So I am thinking about my own homebrewed system, and am wondering how to implement damage. I am considering two types of system here:

  • Damage is rolled separately (base on weapon), and success of the hit roll plays little or no part in determining damage (Most versions of D&D, Savage Worlds etc.)

  • Damage is dependent on how much you exceed your opponent's Defense (Fate, Chimera, Shadows of Esteren). In Shadows of Esteren, damage is your attack's roll + weapon damage - opponent's defense.

My questions are:

  1. From a game design perspective, is there a reason to favour one over the other?

  2. While basing damage on degree of success "feels" realistic, are there any pitfalls associated with it? Generalizations based on existing games which use this type of mechanic are welcomed.


Hit Points and Damage are abstract, as in D&D. Though I do not see the difference if Hit Points really represent bodily wound or just a combination of luck, skill and toughness. Specifically, I am interested in how the number is generated, not what it means.

What I am looking for

This is some analysis I haven't verified, and I am putting here as an example of what am I looking for.

Using "damage as another roll", two opponents with vastly different to-hit can still hit each other for the same amount of damage - the chance to hit, and the probability for high or low damage is decoupled. Say A has only 10% chance of hitting B, but do 2d10 damage. B may have 50% chance to hit A, but only still do 2d10 damage. Hence perhaps I could say that A has a ghost of a chance to kill B because each hit - if he does hit - he does 2d10 damage.

The second case, say A has only 10% chance to hit - and for simplicity sake, say for every 1 percentile he succeeded by (say roll d100 under chance to hit), he does 1 damage. Hence A cannot do as much damage to B as B will be to do A, so A is less likely to be a threat, even if he does hit. Unless the system has some sort of exploding dice roll. And B, going by the system, could hit with more damage,making the to-hit chance even more important.

Stuff like this.

  • 1
    \$\begingroup\$ Either works fine, althoug a mix might be better: degrees of success add to the damage roll. Do remeber that the more dice you roll, the slower combat becomes. However, a good test of a combat system is to model a boxing match. I got for Thai boxing myself. A vast majority of systems fail abyssimally at this simple test. \$\endgroup\$ – Sardathrion - against SE abuse Sep 11 '13 at 7:43
  • 1
    \$\begingroup\$ @Sardathrion Mm..this is probably tangential to the question, but what are the particular details of a thai boxing match that most combat systems can't handle? \$\endgroup\$ – Extrakun Sep 11 '13 at 7:45
  • 1
    \$\begingroup\$ @Extrakun: It is very tangential indeed but in a nut shell: everything! ^_~ It take way too long to run, mostly result in victory in the first boxing round due to fumble/critical, the damage inflicted is just wrong, and the flow of the fight is rarely right. This is generally because of the short combat round most game uses, lack of knowledge of trauma, and the failure to graspe statistics form the designers. Try to read the above with an open mind instead of going for the flame thrower... Thank you. \$\endgroup\$ – Sardathrion - against SE abuse Sep 11 '13 at 9:07
  • 2
    \$\begingroup\$ There is (at least) a third case that you have not considered. Some games use the same dice roll to hit and for damage: e.g. three dice of different colours added together to hit; if you hit, the black die is the damage. \$\endgroup\$ – sergut Sep 11 '13 at 10:25
  • 1
    \$\begingroup\$ There might be a fourth case: Rolemaster uses a lookup table of results, probably closest to a combined roll, but it plays quite differently to other combined systems. It avoids maths, but there is the cost of checking through tables of pre-decided results. \$\endgroup\$ – Neil Slater Sep 11 '13 at 14:20

10 Answers 10


There's really a wide range of options here. For example:

  • Roll "to hit" and use fixed damage.
  • Roll "to hit" and damage separately.
  • D&D-style: separate "to hit" and damage, but your hit roll can give you special effects like critical hits.
  • Straight-up "degree of success."
  • Degree of success with modifiers, like axe gives you +2 to final damage.
  • Degree of success translates into damage based on other stats or a chart.
  • Roll a bunch of dice and split them to determine the characteristics of an action.

