About the D&D Combat System

This is a sort of sister thread for my look at Vancian magic from last week. Looking at what you actually do in D&D (generally, not specifically the modern take), this is what I get:

  • Plenty of freeform negotiation of situations (which I’ve sort of already dealt with last year in my discussion of challenge-based adventuring); despite some weak efforts to the contrary, the core D&D experience really runs on the basis of you-imagine, I-imagine, the result of which is a set of mutually accepted (credible, in theory-speak) challenge constraints that are then set in stone until the challenge is completed.
  • The magic system, which is the most important resource subsystem in the game. Increasingly so at higher levels, increasingly so in later editions.
  • The combat system, which is what you do with that positioning and those resources.

So it stands to reason that I’m interested in tackling the combat system now.

The basic system

Urban legend has it that the iconic D&D combat system was originally designed for modern era warship battles. This is sort of believable, looking at the dynamics.

Regardless, the system does have its strengths – it is not the most popular rpg combat system by accident. For one thing, the system is simple and quick to use. The great amount of abstraction is a big benefit, as is the resource attrition that allows combat to be expectable, controllable and survivable for the player characters. Almost all competing systems through the decades have provided “increased realism” (read: lethality) as their marketing point, so D&D stands in stark contrast to that: it’s procedural, easy and really just a most basic pacing device for something we could be resolving with just one die roll if we wanted to.

The heart of the D&D combat system is as follows, stripped from all extraneous elements: each turn each combatant chooses his target and tries to strike them. All combatants have a strike value and defense value, which are compared during the attack. A successful attack reduces the defender’s staying power. The combat ends when one party decides to retreat or is reduced to zero staying power.

This system is so familiar and endemic that we can’t even really appreciate it for what it is. I myself intensely dislike almost all applications of the D&D model of combat (except for computer games, in which I can stand it), but looking at the basic form, there is nothing objectionable in it: my dislike stems solely from the way the details are implemented, with weird weaponry and armour fetishes, vague rules for using terrain, efforts at tying abstract “hit points” down into some sort of in-fiction health tokens and so on and so forth.

Let’s look at how that combat system looks after we rework it into a modern conflict resolution system. I suspect that I’m going to like it.

Hit points

Hit points depict a character’s capability for continuing this fight. They’re an abstract measurement of combat experience, tenaciousness, preparation and so on and so forth – essentially, we justify hit points a posteori as anything that explains why the combat is not over yet. Hit points are specifically not something you can restore with magic, simply because there is no specific one thing to restore. Magic could give you extra hit points, but then that’d be because the magic gives you more will to fight, more energy, more of something that allows you to keep fighting.

Hit points are an abstract measure that determines when the combat starts causing consequences for you. A character with hit points won’t normally be injured seriously by the fighting – this is a cinematic concession, one might say. A character who loses his hit points does not benefit of this plot protection any more, and will be in danger of getting injured. As we’ll be seeing, the combat system is such that this is a quite dire situation: you’re probably not going to last many direct blows from that sword when it actually starts hitting.

Hit points are rolled at the start of combat. In my own homebrew sensibilities it’s likely that only fighters (which in my homebrew means “anybody with fighting experience, as depicted by their fighter class levels”) will gain hit points in every combat. Other characters might gain hit points if they surprised their opponent, had strong motivations for starting the fight or had some other conditions that favoured them in this fight.

Fighters would probably gain something like 1d6 hit points per level for my homebrew purposes, and of course they’d also get the same sort of additional conditional points the other characters could benefit from. Constitution or other abilities wouldn’t come into it (my homebrew doesn’t even have Constitution as an ability, now that I think of it), as they influence combat in other ways. I’m not so sure that I like that 1d6 thing, as I don’t know yet if I’m going to really use different die sizes enough to justify using them at all, as far as system elegance goes. I might go with +1 hit point per level just as easily, or even a pool of d20 equal to level, pick the highest. A system aesthetics issue, essentially.

An obvious example of a feat or special ability certain fighters might have would be to gain more or less hit points per level. Another obvious one would be a maneuver where the character gets to reroll his hit dice to try to improve them in the middle of combat. After all, a low-level fighter will be especially prone to swings in his hit point total from fight to fight, so something like that could be useful.

There is probably not much in the way of injury management in between fights. This is simply because a fighter who gets injured is not likely to be fighting another fight in true D&D style right afterwards – getting injured is serious, it might kill you. Losing hit points, on the other hand, is no big deal: those come back for each fight. It’s like those per-encounter powers in 4th edition D&D, one might say.

Combat rounds

Nowadays D&D is big on carefully lineated initiative order, but this has not always been the case. For the longest time D&D only determined initiative for each side in the combat separately, I seem to remember from my reading. On the level of abstraction D&D combat really moves, I find exact initiative order to be somewhat weird. I might well go for simultaneous action, in fact – just let everybody roll their dice and see what happens.

