Elliot Wilen ([info]ewilen) wrote,
@ 2008-11-04 23:23:00
Previous Entry  Add to memories!  Tell a Friend  Next Entry
Making D&D combat realistic
This post comes out of thoughts in this thread over on theRPGsite. Basically, the question was which RPGs offer a realistic model of combat, and based on the way the question was phrased, I took this to mean that the model produces results which, when interpreted at the level of abstraction of the system, are commensurate with the inputs.

I wrote,

Well, in combat between humans, without magic [...] if you take the relative experience, armor, and weaponry of combatants into account as inputs, D&D will yield a result that's roughly commensurate given the abstract nature of hit points. In a melee you will end up with one man dead, surrendered, or fled, and the other victorious although possibly weakened somewhat. There are some areas I would call into question:

• Shields are relatively undervalued in terms of their overall contribution to survival. The relative significance of other types of armor compared to each other could also be questioned, but it's not egregious.

• The value of experience also seems to be distorted. At the low end, the likelihood that a trained but "inexperienced" (1st-level) warrior will die in a stand-up fight with an untrained civilian seems rather high. At the high end, it is possible for a fighter to outclass even an experienced opponent to such a degree that there's effectively no doubt as to the outcome.

[...]

Another quibble that's not in quite the same class is the handling of missile weapons. While it's easy to rationalize the loss of hit points by the victor in melee, it's pretty hard to explain the ablative "weakening" of a combatant by arrows that narrowly miss or glance off armor (i.e., cause HP loss but don't kill). And this has another significant effect on the outcome in that archers expend arrows as they attack.

Nevertheless you could still argue that D&D is more realistic than many other games, since those are far more likely to produce outcomes where combatants unambiguously suffer major wounds yet keep fighting and defeat their opponents. [...] I think that most actual wounds inflicted by melee weapons will either be fairly insignificant or will effectively take you out of the fight altogether.


Later I wrote that the main obstacle to achieving "verisimilitude" in D&D was the use of abstract hit points to represent "survivability".

So this has me thinking what might be the minimum changes I could apply to the basic D&D combat system (I mean pre-AD&D 2e) to improve its verisimilitude and realism. These are just some ideas, written up in the spirit of keeping the game simple, and also without too much concern for "fun". What I mean by that is, I think most RPG players dislike high lethality, especially for "high level" characters. These rules changes by contrast make the game a lot less safe; they might be more acceptable as a set of skirmish rules than as an RPG module.

#1: Instead of lowering AC by 1, shields lower AC by 5. This is a rough value, and can also be modified to represent large vs. small shields.

#2: Characters start at 2nd level and cap out at 5th level. Again, this is a rough limit. Furthermore, if you also follow my recommendation of doing away with escalating hit dice/level, it may not be necessary to implement this change. If you do still use hit dice/level, I'd recommend giving max HP (+ Con bonus) for the first die, then average HP for each subsequent die (with or without Con bonus). Otherwise, if you want some randomness, then consider a method that has a lower variance than just rolling a single die. However, if you want to be really hardcore, go ahead--the fact a some fifth-level fighter could end up with 7 HP simply reflects the fact that life is unfair and "living up to your potential" might still mean you suck.

#3: Probably not necessary if you use #4, but an alternative missile weapon system is needed. I can't be bothered to develop the details right now; the basic idea is that if you are hit by a missile, you will be at least significantly wounded regardless of your hit points. It could be a chart per weapon, or some kind of saving throw after you compare rolled damage vs. Constitution, and results could be expressed in terms of % of starting HP lost, or direct description ("no effect", "incapacitated", "dead").

#4: This is really the most interesting bit...it may also be stolen from somewhere (Pendragon?). The basic idea is that survival is achieved by not being hit, and in a normal combat, one maneuvers to an opponent's disadvantage such that, when all is said and done, striking your opponent generally means that you are not struck back. Mechanically, we do it this way: do not add hit dice when going up levels. Then, when in combat, declare a target and roll a d20. This die is both your attack and defense. In order to hit, you must roll greater than the number needed to hit the target's AC (as in the standard combat tables, or via THAC0 calculation for you newfangled types). But if two people are facing each other and both hit, only the one who rolls lower actually hits. (If they roll equal, both hit.)

("Rolls lower" seems right, but "rolls higher" might also work.)

Note that if you don't want to attack, you don't have to; your die is just your defense in that case; your opponent must "beat" your die (roll lower--or higher) while also beating your AC. I haven't given much thought as to whether an "all out dodge/parry/block" needs to added; in principle, under the model I propose, you could argue that if you aren't threatening your opponent, he will have an easier time striking you. But conversely you could argue that people sometimes fight without regard to preserving themselves, and if you're up against someone like that, you'll be less likely to be hit if you go into full defense rather than threatening them.

