[theory] Question About a Combat Mechanic

[Amphetryon]

Note: Mods please move this to another forum if need be.

As I suspect is the case with many here, I am working on a fantasy RPG design in its infancy stages. I am currently working on the combat section. I am trying to find another game system - if such system exists - where combat resolution works as follows:

a)Combat is resolved by both combatants rolling an equal number of dice and

b)Combatants declare before rolling which, and how many, of those dice are assigned to 'offense' and 'defense'. Such declarations should be simultaneous to avoid gaming the system in that aspect. If the dice are in some way already indicative of their offensive or defensive function, so much the better.

For example: Pat and Robin meet on the battlefield, swords drawn. Both get five d10s for combat dice. Pat writes down 4 offensive and 1 defensive dice, while Robin writes down 3 offensive dice and 2 defensive dice. Pat rolls 6 5 4 2 for offense and 3 for defense; Robin rolls 10 9 1 for offense and 8 6 for defense. Paired numbers - the 6 rolled on offense for Pat and defense for Robin - cancel out; Pat's 5 for offense is beaten by Robin's remaining 8 for defense, and Pat's 4 and 2 get through unopposed. Robin's offensive 1 gets beaten by Pat's defensive 3, while Robin's 10 and 9 get through unopposed.

[Psyx]

There are three to my knowledge that use a similar mechanic. all of them are awesome :smallbiggrin: :

1) Riddle of Steel essentially works simultaneously and each combatant rolls multiple d10 dependant on their skill, with dice being kind of rationed for the turn between offence and defence. [this is a gross over-simplification]

2) Pendragon sees both combatants roll 1d20 with the one getting highest but still under their skill roll getting the hit in.

3) My home-brew system gives combatants a number of dice equal to their skill level which are split between offence and defence and rolled simultaneously.

[Glimbur]

My first concern with this approach is that it has no provision for a difference in skill. That is easily corrected. You might also suggest having two colors of dice, one for offense and one for defense. That would make life easier.

I seem to recall White Wolf games using a dice pool which can be divided between multiple actions but I do not recall specifics, sorry.

[Amphetryon]

My first concern with this approach is that it has no provision for a difference in skill. That is easily corrected. You might also suggest having two colors of dice, one for offense and one for defense. That would make life easier. <snip>


While the thought process on this mechanic is still very much in raw form, I was envisioning the difference in skill reflecting in damage dealt and deflected. I agree "difference in skill" is a necessary additional parameter. The last sentence of point A does suggest that some way of telling which dice are offense and which are defense - like color-coding for convenience - is a good idea, or at least it is intended to suggest as much.

To further expand on the mechanic, how should such a combat mechanic - if it exists - deal with a scenario where one side is outnumbered? What happens if Kim is fighting at Pat's side? Should Robin get an addition 5 combat dice to deal with Kim, essentially making it 10 dice vs 10 dice? Should Robin have to split the 5 combat dice amongst Pat and Kim? Should the overall number of combat dice that Pat and Kim use to fight Robin be diminished to reduce the imbalance that might otherwise be present? Something else?

[Person_Man]

The concept reminds me of Dragon Dice (http://en.wikipedia.org/wiki/Dragon_Dice). It's been years since I've played, but basically two players roll dice at the same time, and there are many different (collectible) dice to represent different things.

Instead of a strict offense/defense, I would suggest an Earth/Wind/Fire/Water or Rock/Paper/Scissors/Lizard/Spock type mechanic. (Strength/Dexterity/Guile?) If you have only 2 types of dice, there's limited strategic complexity.

[Oracle_Hunter]

Differences in Skill can be best modeled by using different DCs and using a "Success Test" mechanic rather than high-low comparisons.

For example:

- Assume that the base DC for combat checks is a roll of 10 on a d10
- Stats (e.g. Offense, Defense) reduce the DC for succeeding on an Offensive or Defensive check by their rating
- Each side rolls 10 total d10s each round of combat

Bob the Bruiser has Offense 5, Defense 2. When he rolls an Offense Die he only needs a 5+ to succeed (10 - 5 = 5) but he needs an 8+ to succeed on a Defense Die (10 - 2 = 8).

Denny the Defender has Offense 3 and Defense 8; he needs 7+ and 2+ respectively.

Bob decides to go all out - 9 dice on Offense, 1 Die on Defense. He's good at what he does and he really wants to overpower Denny's known skill.

Denny, however, is going for a more balanced strategy - 6 on Offense, 4 on Defense.

Bob rolls:
OFF 8 10 10 8 2 10 7 10 7 [8 Successes (!)]
DEF 5 [0 Successes]

Denny rolls:
OFF 2 10 4 10 4 4 [2 Successes]
DEF 6 1 2 3 [3 Successes]

The pair then compares their OFF v. their opponent's DEF

Bob got 8 OFF which is then subtracted by Denny's 3 DEF - 5 hits get through on Denny. Ouch!

Denny got 2 OFF which is subtracted from Bob's 0 DEF - 2 hits get through on Bob
By fiddling with the total number of Offense & Defense ranks a character has you can make them more or less skilled in combat. And, by splitting combat skill into two parts (Offense and Defense) you can craft characters who are trained differently even if they are of the same skill level.

EDIT: How you deal with multiple people in a combat depends on how you want to handle combats generally. If mass combats are the norm, then you can have each member can roll individually and then you compare the overall number of successes and failures. If you want a more individual-focused game, then it makes sense to penalize people who decide to "aid" a battle rather than fight their own. Let the assistant roll normally but only be able to contribute half of their total Successes (for each category) to their ally's total.