Doing simultaneous action allows characters to kill each other simultaneously, which is just fine and dandy with me. Surprise, which apparently was the big deal with initiative in the first place, can be handled as increased hit points for the surprising party, as to-hit bonuses or as extra rounds of action for the party with the initiative. No problem here.

In practice I’d likely steal a page from Trollbabe and have each round being with a “free-and-clear” phase during which the combatants decide what they’re going to do during the round. This can easily be done in public – if a player wants to react to something another player is doing, then let the two hash out their actions; if they get stuck in an infinite loop of second-guessing (which would be really rare), just have the GM decide that the two characters bash each other in the head. There’s not that much to decide on in D&D combat, anyway: you just need to choose who you’re going to whack, pretty much. Battlefield-positioning won’t be important for the wast majority of situations, you can whack anybody in the battle as far as I’m concerned. If the fiction has characters in hard to reach places or protected by others or some other such things, then I’d give you penalties to your attack check, simply enough.

One shouldn’t forget special actions, however; running away is a very important action in D&D combat, as it is the only way a traditional D&D game has for controlling the stakes in combat. D&D combat has traditionally been a bloodthirsty affair of kill-or-be-killed, and I probably wouldn’t change that a lot. The default consequence of losing would still be death, it seems to me.

But, special actions: my combat system wouldn’t mess with separate movement actions and action actions and free actions and all that 3rd edition shit – you pick one thing to do for one round, and that’s that. Running away is a fine special action if you don’t want to whack anybody, and there might be other things important enough to justify as combat actions. I’m considering surrender, myself: perhaps it could be mechanized the slightest bit consequence-wise, even, to make it more appealing. Spell-casting would be a combat action, too, I imagine – at least for spells intended to be cast that quickly.

D&D has long put stock on multiple attacks per round, and I sort of like the idea. Traditional implementations suck, of course, due to how they increase whiffing and unnecessary dice-rolling. But I’m just going to design around that because I have another special power I want fighters to be able to get: a feat (or some such) that allows the fighter to take actions equal to his fighter level divided by the fighter level of his enemies. So a 3rd level fighter fighting 1st level fighters could take three attacks, for example, while the opponents only get one. A 10th level fighter fighting a 5th level fighter would get two attacks against the opponent’s one. An 8th level fighter could take one action against a 5th level fighter and one action against a 3rd level fighter. That sort of thing, in any combination practicable on the field. I like this mechanic a lot because it scales: evenly leveled or nearly so opponents will always get just the one attack against each other, so the number of attacks will not get ridiculous in high-level fights.

Attacking and defending

For my homebrew purposes attack checks are ability checks just like everything else: Ability + d20. (Frequent readers will remember that I do Abilities as 3d6 in order, so most characters won’t have that many points of difference here.) Only fighters are getting experience-based bonuses on this check: +1 per level. Others might take some feat or such that allows them +½ per level or something, I don’t care.

The target number for the attack check is a bit more tricky. I’m quite partial to simply having the opponents compare their checks against each other and have the better check hit while the other misses. This is especially pertinent due to how I don’t much like armor and armor class: from my viewpoint that’s again weirdo aesthetics getting in the way of swingin’ fiction – the D&D combat system is much too abstract to have things like armor be the main determinant in whether you hit an opponent or not. I’d rather have armor provide more hit points.

Another option would be a special defense roll, but that solution sucks, too: too much dice rolling for little discernible gain.

Perhaps my favourite solution is to have the roll be against a semi-static difficulty level that might perhaps be changed for exceptional opponents. I’m thinking that rolling against 20 would be fine for my homebrew – amateur boxers can still hit each other half the time, even the most incapable characters have some chance to hit, and professionals or naturally talented characters will hit all the time. Perhaps another obvious fighter feat would be something that allows you to add your fighter level to the difficulty of hitting you.

(An obvious observation in choosing a static target number, by the way, is that characters might easily get into a situation where their actual bonus is so high that they’ll hit all the time. My system in general can handle this just fine: not only are hits scalable as we’ll see, but I’m probably going to cap that experience bonus for fighters at +5 or something as long as the fighter is not magical in nature. Plenty of ways to deal with static difficulties, and they have some philosophical benefits: by not scaling the difficulty of different accomplishments as characters increase in level, you actually allow the characters to genuinely progress in the game; I don’t much like the modern D&D style of just increasing the opposing numbers as the characters gain in power – why bother with gaining experience if you’re still going to be rolling a 50/50 check to do the same thing you did before?)