Does the "Degree of Success" Approach Require More Arithmetic?

Not inherently.

In particular, if you use roll-and-count dice pools rather than roll-and-add math, then players can find the degree of success just by sorting dice side-by-side. This still takes time but isn't as sensitive to players' math skills or level of fatigue.

Does Using "Degrees of Success" Over "Hit, Then Damage" Restrict Tactical Options or Character Variety?

Not inherently.

For instance, if you want to create a niche for a big-damage weapon, it's as easy as "If you win the attack, +2s to damage." This weapon has a very different feel from a more "balanced" weapon with "+1d to attack and defense rolls." Ditto you can create a niche for a big-damage character by just writing something like "Add your Strength to damage when you hit." (Or how about "Margin of success add damage to your hit, up to your Strength number?")

One notable way that "degrees of success" changes back-and-forth damage math is that defense still benefits you even if it falls short of stopping the attack. Consider a system where we roll attack vs. parry — I roll 5 on my attack, you get 4 on your defense: in a "to hit, then damage" system, that's a straight-up failure for you and I get to roll my damage just as if we had 5 vs. 1; in a "degrees of success" system, you've turned a potential killing blow into a glancing one.

The "Realism" Question

Defining what's "realistic" is incredibly tricky here.

Smaller weapons aren't inherently more "agile," because you don't fight with a longsword by just sweeping it back and forth in wide arcs.

Real armor both deflects blows and dissipates some of the their energy. Weapons designed to defeat armor act very differently than those which aren't.

Realistically, a cunning thief with a dagger isn't going to be able to hurt an enemy in gothic plate with "fast" or even "precise" attacks. She needs to straight-up wrestle him into a position where that blade can actually do its job.

If you've chosen to use abstract hit points, your system is already privileging gameplay or genre considerations over "realism." Embrace that! Figure out how you want combat to feel and that'll drive whether you use separate or combined damage, static defense or opposed checks, &c. Just remember that a lot of what seems "realistic" based on other media is just a set of narrative conventions.

I Want Fast Resolution, What Should I Do?

I don't think there's a major difference between separate damage rolls or degree of success, at least compared to other factors like how many modifiers are involved or the size of a dice pool.

Really, though, if you want the fastest resolution of combat, decrease the number of actions involved overall (one opposed die roll with ten situational modifiers is still going to be faster than an entire combat encounter in D&D3/D&D4), and reduce decision points that are exceedingly opaque or offer only trivial benefits.


If damage is another roll, then the average damage you do will increase linearly with your chance of hitting.

If damage is only by degree of success, it'll scale quadratically, because the same modifier is used to determine both whether you hit and how much damage you do.

If damage is a mix of other modifiers and the degree of success, then surprise! It'll be a mix of quadratic and linear terms.


There's been a recent discussion on the Savage Worlds forum about this general topic (it's one that comes up there fairly regularly). Although many people seemed to like the idea of continuing to pile on damage indefinitely as the to-hit roll got better, the ultimate result ended up being an agreement that it's faster to have separate hit and damage rolls. Having to calculate exactly how much the hit roll was made by tends, for most groups, to slow the game down significantly.

(For context, Savage Worlds uses exploding dice, so you may have to roll multiple times to find out exactly how much a hit roll was made by. The standard rules do include a damage bonus if you make your hit roll by a good margin, but it's a one-time thing - once you've reached that margin, you get the bonus and can stop rolling.)


A combined to-hit and damage wraps up all factors into a single hit-and-damage potential. The game has assumed perhaps that being more agile is equivalent to being stronger, in terms of end result.

Whilst this may well work in the game system, and be as good as model as anything that uses Hit Points, by default it removes some elements of choice from the game.

For instance, where would Power Attack sit in such a game system - it works as a gambler's mini-game, and adds to excitement in combat - "Do I take a higher risk of missing for better damage, how lucky do I feel?". More subtly, character design based on trade-offs of accuracy versus damage is harder to implement. Neither of these are impossible with a single-roll system, but any implementation would generally have to add to the mental arithmetic load already present in the combined roll.