[Psyx]

I use different coloured dice. You can reflect skill type by giving more dice, by changing some kind of target number for successes, or by even using differing die types.

As regards multiple foes, I make players split their pool. 4 dice, 2 foes at once... put one dice into defence against each, and have two dice left to attack one of them, for example.

[Britter]

Burning Wheel uses that approach with certain parts of it's combat mechanic, though not using equal pools.

[Amphetryon]

EDIT: How you deal with multiple people in a combat depends on how you want to handle combats generally. <snip> If you want a more individual-focused game, then it makes sense to penalize people who decide to "aid" a battle rather than fight their own. Let the assistant roll normally but only be able to contribute half of their total Successes (for each category) to their ally's total. Assuming this is the approach favored, should the outnumbered combatant have any further modification to the "dice pool" when facing multiple foes, or is the reduction in "aid-related" dice added to the total the only modification you think necessary?

[Oracle_Hunter]

Assuming this is the approach favored, should the outnumbered combatant have any further modification to the "dice pool" when facing multiple foes, or is the reduction in "aid-related" dice added to the total the only modification you think necessary?

Oracle Hunter's Second Rule of Game Design
Never design two mechanics that serve the same purpose

A variant of the KISS principle (http://en.wikipedia.org/wiki/KISS_principle), it never helps to add an extra modification for a given event when it doesn't change the net effect.

Here, being outnumbered is supposed to place the the other party at a disadvanage. As written, my proposal provides a mechanic for doing so. It is always better to simply intensify a mechanic than to add a second mechanic that serves the same general purpose.

So no. If each additional ally in the scrum provides additional dice to the roll then the party without extra allies will always be at a probable disadvantage. You can tweak the mechanics to cause those allies to provide greater or fewer dice later depending on how useful you want "having numbers" to be.

[AyeGill]

I would give a small bonus to lone fighters against multiples, but not much. I'd say that, assuming the same system that Oracle_Hunter proposed, you choose a primary target of your defense(and in the case of more than two opponents, a secondary, tertiary, etc). You subtract you DEF from his OFF, and any leftover DEF, plus an amount of extra dice equal to your defense skill, gets compared with the secondary target. Repeat for tertiary, and so on.

Note: in the case of very high DEF skills, you might actually get a better defense on your secondary than on your primary. Maybe lose a dice from each subsequent defense for each hit you take?

[nedz]

If you havn't already, then you should check out the Forge (http://www.indie-rpgs.com/articles/)

[Oracle_Hunter]

The next question I have to ask is : what are you trying to accomplish with a system like this? :smallconfused:

[Amphetryon]

The next question I have to ask is : what are you trying to accomplish with a system like this? :smallconfused:
I'll try to explain some goals of this. Let me know if further clarification is needed.


Create a more dynamic defense than "AC" is in d20, while being less "swingy".

Avoid the issue I've had with past iterations of Shadowrun (I've created a character for the latest iteration but the GM declared the system 'too lethal and therefore borked' before we got underway), where you needed a bucket to hold all your d6s.

Create an attack mechanism and defense mechanism that, at least at the start and at least in theory, balance against each other. By keeping the number of offensive and defensive dice as a constant, I hope to eliminate scenarios where fights have essentially predetermined outcomes because you're throwing 18 dice and your opponent is only throwing 7, for instance. Conversely, the choices made for dice to put into offense or defense, respectively, hopefully allow a character to be more aggressive or more reactive, as befits the player and concept. Damage and skill differentials - as yet largely undefined - can still weigh a fight one way or another, of course.

[Oracle_Hunter]

Fair enough.

Well then... what about the system I've come up with? Thoughts?

[stainboy]

I'd suggest you have one defense pool apply against all attackers. Alice is fighting Bob and Carol. Each combatant has 5 dice. Alice splits 2 offense, 3 defense. She rolls 3 defense dice against Bob, and 3 defense dice against Carol.

If you want to add a slight advantage for assist-training, you could have the defender roll once and apply that roll against all attackers, but an equal result from any attacker removes that die against all attackers. Again, Alice is rolling 3 defense, and we'll say Bob and Carol each decide to roll 2 offense.
Alice (defender) rolls 1, 3, 6.
Bob (attacker 1) rolls 2, 5.
Carol (attacker 2) rolls 6, 9.
If Bob or Carol were attacking alone, Carol would hit and Bob would miss. However, the 6 from Carol's roll cancels the 6 from Alice's roll, which makes Bob's 5 high enough to hit also.

Whether to reward assist-training at all is up to you though. Most games already reward assist-training just because one dead opponent is better than two wounded ones.

[Psyx]

Assuming this is the approach favored, should the outnumbered combatant have any further modification to the "dice pool" when facing multiple foes, or is the reduction in "aid-related" dice added to the total the only modification you think necessary?

Splitting dice is penalty enough.

You might want to put in a penalty for defending to the rear/flank, but this is a bit of a complication and doesn't really add much.


The next question I have to ask is : what are you trying to accomplish with a system like this?

A better, more accurate simulation which not only has a defence in relation to the protagonist's level of skill but also puts offence/defence options firmly in player's hands, while also getting away from the terrible 'ugo-igo' mechanic prevalent in RPGs and making the game more tactile?


I'd suggest you have one defense pool apply against all attackers. Alice is fighting Bob and Carol. Each combatant has 5 dice. Alice splits 2 offense, 3 defense. She rolls 3 defense dice against Bob, and 3 defense dice against Carol.