Ways for that striking difficulty to change… well, most things that affect combat are such that I’d rather make them give hit points, really. Increased defense would only be gained from some rather generic and generically powerful force-multipliers. I already mentioned the fighter’s experience level as a possible source of defense, and I could see some magic that gave defense, too. Also, targeted special attacks: should the opponent try something other than just decreasing hit points from the target, the difficulty level could be different. Instant decapitation could be at difficulty 30, for instance.

Perhaps an obvious action choice in combat would be going full defense, at which point the character could make an ability check to defend himself from an attack instead of trusting in the static defense modifier. Sounds reasonable to me. Perhaps make that “defensive fighting” and allow the defender to damage the attacker if he succeeds in his check much better than the opponent. Or just allow the defender to get more hit points as a reward for successful defense. These sorts of options might be important for high-level combat, if I’m not providing enough opportunities for increasing the difficulty of scoring a strike.

Hit damage

Now, hitting opponents is just as abstract in the core D&D system as everything else. That loss of hit points caused by a successful “hit” can be explained as many things; essentially, it just means that the striking character manages to somehow erode the fighting resources of the opposition. I’m specifically going to say that getting a “hit” in my homebrew won’t be the same as injuring the opponent in the fiction: that’s different.

How much damage? I’m tempted to get rid of a separate damage check, frankly: I never was very interested in the specific weapons that characters carry, just as I’m not interested in their armor. The background and skills of the character are much more interesting. So if I’m going to have some sort of separate damage check, it won’t be based on how large a weapon the character carries: such a set-up would make no sense at all when hit points are not about concretely hitting the opponent, anyway. That’s battleship thinking, and people are not battleships.

Instead, I’d most likely want to be simple and just let the hit point reduction be taken directly from the hit check: the difference between the check and the difficulty is probably a pretty good number to deduct from those hit points. It won’t be a big number for low-level fighters, but then such fighters won’t have many hit points, anyway.

I could also see the hit point damage getting multiplied by choosing to use some special maneuvers. Like, instead of taking the basic hit against difficulty 20, the character does something risky, pays 5 hit points of his own and gets double damage against his opponent. Or perhaps it’s a check against difficulty 25, but the damage is multiplied, so if you’re going to roll over 30, it’s a good deal. Whatever, lots of options for depicting different combat details.

Characters who go to negative hit points are nearing their end. The negative hit points by themselves won’t be a problem, though: the problem is that the next time the opponent succeeds in his attack, the weapon actually strikes home. This could be direct ability damage as I’ve done lately, but I’m actually sort of leaning towards separate consequential conditions nowadays: a first-degree success (0-4 points over the target number) allows you to give the opponent a wound that has some sort of consequence, a second-degree success (5-9 points over the target) strikes an incapasitating wound, a third-degree success kills outright. That sort of thing. Essentially, not having hit points opens the character up to actually dangerous attacks. The difficulty of this check might be different, too: perhaps it’s an opposed check finally, as it’s the opposing character’s life on the line. Maybe the opponent actually can’t act anymore normally after losing his hit point buffer, and he now has to take these defensive actions to keep himself alive while hoping to score a high enough success to get some more hit points – after successful defense gave him back enough points, he could go back to the normal combat mechanics.

Fighters as a character class

It’s pretty obvious that in my homebrew fighters are the ones who rule the roost on the battlefield: they get the experience bonuses to hit, they get the extra hit points, they get the extra attacks and so on. This is just fine with me: I don’t much like the way all characters improve in their fighting skills in traditional D&D implementations. Much more fun from my viewpoint if the player has to balance a multiclassing conundrum: if he wants his character to actually do anything in combat, he better take some fighter levels. That’s what it means to have combat skills, taking fighter levels.

(For fictional purposes, note that what I’m calling “fighters” can be whatever martially inclined character classes I’d care to have in my campaign. So “barbarians”, “ninjas” and any other things that train for combat would get some subset of the various benefits I’ve listed here.)

The legacy

If this system seems just a bit like the excellent Glorantha roleplaying game Heroquest, then that’s because the Heroquest complex conflict system is actually just the D&D combat system taken through this same wrangler. One would think that the game of the decade, the masterpiece of Robin D. Laws design would make a complete break with the mechanical principles of old games, but actually D&D already had a rather elegant conflict system lurking underneath all that detail. Isn’t roleplaying interesting?

Another system that is very similar is of course Solar System, which was created by Clinton R. Nixon as a sort of D&D love letter anyway. In general, all rpg conflict systems that feature ablative buffer resources that characters whittle down in an effort to overcome their opposition are in many ways just elaborations and streamlining on D&D’s quirky battleship-sinking rules. One would be justified in saying that those rules capture something rather basic about dramatic pacing – the most important success of the D&D combat rules is that they tell us in clear terms how long the fight is still going to last, how well individuals are doing and when we should add another snippet of fiction in our unfolding fight choreography. Just don’t try to use the hit points as fictional representation of getting hit with weapons – that makes sense for ships with thick hulls and redundant systems, but humans are rather more frail. D&D combat went to hell in a handbasket when hit points were equated with physical well-fare. (Pretty sad that this apparently happened with the very first versions of the game, which already included such pearls of wisdom as the “Cure Light Wounds” spell.)