I am not saying by the way this is completely a bad thing - too much choice on minor issues repeated again and again is a good way to slow down any game. A character in D&D 3.5 who has Power Attack and Expertise, with a player not great at tracking all the -1s, +1s and +2s, and halving for off-hand weapon, might be a good example of this. Give them some Strength damage, and Bulls Strength to compensate, and figuring out whether you have a near miss or a hit isn't the most fun part of the game.

Another way to put this:

In your imaginary game design, you have two character concepts. One is "Conan: Powerhouse with Huge Sword", the other is "Errol: The Flashing Blade". They are both combat-centric characters, and have invested similar amounts into that side of their characters. In your game system, would these two characters play differently in combat, and if so, how?

A separate to-hit and damage system already has a way to differentiate the characters. Conan hits less often, but rolls bigger damage dice than Errol.

A combined system needs to either not do this (which is fine, especially if combat is a minor part of the game), or find another way to give the right "feel" to the two character concepts.

Both systems can build further with special effects and results.

  • 1
    \$\begingroup\$ For me, this is a good answer, because it states the implication of the mechanics - " The game has assumed perhaps that being more agile is equivalent to being stronger, in terms of end result." \$\endgroup\$ – Extrakun Sep 11 '13 at 16:01
  • 1
    \$\begingroup\$ Feng Shui (and several other games) use a system where damage is equal to the amount by which the to-hit roll succeeded plus a constant that's based on how much damage the weapon/spell/individual making the attack is likely to do, compared to other attacks. This allows for Conan and Errol to play differently, without requiring more than a single to-hit roll. \$\endgroup\$ – GMJoe Sep 12 '13 at 4:47
  • \$\begingroup\$ @GMJoe - Yes, but it involves bringing in and adding another number. You can broadly make both approaches the same by doing such a thing. However, there are stricter limits on to-hit variation in such a combined system, because you also have to allow for variation in defences, and the stage at which a "Conan" would get locked out happens faster. I'd guess Errol and Conan would have combat ratings maybe 14 and 12 in Feng Shui, with base damage 4 and 7? (Long time since I played) \$\endgroup\$ – Neil Slater Sep 12 '13 at 7:27
  • \$\begingroup\$ @NeilSlater Indeed, it does involve bringing an extra number in. Still, because under most circumstances the extra number is constant for a given character (Feng Shui cunningly applies all circumstantial modifiers to the to-hit roll) the slowdown is still pretty low, while still enabling Conan vs. Errol, which is perhaps a little closer to having your cake and eating it too. Wouldn't you have to account for variation in defenses in any system? \$\endgroup\$ – GMJoe Sep 12 '13 at 8:24
  • 2
    \$\begingroup\$ @NeilSlater ...And that brings us back to the point we share: the mechanics have to fit the 'feel' you want things to have in-game; Or, to put it another way, you need the things that are important and worth thinking about in the mechanics to match up with the things that are important and worth focusing on in the fiction. How Errol and Conan are differentiated - and what method of differentiating them is best - depends on what methods of differentiation you want to be important in the game. \$\endgroup\$ – GMJoe Sep 17 '13 at 5:08

Using "damage as another roll", two opponents with vastly different to-hit can still hit each other for the same amount of damage - the chance to hit, and the probability for high or low damage is decoupled. Say A has only 10% chance of hitting B, but do 2d10 damage. B may have 50% chance to hit A, but only still do 2d10 damage. Hence perhaps I could say that A has a ghost of a chance to kill B because each hit - if he does hit - he does 2d10 damage.

While this is true, it's only because they're both doing 2d10 damage, which given the hit chance difference suggests that B is simply stronger than A and should be expected to win more often. So I think that result is entirely predictable. With those numbers over the long term B is simply going to do far more damage than A.