The problem with that approach is that it's not at all realistic, in a mechanic that is trying to step towards realism. Being outnumbered should be bad news. Speaking from experience; splitting defence dice between attackers works really well.

It depends on what die type you use (I use d20s), but I'd also look towards capping the number of dice rolled by capping the skill level. By the end of my campaign we had a player at 8 dice, with invariably a few more thrown in due to stances and luck points. It was getting a little unwieldy, but that's also partly the fault of additional combat manoeuvres being in play.

[Amphetryon]

Fair enough.

Well then... what about the system I've come up with? Thoughts?

As I read it right now (before coffee, to be fair), it is reminiscent of the Shadowrun mechanic, which has pluses and minuses. I am initially leery of the 'flat DC equals success' aspect of your proposal, as one of the things I disliked about that system was how it felt like we were just 'comparing sixes,' which got cumbersome and slowed down the flow of the combat and the narrative, at least to my perception.

The Offensive die modifier would be "Coordination" (fine motor skills) while the Defensive die modifier would be "Balance" (ability to maneuver in space), by the way.

[Psyx]

Well then... what about the system I've come up with? Thoughts?

If it works, it works!

I personally avoided the flat DC, and made it a comparison of highest rolls, defender winning ties.

ie: 'A' has four dice. 'B' has 5. A splits off/def 2/2, B splits 3/2.

Both roll their d10 (lets say). A gets a 7,7/3,8. B gets 8,3,8/5,2

A's highest offence is a 7, B's highest defence is a 5. A has hit B.
However, B has got an 8 offence... as has A. Defender wins ties, so B has not struck A.

Alternatively in the even of ties, you might want to go to 'next highest dice' which in our case is another tie... and then B has a '3' and A has no more results, so B sneaks in a hit as well. Ai-Uchi!

Going for opposed rolls makes players actively interested in the other player's dice, which is kind cool and has a nice feel to it.
Depending on what you're doing in your system you might want to give modifiers to dice based on stats, so that one perhaps adds 'co-ordination' (in your game's case) while the other 'balance'.

You might want to think about what happens when stat modifiers take the roll over the maximum on the dice. I'd personally cap the roll at the maximum on the dice. Otherwise a character with good stats but poor training could be 'un-parryable' by a swordmaster who has average stats.

[Amphetryon]

<snip>
You might want to think about what happens when stat modifiers take the roll over the maximum on the dice. I'd personally cap the roll at the maximum on the dice. Otherwise a character with good stats but poor training could be 'un-parryable' by a swordmaster who has average stats.

Continuing the D10 example, my initial thought is that stat modifiers cannot push the total Offense or Defense roll above 11. Ours go to 11! :smallbiggrin:

[Psyx]

That'll almost certainly work. I would recommend kicking the numbers around in your head and doing some play-testing too in order to tune things. I found a few minor problems that I needed to tune with my method, but this kind of combat is a lot of fun for the players.

I cap at (say 10), but multiple 10s beat single ones. ie 'two tens' in offence beats 'ten' in defence. This is the only real case where the second-highest die roll counts.

[AyeGill]

I suggest having a natural ten always beating anything else(and in the case of two natural tens, the one with the highest modifier wins)

[Oracle_Hunter]

I am initially leery of the 'flat DC equals success' aspect of your proposal, as one of the things I disliked about that system was how it felt like we were just 'comparing sixes,' which got cumbersome and slowed down the flow of the combat and the narrative, at least to my perception.

RESPONSES
(1) Having to compare individual die rolls across the table will be much slower than comparing Successes v. Successes.
Sorting every die by number and then lining them up against each other is going to require a lot of "pair comparisons" to resolve a single combat. Plus, you'll need to have a mechanic to resolve ties - which can get wonky depending on the order of how you "cancel" dice.

A Success v. Success mechanic allows each side to roll their dice and then simply pull out the number of "yes" dice they have. Resolving the action now becomes a question of "5 against 3" instead of "1 vs. 1, 1 vs. 1, 1 vs. 1"

(2) Straight Die Rolls give significant advantage to the side with more Offense or Defense dice.
Provided a single die cannot eliminate more than one of its opposing number, a straight number comparison means that the side which rolled fewer dice is going to have fewer dice. Since you've fixed the total number of die to be rolled this gives a pure mathematical advantage to simple assigning more dice to a roll regardless of a character's actual "skill" in combat.

Success Tests, on the other hand, provide an advantage by rolling more dice but by no means ensures that the fellow rolling 9 dice is always going to beat the person rolling 1.

(3) Variable DCs are a good way to model varying skill levels in a fixed die total environment.
As suggested by Point #2, the original set-up has very little scope for modeling the skill-at-combat of any given character. Either you grant free successes (a blunt instrument), the ability to arbitrarially eliminate enemy dice post-roll (somewhat better), or +/- on roll values - none of which sound like they jive with your aims.
Summary
My proposed mechanic should be very quick to resolve and flexible for depicting differing skill levels - both on the offense and the defense.

N.B. There are no other DC modifiers aside from the character's Coordination and Balance values. This means each Player will know what their DC is for a check at all time - they don't even need the DM to describe "situational modifiers" and the like before rolling. Very fast.

* * * *

Now, I depicted CORD and BAL as 8-point stats because that made the mathematical relations clearer but you can re-jigger that however you want provided the underlying mechanic (i.e. more skill = lower DC) remains the same.