7 Responses to “About the D&D Combat System”

  1. Marshall Burns Says:

    “Hit points are an abstract measure that determines when the combat starts causing consequences for you. A character with hit points won’t normally be injured seriously by the fighting – this is a cinematic concession, one might say.”

    Hey, that’s the way The Rustbelt works, with Blood, Sweat, and Tears in the place of hit points. And with the distinction that you can opt for actual injury at any time to save your buffers for later.

    Actually, there’s a lot of this that is much like The Rustbelt. Except that battlefield positioning matters in The Rustbelt, despite the lack of a gridmap.

  2. Johannes Kellomäki Says:


    I love your work on this primitive DnD. I think that your idea of hit point gambling is an exellent one.

    Did I get it right: Lets assume that a hit causes normally one point of attrition (“damage”). If you choose to gamble one of your own hit points when you hit, your blow would cause three points of damage – if it strikes home. If not, your gambled hit point would be lost.

    I guess you could gamble your character to negative hit pointsif you so choose. That would just affect your injury checks or something.

    Initiative could be an issue if gambling is not limited.

  3. Eero Tuovinen Says:

    That’s a simple method, yes. I’m not particularly worried with the specifics at this stage – my play has lately been pretty loose on the rules, as I try to hone the method and principles; the specific rules are an afterthought, really.

    I could imagine binding the hit point thing to some sort of in-fiction stuff or requiring characters to do some special maneuvers to gain access to those gambling options, but the simplest option is certainly to do it like Heroquest does: just have the player decide for himself how decisive a maneuver he’s going to announce, and gamble a number of hit points on that basis. This has the benefit of letting the player sort of control the pacing of the fight, too. Probably too simple for a challengeful game, though.

  4. K. Bailey Says:

    Interesting article. I hadn’t heard the urban myth about battleships, and I don’t really believe it, because the descent from Chainmail is easy to see. But I’ll grant it as a rhetorical device.

    One of the things D&D combat has to handle is an insane amount of variety. Large amounts of opponents on one or both sides, strange situations, monsters of all different kinds, magical effects. Designing a system around the expected case of small numbers of human-like combatants of normal-to-expert skill may result in situations where the system seems arbitrary. You can special case them as they arise, which is fun enough, but then those special cases accumulate…

  5. Abkajud Says:

    Very interesting post as usual, Eero! 🙂

    I’m intrigued by the apparent similarities between this D&D homebrew of yours, Rustbelt (judging solely from the above comment that references it), and my game, Mask of the Emperor – specifically, I really like the idea that the resources being spent, depleted, frittered away, etc., in combat play represent advantage, rather than Life Force.

    Certainly, for a generation raised on video games (mine and yours, I imagine), it’s intuitive that hit points represent your actual, physical state. But what if we prefer a little realism in the narrative, at least a *kind* of realism, one that doesn’t include getting hacked to pieces during every single fight, only to patch it all up instantly with a health-kit/life potion/buff-granting foodstuffs?

    Good stuff. I hope to run across a more detailed collection of these ideas one of these days, that I might steal it for my purposes!

  6. ChristopherA Says:

    I like your attempt to make a system that actually follows the old D&D idea that hit points are abstract and don’t refer to physical wounds. I’ve been meaning to do something like that myself.

    I always thought it was strange that 1st edition D&D had one small rules section claiming that hit points didn’t represent real damage, but for all other purposes they were, in fact, treated exactly like real damage. You example of Cure Light Wounds is easily excused (it is really healing your scratches and giving you a non-physical boost). It was the fact that, without magic, wounds would take weeks to heal. Or getting hit with a harpoon, and being tied to the harpoon rope until you took damage removing the harpoon. Or getting hit with a fireball and surviving, despite having the metallic equipment you are carrying melted by the heat. Huh?

  7. Kari Says:

    I realize this is a very, very old thread, but I’m going to comment anyway! If hitpoints are an abstraction of fighting ability in general, why even bother with attack rolls? Wouldn’t hitpoints be the ultimate abstraction of armour, endurance and dexterity alone? Just roll until you reduce HP to zero and then start causing some real injuries. Doesn’t really make sense to abstract everything but then leave the chance to hit as some weird arbitrary number.
    Would make combat kind of like a dice based game of Black Jack I suppose, where the better you were at fighting, the better your odds would be.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: