save versus poison does not seem cumbersome, so I don’t see how this will be much different, though I suppose getting hit happens more frequently than getting hit and poisoned). In any case, figuring things like this out is what testing is for.

I think this rule would also work with my recent 2d6 fantasy game, without the +1 AC bonus (since the numeric armor scale is less extensive), and with only +1 to saving throws versus area effects.

Using Attack Ranks Defensively

5 Replies
Battle of Castillon (source)

In the ratling post, I alluded to an ability to use attack ranks for things other than just an increased chance to hit. Specifically, ratlings can use attack ranks to improve their armor class on a round by round basis, with each rank so applied improving the effective AC by one armor category. Ratlings naturally have light armor (AC 7). For example, a ratling who fights at attack rank 1 may choose each round whether she wants to have AC 7 and attack rank 1, or AC 5 and attack rank 0.

Fighters (but no other class) may also use attack ranks defensively in a similar way to improve their AC from round to round. In addition, fighters may spend their attack ranks to improve the AC of companions they are defending if that makes sense situationally. For example, consider a first level fighter wearing medium armor, using a shield, and fighting near an unarmored magic-user. This fighter has AC 4, the magic-user has AC 9. The fighter has the following options:

  1. Fight at attack rank 2. All armor classes remain unmodified.
  2. Fight at attack rank 1 and improve her own armor category one step (AC 2).
  3. Fight at attack rank 1 and improve the magic-user’s armor category one step (AC 7).
  4. Fight at attack rank 0 and improve the magic-user’s armor category two steps (AC 5).

Armor class may never be improved beyond AC 2. Defensive assistance from multiple fighters does not stack. Defensive assistance may be divided between multiple companions if that makes sense. I am not sure whether or not this floating defensive bonus should stack with armor. For now it does (as I would like to see people use this option some of the time). Fighter retainers may be instructed to defend their employer. Fighters must be armed with melee weapons or shields when defending companions (not missile weapons).

This ability is similar to the defense trait from the Hexagram path of arms. I think it is a nice addition to the fighter class, giving them some ability to defend other characters mechanically without resorting to other cumbersome subsystems. I could also imagine attack ranks potentially being usable for other things too, though I don’t want to overload the system too much.

Adjusted Attack Ranks

3 Replies
Battle from Holkham Bible (source)

The OD&D alternative combat system implies the use of attack ranks (discussed here and here previously). Based on my experience running Pahvelorn so far, I would like to adjust slightly how the various classes interact with this system.

Like armor, there are four categories of combat focus; each one corresponds to one of the four core classes. In the table below, focus is how much effort the class spends on improving combat skill. Begin is what attack rank members of the class have at level one. Improvement is a list of the levels where the class gains an attack rank. Max is the maximum attack rank attainable by members of the class. Regarding the maximum, keep in mind that no creature ever has AC less than 2 in this game.

Combat Competency By Class
Focus Class Begin Improvement Max
None
magic-user
0
none
0
Light
thief
1
5, 9
2
Medium
cleric
1
5, 9, 13
4
Heavy
fighter
2
4, 7, 10, 13
6

(Edit: starting thief attack rank changed from 0 to 1.)

The various classes are balanced around these trade-offs. For example, a magic-user is assumed to spend their time studying the stars and experimenting with new magic formulae. If they instead spend their time drilling and practicing skill at arms, their magic will suffer. Such things will be handled on a case by case basis within the game, and come with special requirements, such as finding a teacher.

Attack Rank
AC 0 1 2 3 4 5 6
2 18 16 14 12 10 8 6
3 17 15 13 11 9 7 5
4 16 14 12 10 8 6 4
5 15 13 11 9 7 5 3
6 14 12 10 8 6 4 2
7 13 11 9 7 5 3 2
8 12 10 8 6 4 2 2
9 11 9 7 5 3 2 2

Some further notes. Magic-user combat skill has obviously been pared down. Max thief combat skill is also capped lower, but note that a surprise attack allows the thief to attack as if with two extra attack ranks (that’s the translation of +4 to attack). The matrix has also been adjusted to reflect that a 1 is always a miss.

Attack rank attained will also affect what enchanted weapons a character can master (that is, use with full potential). This is an idea that I have been playing around with in Hexagram, and I believe it fits here as well, and is a more elegant solution than than just restricting weapons by class.

Silvered weapons

15 Replies
Image from Wikipedia

Some foes, such as lycanthropes and wights, are immune to standard weapons but vulnerable to silver, and most or all versions of D&D include silver versions of various items in the equipment list. OD&D prices silver arrows at 5 GP per arrow (compared to a quiver of 20 arrows for 10 GP), and Moldvay prices silver daggers at 30 GP (ten times as much as the standard dagger). Silver crosses are also available for 25 GP (the simple wooden variety is only 2 GP).