In say 10 rounds of combat, B has a decent chance to hit 5 times, which is 10d10 total damage. A would need to get extremely lucky to come anywhere close to that. If A hits once (which is reasonably likely), they're doing 2d10. Given that B's damage output over the long term is 5x A's damage output, this isn't a fair comparison.

In order to make it fair, A would need to do 10d10 per hit. Then over a long combat their damage output would be more likely to be similar, except that A's will be spikey and B's will be more consistent.

The second case, say A has only 10% chance to hit - and for simplicity sake, say for every 1 percentile he succeeded by (say roll d100 under chance to hit), he does 1 damage. Hence A cannot do as much damage to B as B will be to do A, so A is less likely to be a threat, even if he does hit. Unless the system has some sort of exploding dice roll. And B, going by the system, could hit with more damage,making the to-hit chance even more important.

This example is actually the same as your other one. B is going to do considerably more damage than A, in this case because he has so much extra margin on his to-hit to add damage from it. In the previous case B was just doing damage far more often, but the result was the same. The main difference is that calculating how much damage you're doing is more complicated for the player in this system.

The other issue with adding bonuses from your attack roll to damage is that it becomes harder to model a scenario where A is big, slow, and wielding something like a tree trunk (which would be hard to hit with but would hit REALLY hard if it did), while B is small, fast, agile, and using a much weaker weapon. If B's chance to hit A is 90%, some of the time B is going to roll really well and you'll be adding a ton of bonus "to hit" damage, which means making this case where he doesn't hit very hard doesn't work very well. In this system a bonus to hit is also a bonus to damage, which makes it far better than a bonus to damage alone.

It's important to keep in mind that in either system, a bonus to your to-hit is adding damage in the end by letting you hit more often (and thus rolling your damage). Making a bonus to-hit also directly increase what you get to roll for damage in addition to just hitting (and doing damage) more often gives it extra effectiveness as a bonus as it's really increasing your damage twice.


In the long run there is little difference. Every combination of dice means a probability of damage and that is what you see over a long time (although at some specific events it could feel different). Edit: the answer by Tridus says this same thing going in detail through your example.

But gaming is more than just maths. Rolling more or less dice feels different, and has an impact on the gaming experience. Depending on the type of game that you want to end up with, one system may be better suited than other.

Damage as dice roll

This system means rolling more dice, which is good for some players and bad for others. The first advantage of decoupling attack and damage opens the door to tactical decisions: do you prefer to hit often for little damage or to hit fewer times but harder? Decisions like "sword vs axe" or "power attack" in DnD or "attacking body or head" or "Agility vs Strength" on Savage Worlds are part of what makes those games fun.

Rolling more dice usually results in longer combats, but many gaming groups counter this by rolling all the dice at the same time (hit and damage) and ignoring the damage dice when they do not hit. Some players like the increased drama of rolling more dice once you have hit.

Note than DnD3 and Savage Worlds are not pure "damage as dice roll" systems since both have core mechanisms to do more damage if you hit well (critical hits in DnD3, raises in Savage Worlds).

Pros: deeper tactical decision-making, more gaming drama.

Cons: longer combats (in real time, not in-game time).

Damage as degree of "hit success"

This system means rolling less dice but doing more mental arithmetic, which once again is good for some players and bad for others. Some people are not good at adding and subtracting numbers quickly. Of course, everyone can get better with practice.

It also makes the games simpler, not just faster. A good attack is a good attack, period.

Some players feel it is "more realistic" to do more damage if you hit better, but in the end it is just a way of modelling reality. You can model a pierced lung both as a very good hit or as a hit with a very good damage roll.

Pros: simpler games, faster combat (if you are good at mental arithmetic).

Cons: less tactical depth, some people are bad at mental arithmetic.

All in one roll

Some games (like those based on the Sombra system) use just one roll for everything. In Comandos, you roll 3d10 and you try to get over the difficulty to hit. If you do, the coloured die is your damage (unless it is heavy damage, in which case it is the other two).

In a way, this is similar to playing DnD and roll the d20 and damage dice together.