Also: It sounds like you have a larger system in mind already; that would have been nice to know before trying to figure out how whether a mechanic would work or not :smalltongue:

[Amphetryon]

RESPONSES


Also: It sounds like you have a larger system in mind already; that would have been nice to know before trying to figure out how whether a mechanic would work or not :smalltongue:

As I suspect is the case with many here, I am working on a fantasy RPG design in its infancy stages. I am currently working on the combat section. I am trying to find another game system - if such system exists - where combat resolution works as follows:
:smallwink:


With coffee in me I can see more merit to your proposal, though I am still not convinced. I'm not a huge fan of a mechanic that creates an unbeatable success, for my major concern. While the D&D example of "a roll of 20 always hits" is a popular one, it's not necessarily one I'm seeking to emulate.

I would like the 'skill at combat' to be more reflected in the damage done and defended within combat than in the hits landed, if that makes sense. My thinking here is it's a relatively easy matter to physically connect with the opponent but a relatively harder matter to do damage as a result of that contact.

I hope I'm being clear.

[Oracle_Hunter]

With coffee in me I can see more merit to your proposal, though I am still not convinced. I'm not a huge fan of a mechanic that creates an unbeatable success, for my major concern. While the D&D example of "a roll of 20 always hits" is a popular one, it's not necessarily one I'm seeking to emulate.
But... the Success Test isn't unbeatable any more than rolling more dice is unbeatable under your original proposal :smallconfused:

Remember that the base mechanic is "compare the number of successes; highest number wins." If Bob rolls 5 OFF dice and only gets 2 Successes, he fails to hit Denny if Denny rolls at least 2 Successes in his DEF dice (assuming ties go to the Defender)

If that's not what you're looking for, what are you looking for?

[Amphetryon]

But... the Success Test isn't unbeatable any more than rolling more dice is unbeatable under your original proposal :smallconfused:

Remember that the base mechanic is "compare the number of successes; highest number wins." If Bob rolls 5 OFF dice and only gets 2 Successes, he fails to hit Denny if Denny rolls at least 2 Successes in his DEF dice (assuming ties go to the Defender)

If that's not what you're looking for, what are you looking for?

Given that both combatants are rolling the same number of dice in my original proposal, I am not sure where "rolling more dice" is coming from, or to what it refers, specifically.

In my initial proposal, if I roll a 10, and my opponent rolls a 10, the scores cancel out. I included matching 6's in my example as an illustration of what happens in a tie; I hope it did not come across as if only that specific integer produced that specific result. As I understand your DC proposal ("as I understand" being an important qualifier), a roll of 10 would be an automatic success. Do you see the difference? In your proposal, I'm understanding both parties to compare things first against a static DC and then against each others' successes. My initial proposal had no static DC, comparing rolls directly.

[Oracle_Hunter]

As I understood it, you gave each Player 5 dice and told them to split them into Offense and Defense pools. Then they roll.

If Bob puts 4 dice in Offense and Denny puts 1 die in Defense, Bob will always "hit" with at least 3. Conversely, Denny can put 5 dice in Defense and 0 in Offense and never be hit unless Bob puts all 5 dice in offense and rolls one more 10 than Denny does.

Is that how you want your system to work?

EDIT: As for my system, while a roll of 10 would always be a success, it is a success that can be canceled out by any other success from the opponent.

So even if Percy the Peasant has OFF 0 but rolls a 10, Kenny the Knight with BAL 8 will merely need to roll a 2 to cancel it out. Yes, this means that even an unarmed peasant is some threat, but it's going to be very hard for him to do damage and very easy for a trained fighter to avoid the attack.

[Amphetryon]

As I understood it, you gave each Player 5 dice and told them to split them into Offense and Defense pools. Then they roll.
So far so good; both players have 5 dice to use as they see fit for combat.


If Bob puts 4 dice in Offense and Denny puts 1 die in Defense, Bob will always "hit" with at least 3. Conversely, Denny can put 5 dice in Defense and 0 in Offense and never be hit unless Bob puts all 5 dice in offense and rolls one more 10 than Denny does.

Is that how you want your system to work? The first sentence is correct. The second is inaccurate on a couple of points. The first point is that, by assigning all 5 dice to Defense, Denny is in what D&D calls "Total Defense Mode", unable to hurt Bob at all, which would seem to be an unusual choice in this system - hopefully for reasons that become clear. The second point is that Denny's rolls, and Bob's, could still result in Denny being hit multiple times if Bob rolls (for example) 7 7 7 7 and Denny rolls 4 4 4 4 6 on defense. The third point is that rolling "one more 10 than Denny" reads from here as an implication that a 10 always hits. That's not a valid assumption in my proposal, though it's close to valid as I understand your proposal. Any ties cancel out, as I said, regardless of whether that tie/matching roll is on a 10, a 6, or a 1.

[Psyx]

Sorting every die by number and then lining them up against each other is going to require a lot of "pair comparisons" to resolve a single combat.

If only the highest die result counts, it's not a problem. Typically paired results are unusual, apart from at the top end, where players shout 'two twenties'. I've not had issue with this, using d20s.


Straight Die Rolls give significant advantage to the side with more Offense or Defense dice.

Yes, it does. But then a highly skilled boxer/fencer/combatant really shouldn't loose to someone with less training, unless they've been VERY lucky.


I would like the 'skill at combat' to be more reflected in the damage done and defended within combat than in the hits landed, if that makes sense. My thinking here is it's a relatively easy matter to physically connect with the opponent but a relatively harder matter to do damage as a result of that contact.