However, treasure hunting adventurers are not very price sensitive regarding mundane equipment, so increased cost does not have much effect other than during initial equipment buying. Thus, there should be some trade-off to using a silvered weapon other than just costing more initially. Otherwise, players will just outfit everyone with silver versions of everything, and then combattants will be assumed to always use silvered weapons, just in case. Most things that are unproblematically better are boring. So there should be some reason to not use silver weapons all the time.

A silvered weapon is not actually made of solid silver. Rather, it is an iron or steel implement that has silver bound to the blade in a process similar to gilding. Perhaps a ritual and some hedge magic or blessing is also required as part of the procedure. As it is used, the silver wears off. This process of wearing off is actually critical to the effective functioning of the silver weapon — you are essentially leaving traces of poison in the argyrophobic  creature.

A “silver die” (d6) should be rolled along with every damage die. On a silver die roll of 1, the silvering process has worn off, and must be re-silvered. Needing to roll an extra die also draws attention to the use of a silver weapon, making it more of an explicit choice, and less of a default. This makes silver arrows potentially more cost-effective than most silvered melee weapons (though note you can’t effectively fight with a ranged weapon if you are in melee). Also, a miss with a silvered melee weapon will not potentially degrade the weapon, but an arrow that misses may be damaged or lost. So the value comparison is not direct.

Most metal weapons can be silvered. The cost (following Moldvay) is ten times the normal weapon, and takes a skilled smith one week to complete. Given the cost of silvering, it makes sense to only use silver weapons when they are likely to make a difference. This is in effect a form of melee ammunition.

Silvered plate armor is available too, at the same cost multiple. Argyrophobic foes will generally prioritize attacking characters that are not wearing silvered armor, and will usually take a penalty when attacking combattants armored in silver (though this varies based on the specific creature). Silvered armor will also wear out in a similar manner (a silver die should be rolled per attack that is landed on the wearer).

Armor by Class

7 Replies
Image from Wikipedia

Image from Wikipedia

Armor is something I have never been totally satisfied with in D&D. It’s not because I object to the abstraction of armor making characters harder to hit rather than preventing injury (though I have written about that before). Rather, I don’t much like the idea that a magic-user can’t strap on a breastplate and get some benefit from it. This is not a major problem for me; I don’t have any issue playing traditional B/X with all the armor and weapon restrictions and having a great time. But it does seem like an imperfection.

That being said, I do think fighters should be better at using armor. The weapon restrictions problem, which is similar, has been very elegantly handled, in my opinion, using damage by hit die rather than by weapon (which I originally based on Akrasia’s damage by class). Perhaps something similar could be done for armor without totally modifying the underlying system?

First, I am going to fall back on my earlier assumption of a threefold armor classification (light, medium, and heavy corresponding to leather, chain, and plate). Traditionally, there are also several types of character class regarding armor use. The heavily armored classes can use plate (fighters and clerics). The lightly armored classes, such as thieves, which can only wear leather. And finally the unarmored classes, such as magic-users, which can wear no armor.

My basic idea is that by default, armor grants one point of AC per tier. So, for example, leather armor would grant +1 AC and plate armor would grant +3 AC. Someone with no armor skill (such as a zero level character or a magic-user) who straps on a suit of plate thus only gets +3 AC.

Additionally, characters gain an AC bonus equal to their armor skill if their armor skill is equal to or less than the AC bonus of the armor in question. The armor skills by class are:

  • Cleric (heavy): +3
  • Fighter (heavy): +3
  • Magic-user (none): +0
  • Thief (light): +1

In other words, you get to double the AC bonus when using armor of a tier less than or equal to your (class-based) armor skill. None of the big four classes seem to have medium armor skill (though in terms of balance, the cleric probably should). However, there are a number of subclasses that seem like they would naturally have medium armor skill, such as the ranger.

Effective AC Bonus
Class Armor Skill Light (leather) Medium (chain) Heavy (plate)
Cleric
+3
+1 +1 = +2
+2 +2 = +4
+3 +3 = +6
Fighter
+3
+1 +1 = +2
+2 +2 = +4
+3 +3 = +6
Magic-User
+0
+1 +0 = +1
+2 +0 = +2
+3 +0 = +3
Thief
+1
+1 +1 = +2
+2 +0 = +2
+3 +0 = +3

In the table above, each armor column has three numbers: the inherent armor bonus, plus the class armor skill bonus, which together = the full AC bonus. The armor skill bonus only kicks in if it is not greater than the armor bonus. So, for example, a thief, who only has an armor skill of +1, gets no armor skill bonus when wearing chain, making chain equivalent in terms of protection to leather (though the thief can still wear chain if circumstances require it).Similarly, for example, magic-users benefit much less than fighters from plate, but a magic-user in plate is still better armored than a magic-user in leather. In effect, this system defines how much protection comes from just wearing the armor and how much protection comes from skill at fighting in armor. Some classes may get less benefit from wearing armor, but all penalties, whether they are encumbrance, speed reduction, thief skill impairment, spell failure chance, or anything else, apply in full. All else being equal, a thief will thus gain the same AC bonus from leather as they would from chain, but will take fewer penalties in the leather. The same thief will be slightly more protected in plate, though presumably at the cost of rather still penalties to sneaking and maybe fleeing (depending on the specific armor penalty rules that are active).