Pros: faster combat.

Cons: less tactical depth, usually requires different colours for dice to know which is which.


(As a personal note, I prefer faster combats and suck at mental arithmetic, so I roll all dice together.)

  • 1
    \$\begingroup\$ There are also systems where you choose your to hit and damage dice from the same pool, so that you actually trade off between spending your "good" dice on either success or damage. It adds a bit of tactical risk to constructing your result after the roll, which some people enjoy. It also gives you opportunities for interesting situations like "I'll sacrifice effectiveness of this attack because (for whatever reason) I just need to make contact", or "Because my opponent is large and slow but heavily armored, I don't have to worry so much about hitting, but I've got to make it count." \$\endgroup\$ – Zimul8r Sep 11 '13 at 13:22
  • \$\begingroup\$ That is very interesting. I do not know those games. Feel free to edit the answer or post an informative link and I will improve it. \$\endgroup\$ – sergut Sep 11 '13 at 15:10
  • \$\begingroup\$ Mmmm... a negative vote? Why? Is the answer incorrect or unclear? \$\endgroup\$ – sergut Sep 16 '13 at 16:15

As an alternative you could have a system similar to Silent Death where you have a target number to hit, and you roll dice based on your weapon as well as a die (or dice) based on your skill with it, and the to-hit roll is ALSO the damage roll.

If the dice total equals or exceeds the target number you hit.

If you hit, damage is based on the the dice and the weapon. So a dagger might do "low" damage, whilst a broadsword does "high" and a mace "medium". You even can still have crits when you roll doubles or triples, and Magic items could do either flat bonus to damage, or even do ALL damage.

Example, my fighter is skilled with a broadsword so he gets a d8. The broadsword is 2d6 Medium. I attack an armor 12 opponent and roll 3,4, and 5. That's a hit for 4 damage (medium die). If it'd rolled 3,5,5 it would have been 10 damage.

Note one peculiarity of this system is Medium damage weapons are more likely to crit than high or low damage (since the medium die is always part of a doubles roll).


Rolling damage separately and damage based on success is an important decision in game design, and I'll break it down into several categories for easier assessment.


Based on Dice:

Rolling attacks then damage quickly becomes a lengthy affair; it's doable, and at times the lack of math is preferable, especially if you're dealing with large numbers, though this will also depend on the number of dice versus their size; 1d20's easier to read and tabulate than 5d4.

Based on Success:

By far, adjusting damage based on success is quicker here. Some of this is math dependent, but my personal experience leans towards it being quicker to just apply the modifier than to have to gather up dice and roll again, even if you assume that you hit when you go to roll dice. That said, it tends to become more difficult in cases of large degrees of scaling (i.e. 10% per point over requires players to find 70%, versus having a table of three possible values), and the math can take time.


Based on Dice:

Results will always be within an easily determined expected range (i.e. d8). Optimal reliability. So long as a character hits, they're guaranteed to do a certain amount of damage based on their weapon type, so there's a lot less fluke involved.

Based on Success:

Results will vary wildly based on how well an attack connects. Not exactly a horrible way of doing things, but it can get out of hand, and since it relies on a prior roll, meaning that it's possible that a character who constantly barely succeeds will do nothing and a character who succeeds astronomically once in a blue moon can do epic levels of damage just based on how the dice fall, meaning that this only works well in a finitely scaled method or if you are fine with absurd scaling (e.g. a pistol conceivably taking down a tank) or implementing another system (i.e. something like hardened armor in Shadowrun).

Ease of Concept

Based on Dice:

Heavily system independent, but very dependent on individual cases. If you always roll the same dice, it'll be pretty easy to remember this, but if each weapon has different dice then the players need to note them down.

Based on Effect:

Ditto from above, unless you're working with an abstract system that doesn't care about weapons on an individual basis. In addition, direct linear scaling doesn't make sense (i.e. 1 success for 1 point of damage in Shadowrun), at least unless damage is "exponential" in effect (i.e. a 6 point damage source is not nearly as dangerous as an 8 point damage source), which means you should then integrate a resistance system which is an added degree of complexity.