Depending on armour, of course. If armour is well-modelled, then it can be hard to do damage, but unarmoured combat is very deadly.


Bob will always "hit" with at least 3.

Not if you use 'highest dice only'. You might want to put a 'base DC' in there too equating to a target's Agility or something if you like, too.


The joy of this kind of system is that it's very easy to then add abilities that enhance (and complicate!) matters in various ways: Stances, the ability to riposte, skills to breach armour, etc. They can be bought as mini-feats for a fixed price, allowing PCs to individualise their fighting style.

[Oracle_Hunter]

The first sentence is correct. The second is inaccurate on a couple of points. The first point is that, by assigning all 5 dice to Defense, Denny is in what D&D calls "Total Defense Mode", unable to hurt Bob at all, which would seem to be an unusual choice in this system - hopefully for reasons that become clear. The second point is that Denny's rolls, and Bob's, could still result in Denny being hit multiple times if Bob rolls (for example) 7 7 7 7 and Denny rolls 4 4 4 4 6 on defense.
So you can guarantee hits but not defense? What happens to that last defense die which is not canceled out? :smallconfused:

EDIT: I suspect that you are implying that Defense Dice that are unopposed are discarded.

So if the Offense pool was a single die that rolled a 10, it wouldn't matter that the Defense pool was a 9 8 3 4 5 - the Offense would still hit. Was that intentional? It makes Big Offense Pools a great idea and Big Defense Pools a bad one.

[Amphetryon]

So you can guarantee hits but not defense? What happens to that last defense die which is not canceled out? :smallconfused:

EDIT: I suspect that you are implying that Defense Dice that are unopposed are discarded.

So if the Offense pool was a single die that rolled a 10, it wouldn't matter that the Defense pool was a 9 8 3 4 5 - the Offense would still hit. Was that intentional? It makes Big Offense Pools a great idea and Big Defense Pools a bad one.Unopposed Defense dice have nothing to defend against. You could conceptualize them as representing a defensive pose that is not assaulted, for example.

Could you explain, please, what you mean when you say 'guarantee hits'? I'm reading that as an assumption that a given roll will hit with certainty, which is not accurate in my proposal. If Bob rolls 5 Offense dice and Dave rolls 5 Defense dice, there's no guarantee that any of Bob's hits get through. It means, as I said before, that Dave is doing no attacking on his own, so has no chance at hitting Bob. (Maybe Dave thinks the fight's a misunderstanding. Maybe he's fighting a holding action until his buddies arrive or get past the gate.)

Assigning all dice to Offense and none to Defense would also be a huge gamble, hoping that your opponent does not attack you. Since damage is factored separately and is more related to skill (as currently proposed), if Bob rolls 5 Offensive dice and no Defensive dice, while Dave rolls 2 Offensive dice and 3 Defensive dice, there's a real possibility that both Bob and Dave get hit twice, in which case the relevant skills are highlighted. If Dave is a more skilled combatant than Bob, his two hits could well be more lethal than Bob's were.

[Psyx]

Would you allow multiple hits per round, then?

I think that's perhaps complicating matters and stepping away from reality a little.

Better deliver one solid hit. Swordsmen typically don't train to hit people four times in five seconds, as it's a pointless exercise.

[Amphetryon]

Would you allow multiple hits per round, then?

I think that's perhaps complicating matters and stepping away from reality a little.

Better deliver one solid hit. Swordsmen typically don't train to hit people four times in five seconds, as it's a pointless exercise.

Yup, as many as (in the current model) five.

I've not defined the length of time in a combat round to date, and haven't done so in this thread either, so I'm not sure where 'five seconds' comes from.

[Oracle_Hunter]

Could you explain, please, what you mean when you say 'guarantee hits'? I'm reading that as an assumption that a given roll will hit with certainty, which is not accurate in my proposal. If Bob rolls 5 Offense dice and Dave rolls 5 Defense dice, there's no guarantee that any of Bob's hits get through.
If Bob rolls 3 Offense Dice and Denny rolls 1 Defense Die, it is guaranteed that two Offense Dice will "hit" as there is nothing to oppose them.

If Denny rolls 1 Offense Die and Bob rolls 3 Defense Dice, then there is still a chance that Bob gets hit - there is no guaranteed Defense.

You have given Offense a substantial advantage over Defense, unless you add a new mechanic which allows unopposed Offense Dice to still "miss" for some reason.

[Tvtyrant]

Have you considered doing it diplomacy style; writing out your dice devoted to offense or defense and then revealing them after the roll?

[AyeGill]

Have you considered doing it diplomacy style; writing out your dice devoted to offense or defense and then revealing them after the roll?

how do you determine which of the dice were allocated to either category. Unless you mean that each play allocates dice, rolls, and then picks, say, 2 of those rolls for his 2 offense dice, and let the remaining 3 be his defense rolls. which would make sense.

[Amphetryon]

If Bob rolls 3 Offense Dice and Denny rolls 1 Defense Die, it is guaranteed that two Offense Dice will "hit" as there is nothing to oppose them.

If Denny rolls 1 Offense Die and Bob rolls 3 Defense Dice, then there is still a chance that Bob gets hit - there is no guaranteed Defense.

You have given Offense a substantial advantage over Defense, unless you add a new mechanic which allows unopposed Offense Dice to still "miss" for some reason.
As I've said a few times now, there will be mechanics in place to make it possible for an unopposed Offense Die to do no damage.

One of the things that kept throwing me in your responses was the repetitious use of examples with a 10 in your responses. I kept reading your examples with that number as indicative of a belief that a 10 always hit, which is not the case.