It could even be that you might get a useful effect from allowing players to declare before rolling that they're taking "high hits" or "low hits"--meaning that they'll hit when rolling higher than the opponent (provided they also hit on the combat table), or they'll hit when rolling lower than the opponent (provided they also hit on the combat table). I haven't thought much about this.

Another possible improvement would be to implement a 3e-like concept of "base combat bonus" in lieu of the columns in the combat tables or increasing THAC0 as you go up levels. The main benefit here would be that you'd apply the bonus to your combat die (as a subtraction on defense, if the opponent is trying to roll under, but as an addition for all other cases including attack). The purpose of this wouldn't be so much to move to a 3e-like system of target numbers and "better armor means higher AC"--although you could rejigger the math in that direction, to make things simpler--but mainly to give a defensive boost to more experienced combatants.

Obviously this is very raw, and I'm not sure I'll ever do the calculations/testing to hammer it into shape. But I wanted to get these ideas down before I forget them, and anyone else who wants to play with them might get something out of them.



(6 comments) - (Post a new comment)


[info]jamused
2008-11-05 01:41 pm UTC (link)
If you're trying to change the abstraction from survivability (that which makes veteran troops more likely to prevail over green troops given the same equipment and leadership) to physical damage sustainable... I don't think that's particularly realistic. My understanding is that even to this day most combat is decided by failures of morale rather than actual injury, and certainly in pre-modern combat most deaths were the result of what happened after combat (infection, disease, etc). Even routed troops could mostly be reassembled after the battle (sometimes even during). Troops were destroyed primarily by breaking their morale so thoroughly that they deserted, or by trapping them so that they could not retreat and then slaughtering them even though they were no longer capable of fighting. Quibbles about the effectiveness of shields aside, I think D&D (pre-3.x anyway) does a reasonable job of modeling that: both sides in the battle have their combat readiness degraded and the losing side will have given up the fight and retreated or surrendered after a few deaths have shaken their morale (a big part of the older D&D rules, at least for NPCs and monsters).

Without actually playing a bunch of combats on the surface it seems your revisions make combat much more lethal, but that survivors are more likely to be completely ready for another action, and that disciplined elite troops are reasonably likely to be defeated by an equal number of rabble and virtually certain to be overwhelmed by any more numerous foes.

(Reply to this) (Thread)


[info]ewilen
2008-11-05 09:50 pm UTC (link)
Thanks for the comments. I understand the issues you raise, but morale is something I would layer on top of the combat system, exactly as is done with B/X and AD&D1e (and probably other versions of D&D).

When I wrote "survivability" I was referring only to that which is handled, in the standard system, by giving more "experienced" combatants more hit dice. This serves in lieu of making them mechanically "harder to hit"; the only other defensive advantage a high-level character has is that he's likely to hit his opponent before his opponent hits him. The purpose of option #4 is to fold this factor into an abstract representation of the fluidity of combat. By getting better at combat, you lower the minimum you need to roll to hit a given AC, which simultaneously reduces your opponent's chance of hitting you--since there are more numbers that you can roll to nullify his attack.

Also note that these are skirmish-level rules that don't consider formations; your comments are applicable at a much different scale.

That said, you may have a point about the influence of lost hit points on morale. Under the standard rules this could be seen as a cumulative disorientation at the individual level, which then feeds into morale checks. It doesn't work quite as well for PCs as NPCs, though, since as you note, breaking is generally bad for survival; if PCs aren't subject to morale checks, they'll only retreat when doing so maximizes their chance of survival--which runs counter to the rationale for cumulative hit points.

To address this, we could start down a slippery slope of complication, hopefully catching ourselves before it gets too extreme. Off the top of my head, what I'd propose is giving each combatant a number of "fatigue/disorientation" points, maybe based on Con, Wis, &/or Int. Melee damage is subtracted from these until they're exhausted. At that point, you suffer a penalty on combat rolls, and any future damage comes directly from hit points. Damage also goes directly to hit points if it comes from missiles, or if an opponent achieves a "critical" hit. (This is cribbed somewhat from SPI's Dragonquest.)