Need for Assistive Mechanics

Based on Dice:

Pretty much nil. Your damage system can stand on its own as an independent system, like D&D's, and while it may need a separate set of handling compared to your core system it doesn't need any changes to the core system to reflect it.

Based on Success:

Either you need to intelligently scale your damage system or it's going to be difficult to mesh into the core mechanics well. In my ABACUS system I'm working on, I do this by giving each individual weapon a different amount of success-based damage scaling, which works well since there's no resistance test other than just dropping damage down the scale when an attack hits armor, but in a system that isn't aiming for one roll per action simplicity this is usually handled as a damage resistance roll separately, meaning combat can turn into a complicated affair.


You may be interested in looking into the Ubiquity System (Hollow Earth Expedition, Desolation, All for One). In this system, the attacker makes his attack roll & the defender makes a defense roll, the difference between these rolls is the amount of damage. The Ubiquity uses a skill point system with weapons adding to the roll (e.g., if your character had a melee skill of 6 and used a club 1N, the player would roll 7 dice to see how many successes; the defender with a defense of 4 would roll 4 dice. The difference in the number of attacker's success vs the defender = the amount of damage.)

Hollow Earth Expedition is a pulp RPG. Desolation is a post-apocalyptic fantasy RPG. All for One is a French Revolution RPG.

  • \$\begingroup\$ This is simply an example of a system, it does not address the pros and cons or pitfalls. \$\endgroup\$ – user1861 Jul 18 '15 at 15:04

I am sure you have thought of most of these things for a while, but I have some questions on terminology first.

The Terms

First, you should determine what damage means in your game. Do the characters have some form of hit-points? What do they represent? Is it a representation of the characters capacity to endure physical injury? Or is it how much energy/stamina the character has to fight before he can fight no longer? Or is it a combination of both?

Then, you need to think what does a successful attack mean. Can attacks "miss" their targets? Or are all attacks simply varying degrees of where you actually hit the target? Is a failed attack a bad thing?


In D&D attacks are "hit or miss" which means that a non-successful attack does nothing, while a successful attack causes physical damage. This is represented as a loss of hit-points. When the hit-points reaches 0 the character is incapacitated but not dead, and dies only at -10... Hit-points represent physical damage, and the system also includes "subdual" damage to represent non-lethal penalties to hit-points.

In the Game of Thrones tabletop RPG your hit-points represent your stamina in battle, not physical wounds. When the hit-points are reduced it does not mean the character received physical damage, only that his ability to continue fighting has been diminished. A sturdy blow against an enemies shield can reduce his hit-points if the damage was more than the shield can absorb. Reaching low hit-points induces penalties and physical damage can be caused by special attacks, or critical attacks, or if no defensive items are worn and no dodging is possible.


You need to figure out how you want the flow of your game to work. How many steps do you want each attack in combat to have? How many attacks (rounds) should combat take? Does it make more sense in the game for damage to be physical? Is there magic that can heal injuries instantly? Or do character need to rest in hospital for 6 months after every combat? These are important factors you need to think about.


You should do what fits most with the world you have created around your game. The mechanics represent the way you want the world to feel like, if you want combat to be a central part of the game it should not have draw-backs, A character who fights and incurs some damage should not need to heal for months at a time. So either there is magic and damage is physical and dangerous, or there is no magic and damage is less physical (of course, reducing an enemy to 0 and killing them would be) but physical damage should be more serious in a world without magic.

Personally, I prefer more realistic worlds and mechanics, but that may not fit your game.

  • \$\begingroup\$ Your conclusion makes more sense if the word 'healing' is inserted before each instance of the word 'magic.' More importantly, this answer doesn't answer the question, and could be compressed into a comment that says "We need more information about the role of combat and damage in your game before we can usefully answer your question." So... -1. \$\endgroup\$ – GMJoe Sep 11 '13 at 8:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.