[Oracle_Hunter]

As I've said a few times now, there will be mechanics in place to make it possible for an unopposed Offense Die to do no damage.
That's fine. I was just pointing out a serious hole in the system.

And, as I've said before: better to design a system that doesn't need patching than a system you can patch. The "Success Test" system doesn't need an extra mechanic to do what it's supposed to do: make the allocation of Offense/Defense dice a fair question. Throwing a second mechanic to balance a hole in the first risks imperfect balancing and unneeded complexity.

But if you're fine with that, I'll reserve further comment until I see the completed mechanic :smallsmile:

[Sitzkrieg]

If Bob rolls 3 Offense Dice and Denny rolls 1 Defense Die, it is guaranteed that two Offense Dice will "hit" as there is nothing to oppose them.


I hope I'm not confusing things further if I'm wrong, but I think you might be incorrect here. In a case of 3 Offensive Dice vs. 1 Defensive Dice, any number from 0-3 attacks might hit. For instance, if the attacker rolls 9,8,7, and the defender rolls a 4, then three attacks hit. However, if the attacker rolls 9,8,7, and the defender rolls a 10, then no attacks get through. The attack is only cancelled if it is a exact match. A defense of 8, in this case, would result in two hits, as the 9 gets through unopposed, the eights cancel, and the seven would then be unopposed.

A point that was brought up before, and is highlighted here, is that the rules must specify the order in which die are counted. An enterprising player might claim that the 7 was not sufficient to get past an 8 defense, even the 8 defense was cancelled by the attacker's 8, if the player could claim that the 7 should be counted before the 8's cancel.

[Fiery Diamond]

So, OP, as I understand it, your system is like this (so far, anyway):

Bob and Carl are fighting. They each have 5 d10s. They decided how to allot them between offense and defense and then roll.

Bob decides on 3 offense and 2 defense. The result is 10, 6, 2 // 9, 4.
Carl decides on 3 offense and 2 defense. The result is 8, 7, 5 //4, 3.

They check for ties between offense and defense pairs first.
There aren't any, so they move on to the next step.
Bob compares his offense versus Carl's defense like this:
10-4, 2-3, 6-nothing. (or 6-4, 2-3, 10-nothing)
2 loses to 3, but 6 (or 10) beats 4 and the remaining die is unopposed, so Bob gets 2 hits in.
Carl compares his offense versus Bob's defense like this:
8-9, 7-4, 5-nothing. (or 8-9, 5-4, 7-nothing)
8 loses to 9, but 7 (or 5) beats 4 and the remaining die is unopposed, so Carl also gets 2 hits in.

Bob hits Carl twice and Carl hits Bob twice.

Now, we compare modifiers or scores of Coordination against Balance. Bob has high coordination and low balance, but Carl has moderate coordination and moderate balance. For the sake of illustration, we'll just assign numbers. I have two different ideas for how this might play out, both probably oversimplifications, since you didn't specify.

1) Bob has 8 coordination and 2 balance; Carl has 4 coordination and 5 balance. Bob deals 8-5=3 damage per attack, Carl deals only 4-2=2 damage per attack. (This is actually not the best choice, since if Bob had 4 or more balance, Carl could never hurt him.)

2) Bob has 8 coordination and 2 balance; Carl has 4 coordination and 5 balance. Bob deals (8x2)-5=11 damage total, Carl deals only (4x2)-2=6 damage total. (This still may need some reworking, but it is better than the other, since if Bob has, for example, 4 balance, Carl could never hurt him with a single hit (4x1-4=0) but with two hits could hurt him (4x2-4=4)).

In this scenario, each character was guaranteed a single hit and had the opportunity to get up to three based on their choice of 3 offensive dice versus their opponent's choice of 2 defensive dice. How much damage that guaranteed hit (and any others) dealt was determined by their Coordination and Balance skills.

[Amphetryon]

I hope I'm not confusing things further if I'm wrong, but I think you might be incorrect here. In a case of 3 Offensive Dice vs. 1 Defensive Dice, any number from 0-3 attacks might hit. For instance, if the attacker rolls 9,8,7, and the defender rolls a 4, then three attacks hit. However, if the attacker rolls 9,8,7, and the defender rolls a 10, then no attacks get through. The attack is only cancelled if it is a exact match. A defense of 8, in this case, would result in two hits, as the 9 gets through unopposed, the eights cancel, and the seven would then be unopposed.

This is incorrect; a single die roll is opposed by a single die roll. If the attacker rolls 9 8 7 and the defender rolls a 10, only one attack roll is eliminated.

Rather than allowing the defender to win ties a la Risk, I'm thinking about the notion of the defender being the one to assign untied pairs.

So, Pat and Kim each roll 5 dice, with each using 3 Offensive and 2 Defensive dice. Pat's dice come up 7 7 5 for Offense and 6 8 for Defense; Kim rolls 8 7 5 for Offense and 9 3 for Defense. Kim's Offensive 8 and Pat's Defensive 8 cancel out in whatever appropriately descriptive way you like. Pat, quite reasonably, assigns the Defensive 6 to oppose the Offensive 5 from Kim, and Kim assigns the Defensive 9 to oppose one of Pat's Offensive 7's. Pat hit twice, with the 7 and the 5, while Kim hit once, with the 7. The numeric value of a Defensive die that is beaten by an opposing Offensive die conveys no benefit in reducing damage.