Potential AC bonus from a shield is equal to half the armor skill, rounded down, meaning that thieves (who have a +1 armor skill) and magic-users (who have a +0 armor skill) would gain no benefit from using a shield, and fighters (who have a +3 armor skill) would gain a +1 from using a shield (as would any class that was deemed to have a medium armor skill). This ends up outputting the original numbers, which is nice, but also means that it is impossible to gain any benefit from a shield without armor skill, and I’m not sure that is satisfactory. Provisionally, I think characters without enough armor skill to use a shield should be able to get the +1 AC if they spend their action focusing on using the shield to defend.

Incidentally, this same system could easily be used with Hexagram, substituting the path of battle “defense” trait (or something derived from it) for the armor skill bonus described above.

Abstracting missiles

26 Replies
Image from Wikipedia

Image from Wikipedia

Traditionally, combat rounds were one minute long, and a single attack roll determined the outcome of the entire sequence of feints, parries, dodges, and strikes that occurred durring that time. This is a simple system, and it works well with melee attacks, as they don’t use up any resources. The abstraction breaks down for missile weapons though, which require ammunition. Even if you shorten the round to 10 seconds (B/X) or 6 seconds (Third Edition), that doesn’t really address the problem if you want to keep combat abstract and fast-paced.

I posted a question on Google Plus about this topic, asking people for ideas about how to bridge the abstraction divide between melee and ranged combat. To be honest, tracking ammunition is not really that difficult, so this began more as an intellectual exercise. However, Jeremy M. of Over the Misty Mountains, suggested the following solution, which I quite like:

You could maybe resolve the problem by having several stages of depletion (maybe quiver depletion) and have them roll AFTER each combat. Making it more about how many they recover than how many they have. That keeps the resource management aspect but still skirts counting arrows

Here is a variation on his idea. I also like the option of allowing ranged attackers to spend more ammunition in exchange for a bonus or additional effect. It doesn’t complicate combat much (assuming the potential effects are simple), and gives players with characters wielding missile weapons an interesting choice to make every round.

Abstracting missiles

Collections of ammunition are tracked rather than individual arrows or bullets (quivers, cases, pouches, or whatever makes sense for the weapon in question). Ammunition does not run out durring combat (though see “volley” below). Instead, each weapon has an ammo die which is rolled after combat to determine if there is ammunition remaining. If the ammo check comes up 1, the current quiver is exhausted. The default ammo die is d6, though special weapons may use a different die. Rolling the ammo die also represents collecting any reusable ammunition, and the referee may assign penalties to the ammo check if recovering some ammunition would not be possible situationally (such as if firing across a chasm).

Ammo checks are not used for missile weapons that don’t have aggregate measures of ammunition, such as throwing axes or daggers. Depending on referee ruling, this ammunition rule may be used with bandoliers of throwing knives or shurikens. Referees may also opt to track ammunition more closely for special combats involving attrition, such as if characters are trapped in a foxhole. Special ammunition (where the rarity of an individual missile is important, such as an arrow of slaying) should be handled separately.

Volley

During combat, if using a weapon with ammunition, a ranged attacker may choose to expend more ammo in exchange for a bonus to attack or damage. This option requires an immediate ammo check, and if the ammo check fails the quiver is exhausted by the volley. This ammo check is in addition to the check required post-combat.

Because ammunition is still being tracked in aggregate units (quivers, clips, etc), resource management remains. An archer is potentially rewarded for thinking ahead and bringing extra quivers, though that comes at the cost of having less space for treasure or other equipment. Quivers, at least for arrows, are pretty bulky, and it’s hard to imagine carrying more than 2 or 3 and still operating efficiently.

Note that this also abstracts away reloading times. This is probably a good thing, as reload rates have far too large of an effect on combat efficacy. A two shot per round longbow just blows away a one shot every other round heavy crossbow, so consequently nobody ever uses heavy crossbows. Even if a heavy crossbow had significant bonuses against heavy armor (or something), the variance of the d20 means that you are probably better off with multiple shots and relying on lucky high numbers (though the exact probabilities will depend on other specifics).