Morale checks for NPCs are then triggered and modified based on both loss of "cool" and actual injury. PCs still won't break "irrationally" but there will be an incentive for losing characters to try to break contact, as well as for winning characters to pursue. (I'm assuming here that disorientation is recovered at a fairly rapid rate.)

it seems your revisions make combat much more lethal, but that survivors are more likely to be completely ready for another action
Yes. This comes from the fact that emphasizing "verisimilitude" makes wounding a pretty much all-or-nothing affair. The generic, abstract "weakening" of an individual which the standard rules handle through hit points is lost. If you partially add generic "fatigue/disorientation" points back into the equation, this is mitigated, but now the trick is to balance the numbers.

and that disciplined elite troops are reasonably likely to be defeated by an equal number of rabble and virtually certain to be overwhelmed by any more numerous foes.
I think the former is incorrect especially if there's a difference in armor--chain or better with shield, vs. nothing or leather with or without shield, ought to be a fairly large advantage. It's also a question of what you consider to be the minimum level of a professional soldier compared to random civilian; I may have put it too low at 2nd vs. 0/1st level. As for the second, armor will still make a big difference; giving a defensive bonus for experience (2nd-to-last paragraph above) will also help. But ultimately I believe that numbers matter a lot, especially at the skirmish level, where formations (and therefore disciplined maneuver) don't really apply.

(Reply to this) (Parent)(Thread)


[info]jamused
2008-11-06 08:15 pm UTC (link)
Well, I was thinking of rabble that had been similarly equipped to the veterans (since it's obvious that AC matters)...that is, why go to the expense of training men-at-arms when you can grab peasants and outfit them in chain and shield for nearly the same results? D&D makes the most sense to me when you think about it in the context of a war-game reduced to the scale of individual action (as I understand it, back in the original D&D campaigns that Gygax and Arneson ran players actually would take enough troops as hirelings to fight in formation into the dungeons). It leads to some anomalies, but I think I understand the abstractions chosen.

As for the fatigue/disorientation vs. physical damage, that makes a lot of sense to me; I just accept the level of abstraction where HP represent both (the first die being actual physical resilience, everything after that more on the order of mental toughness, alertness and resilience) so that the difference between veteran and green fighters is how long you can keep them in the field at reasonable effectiveness. I've talked a bit on that notion in this post. That said, I don't actually run D&D. My current system of choice is Savage Worlds, which has a completely different abstraction.

I've a few more thoughts on realism, but this comment is getting long so I think I'll post them on my blog.

Edited at 2008-11-07 04:55 am UTC

(Reply to this) (Parent)


[info]r_earley_clark
2008-11-07 05:31 pm UTC (link)
"#2: Characters start at 2nd level and cap out at 5th level"

Without messing with too much else ( I would favor experience falling into a more _Forward...to Adventure! style to go with it), I think you'd manage to grab a good bit more realism.

Another option might be allow a character to choose some sort of ability at each level, but that begins to seriously alter what the core rules are. Here I'm thinking "spell level or hit points to take at this level? Or maybe I want an inherent armor class reduction (better defensive fighting) instead of increased to-hit or damage".

Darn you Elliot, now I'm flying off into Heartbreaker-ponderations... ;)

(Reply to this)

Keepin' it Real
(Anonymous)
2009-02-04 11:05 pm UTC (link)
The combat systems in pen-and-paper RPGs as pertains to combat are very crude abstractions of reality. The best to be done would be in a videogame. In Spore procedural algorithms animate creature's movements based on the body parts they're given. Similar algorithms could calculate the paths of weapons into bodies, organs, etc. to determine damage. Each body would be its' own set of rules (liver does ____, adrenaline is released [condition]). Such a database of functions need only be very simple, nowhere near a real body's complexity but it'd make the game more believable. Imagine chopping off a goblin's arm, ogres going into shock from blood loss after their thigh arteries are punctured, watching an orc asphyxiate from an arrow to the lung, or a gutted foe nodding off. On the flipside: light cuts on a character could make him move faster and stronger temporarily from adrenaline and a protracted fight would leave poor players with a slow, weakened, slick-gripped PC sweating from exertion. Magic could easily be implemented as equivalent to physical actions. Casting a more powerful spell would be like lifting an opponent above your head to throw him off a cliff and lighter spells would be like sending a blow w/full bodyweight behind it. Because magic effects are more stupendous than the mundane magic-users would be for high skill players; a mage has to make each spell count b/c he burns out faster than fighters.

(Reply to this) (Thread)

Re: Keepin' it Real
[info]ewilen
2009-02-05 02:08 am UTC (link)
If you want to go down that path, sure. But...

1) Detail ≠ realism.

2) If you haven't read Harnmaster, or the various Leading Edge games (Phoenix Command, Rhand, Living Steel, Swords Path Glory etc.), you ought to familiarize yourself with them and incorporate that knowledge into your overview of RPG combat. Basically: what you're suggesting has been tried in P&P games, and it works more or less well if the group can tolerate the complexity.

Computerization is an option but for the sake of argument I assume it's not desired; among other things, many people like to have the algorithms be transparent so that they can strategize during play and tinker with the rules at other times.

(Reply to this) (Parent)


(6 comments) - (Post a new comment)

Create an Account
Forgot your login or password?
Login w/ OpenID
English • Español • Deutsch • Русский…