Pat's Coordination and Balance - or the relevant positive modifiers of those stats - are 2 and 1, whereas Kim has Coordination 4 and Balance 3. Discounting any other factors, Pat's hits each add 2 damage while Pat's Balance reduces the damage from each of Kim's blows by 1, whereas Kim adds 4 damage to the one hit scored and reduces the damage from each of Pat's hits by 3. Pat was hit once, for 10; Kim was hit twice, for 6 and 4.

[Psyx]

This is incorrect; a single die roll is opposed by a single die roll. If the attacker rolls 9 8 7 and the defender rolls a 10, only one attack roll is eliminated.

Hmm. I can see that becoming VERY messy if skill levels are not similar. I'm all for the best man winning, but that throws things wildly in their direction.

How long were you basing your combat rounds?

As for 'Balance'... would that not be better named 'Agility?' Just a thought. It seems more intuitive.

How are you handling different weapon types?

[Sitzkrieg]

This is incorrect; a single die roll is opposed by a single die roll. If the attacker rolls 9 8 7 and the defender rolls a 10, only one attack roll is eliminated.

I see. Well, as long as suggestions are being taken, I would offer the interpretation I had as a possible alternative. This would discourage an all-out offense, because each die assigned will have a diminishing return. In the method that you proposed, each extra die added to the Offense category will actually be worth more than the previous die, because each die added will have a diminishing chance of being blocked. An optimal strategy in the current method would be to all-out attack every time, since 5 Offense is always better than 4 Offense, 1 Defense, which is better than 3 Offense, 2 Defense, etc.
For a brief example of this, consider an otherwise even fight between a character (named Attacker) that all-out attacks versus one (named Defender) who splits it 4Off and 1 Def. In any given round, Attacker is always going to land between 4 and 5 hits. Defender is always going to land 4 hits. Attacker's strategy is clearly superior.
This also holds true if Attacker full-attacks every round, and Defender fully defends (5 on defense) every time. In any given round, Attacker will land between 0 and 5 hits, and Defender will never land any hits.
No matter what split Defender tries, he will always be at a disadvantage to Attacker.
Therefore, everyone in this system will always land exactly five hits, since nobody is going to put anything into Defense except in extremely limited circumstances.

[Amphetryon]

I see. Well, as long as suggestions are being taken, I would offer the interpretation I had as a possible alternative. This would discourage an all-out offense, because each die assigned will have a diminishing return. In the method that you proposed, each extra die added to the Offense category will actually be worth more than the previous die, because each die added will have a diminishing chance of being blocked. An optimal strategy in the current method would be to all-out attack every time, since 5 Offense is always better than 4 Offense, 1 Defense, which is better than 3 Offense, 2 Defense, etc.
For a brief example of this, consider an otherwise even fight between a character (named Attacker) that all-out attacks versus one (named Defender) who splits it 4Off and 1 Def. In any given round, Attacker is always going to land between 4 and 5 hits. Defender is always going to land 4 hits. Attacker's strategy is clearly superior.
This also holds true if Attacker full-attacks every round, and Defender fully defends (5 on defense) every time. In any given round, Attacker will land between 0 and 5 hits, and Defender will never land any hits.
No matter what split Defender tries, he will always be at a disadvantage to Attacker.
Therefore, everyone in this system will always land exactly five hits, since nobody is going to put anything into Defense except in extremely limited circumstances.

The problem that I see is that this appears to presume either a way of going first, a la Initiative, or a way besides the Defense roll to offset damage, or else it presumes that everybody dies every time in every fight....

[Sitzkrieg]

Well, yes, lots of deaths might be a repercussion of the system. That's why it might be best not to use the system as proposed, because as is, putting dice into Attack is simply a better decision than putting them into Defense. That's the fault of the system, not the player. Changing it so that the highest Defense roll is a DC that must be overcome would alleviate the issue.

[Gralamin]

See the problem with this sort of thing is it can be proven to be optimal to go all in attack. I'm not going to do a rigorous proof, but consider as follows:
If you go all in attack, and they go all in defense, luck may decide the outcome.
If you go all in attack, and they go all in defense but 1, you get one hit for sure, and take only 1. (So you take 1, and they take At least 1)
If they go all but 2, you get two hits for sure, and take 2.
If they go in for 2 defense, you get three hits for sure, and take 3.
If they go in for 1 defense, you get four hits for sure, and take 4.
If they go in for 0 defense, you both get five hits for sure.

So, you are always dealing at least as much damage as they are dealing you. So if you go all in attack, you never do worse then your opponent. Obviously this breaks down at a sufficiently large difference in damage, but that has a side effect: When your hits do more damage, it is always to your advantage to go all in. And because of the above reason, the defender will always not really be able to fight back: They cannot harm the opponent without opening them up to a similar or larger amount of harm, which is what every option they have boils down to anyway. They can only rely on luck and delay it.

Edit: Now I see someone else pointed it out. Oh well, at least it confirms it :smallredface:

[Amphetryon]

What if a tie - and only a tie - canceled out not only its opposite number but an additional Offensive die?

[Oracle_Hunter]

What if a tie - and only a tie - canceled out not only its opposite number but an additional Offensive die?
Marginal improvement in Defense, but at the cost of making combat much more "swingey" - which, AFAIK, you wanted to avoid.

Plus, then you have to determine which die it cancels (Defender's Choice? Next highest? Highest?) and you add an extra step to resolving combats - the "locate ties and discard" phase.

[Amphetryon]

Marginal improvement in Defense, but at the cost of making combat much more "swingey" - which, AFAIK, you wanted to avoid.

Plus, then you have to determine which die it cancels (Defender's Choice? Next highest? Highest?) and you add an extra step to resolving combats - the "locate ties and discard" phase.

Thoughts on improvement that don't involve setting a static DC? :smallsmile:

[Oracle_Hunter]

Thoughts on improvement that don't involve setting a static DC? :smallsmile:
...use a Success Test where the DC varies with Skill? :smallconfused:

I'm still not quite sure what your objection to it was but, well, your current mechanic is too rigid to work without resorting to lots of auxiliary rules. The main issue has to do with the concept of "unblocked" Dice: due to having a fixed number of total dice, the amount of dice per pool will almost always be unequal. Therefore, the treatment of these "unblocked" dice is the central issue of your proposed mechanic and I have a hard time thinking of a way to deal with them withotu causing more problems.

The best way I can think of it is by adding a second-level mechanic which allows you to use "passed" dice (winning or unopposed) to achieve certain ends.

For example:
- 3 Passed Offense = +1 Damage
- 1 Passed Defense = -1 Offense Die

But that will just slow down combat further.

I think it's time to re-examine what you want out of this mechanic:



Create a more dynamic defense than "AC" is in d20, while being less "swingy".

Avoid the issue I've had with past iterations of Shadowrun ... where you needed a bucket to hold all your d6s.

Create an attack mechanism and defense mechanism that, at least at the start and at least in theory, balance against each other. By keeping the number of offensive and defensive dice as a constant, I hope to eliminate scenarios where fights have essentially predetermined outcomes because you're throwing 18 dice and your opponent is only throwing 7, for instance.

Conversely, the choices made for dice to put into offense or defense, respectively, hopefully allow a character to be more aggressive or more reactive, as befits the player and concept. Damage and skill differentials - as yet largely undefined - can still weigh a fight one way or another, of course.
I'd summarize it as:
- Consistent Results
- Dynamic Attack v. Dynamic Defense
- Fixed Dice Pool

Currently, the Fixed Dice Pool idea is wrecking the Dynamic concern - although Consistent Results is doing fine.

How do you want to prioritize these three concerns? Or, can you re-define your aims for the system into 3 bullet points - and then rank them?

[Amphetryon]

Prioritizing them, I'd say:


Fixed Dice Pool
Dynamic Attack vs Defense
Consistent Results

[IcarusWings]

A way too help fix it (don't know if it would completely or not) would be to just not have defense dice lost unless it is beaten or cancelled out by a tie. So if defender puts 2 in defense and gets 10 and 1, and attacker puts 5 in attack and gets 9, 8, 7, 8, 6, none of them hit as none of them can beat the 10. But if the defender had rolled 8 and 8 instead, then the 9 beats the first 8, one hit, the 8 cancels the other 8 out, and all the others hit as he has no defense die left. Leading to a total of four hits.

[Oracle_Hunter]

Prioritizing them, I'd say:


Fixed Dice Pool
Dynamic Attack vs Defense
Consistent Results

Then we need a second-order system as well.

If you're really against a Success Test System...
(1) Each Player takes 5 Dice and secretly allocates them to Offense ("OFF") and Defense ("DEF") Pools.

(2) Each Player rolls and orders their OFF and DEF dice from highest to lowest.

(3) Player A compares his OFF against Player B's DEF and vice versa.

Example

OFFA - 10 8 7 6
DEFB - 7 7 7

10 vs. 7
8 vs. 7
7 vs. 7
6 vs. (blank)

OFFB - 3 1
DEFA - 5

3 vs. 5
1 vs. (blank)
In each pairing, the highest number counts as a Success. Each OFF Success inflicts 1 Damage; Each DEF Success prevents 1 Damage. Any unopposed dice adds one point to the Trick Pool ("TRK") of the Player, while ties add a point to the TRK of each Player.

Example (continued)
Player A does 2 Damage to Player B and prevents 1 Damage to himself. He now has 2 Points in TRKA

Player B does no Damage to Player A and prevents 0 Damage. He now has 2 Points in TRKB

At this point Player B will take 2 Damage and Player A will take none

(4) Before Damage is inflicted, Players may choose to spend Trick Points according to the Skills they have to infuence the outcome of combat. Activating a particular Skill can cost anywhere between 1-3 Trick Points. Trick Point usage is determined secretely.

Examples of Skills
Gambling Man (1 point) - Each Player rolls a single die and compares the result (as per combat). The loser takes +1 Damage. In case of a tie, nothing happens.

Lucky Strike (2 points) - +1 Damage inflicted on enemy
Dodge! (2 points) - Enemy inflicts -1 Damage on you

I Don't Think So (3 points) - After both sides have revealed their Tricks for this round, choose one. That Trick has no effect this round.
As you can see, this is getting pretty complicated but... well, how's that look?

[Amphetryon]

I'm not sure I understand the reason for Trick points, other than an additional complication that would take the place of - or mitigate some of - the skill allocations for Coordination and Balance?

[Oracle_Hunter]

I'm not sure I understand the reason for Trick points, other than an additional complication that would take the place of - or mitigate some of - the skill allocations for Coordination and Balance?
It's to mitigate the Offense Dominance of the base "high-low" system you've proposed.

Rather than using Skill Allocations here, you can have characters purchase various Tricks as they advance for a certain number of XP points (or however you want to do it).

I have no idea how you plan to use Balance & Coordination in your proposed system since there are no DCs to modify and you can't add or reduce the die pool.