sanity check on my attack/defense combat pool system

[OgreBattle]

I bring up riddle of steel and Ronin as often as I can, those are dicepool combat games where your character has a combat pool

What I'm aiming for in my heartbreaker is a melee combat mechanic that has players secretly allocate their combat pool into offense, defense, initiative, and then reveal. But simpler than how those already published games do it.

So....

Characters have a combat pool. When melee combat happens you divide the pool into attack and defense. You roll over your TN (varies with character skill 2+ 3+ 4+ 5+ 6+) to land a hit or negate a hit. This is on a d6.

So characters will improve by gaining more combat pool and having better TN's. Yeah "Variable TN with variable pool" is clunky to calculate on the spot but I plan on the pool as something not modified mid-battle.

In a 1 on 1 fight the character with initiative (could be everyone rolls initiative, but I'm thinking of justhaving team initiative and I Go You Go for speed) declares their melee target in range. The player then divides their combat pool into attack/defense, as does their target.

So if a character wants to stall for time they can throw all defense dice. If a character has a lot of armor or toughness to take hits, or it's crucial they end someone now they can throw all attack dice. Attacks land simultaneously so having initiative is a function of deciding who must defend/attack against who.


As a demo starting point let's say two characters have a combat pool of 4 and they hit on a 3+

The combinations possible per fighter...
4 attack / 0 defense
1 attack / 3 defense
2A / 2D
3A / 1D
4A / 0D

Then their target can take one of those 5 options, so 25 possible dueling combinations with just "allocate attack and defense dice". I want to have as few moving parts as possible, I feel just adjusting the ratio of attack/defense dice already works for lots of maneuvers like "a flurry of attacks, a defend/riposte, turtling, a power attack".

The step after landing hits is rolling to wound, armor saves like in my "not warhams" threads.
Additional layers of complexity I'm considering...

-Initiative die, if you throw more than your opponent you will strike and resolve damage (so KO, or inflict stuns that penalize their slower attacks)
-If characters have leftover combat pool there's a 2nd phase where they can spend their pool, no phase after this. So this lets a lower initiative character throw a few dice to defend when a higher initiative character attacks them, then they can use the rest of their dice to attack a target in range. There's a 'soft defender mechanic' here where the armored warrior next to the robed wizard can afford to throw a lot of attack dice at the goblin next to both of them, the goblin defending itself will take dice away (maybe all) from attacking the wizard.

[Orca]

My first thought is that with small dicepools and presumably with the attacker needing a net success you're going to see a lot of attacks fail. 2d vs. 2d, 2/3 success gives a 21/81 or ~ 26% chance of hitting, then you possibly have armor. If you need say 3 such hits to end this duel then that's about 12 rounds. If one side goes to 3d defence, 1d attack then it could last until a player's wrist drops off from rolling too many dice.

[brized]

For the sake of speed of play and fun, maybe max offense should beat max defense at least 30% of the time with equal opponents? Some research on this: https://www.pokerstrategy.com/news/worl ... ly_101770/

Slight tangent, but this is also likely why the minimum viable match-up in fighting games is 3-7, while 2-8 or 1-9 is considered broken.

[OgreBattle]

Yes two fighters going full defense is definitely going to happen with this set up... "well there ARE combat sports matches where both turtle..." so that might be a feature... that also causes the crowd to boo and the fighters to get paid less in the future.

This is also meant for D&D scale dungeon crawls and equal opposition skirmish, so spending all of your dice to defend against one attack can leave you open to another's attack.

30% win rate study is interesting. That does sound about right for ork shooting accuracy prayers (which is actually much worse than 30% because of rolling to wound and save afterwards but psychologically it feels like there's a chance since those are separate steps)


I think the '2nd phase for leftover pool' option will help with turtling:
A & B both have 4 dice in their pool

A goes 4 defense, but B decides to only throw one attack die, B's attack likely fails but now A has 0 dice while B has 3 dice for a 2nd phase attack.

So B has done a 'feint' maneuver. If A had instead gone full attack instead of turtling then B gets smashed.

Looking at a greatly simplified olympic fencing tactical wheel...


So all dice into one big attack is a simple attack, but if the opponent goes heavy on defense they can negate most hits while throwing an attack or two of their own, a "Parry Riposte"

All-in turtling is defeated by having a low dice attack in the first phase, and then attacking in the 2nd phase "Compound Attack" which is feint and attack to another opening, or throwing two attacks in quick succession

That "compound attack" where you only use half of your pool to attack in both phases will be defeated by the opposition dividing their attack/defense all into one phase, the "counterattack"

[deaddmwalking]

So let's say assume that it takes 3 hits to incapacitate your opponent. Assuming that, regardless of the attack dice they can only get '1 hit' in a round, you know that you're going to have two rounds where you can go 'full attack' before you're really in trouble.

Your opponent really can't go 'full defense' - they should try to get at least one hit because if they stay on full defense against full offense indefinitely, they'll lose.

To me, it looks like the Nash Equilibrium is 'always full attack'.

[brized]

It'll get complicated with more than just 1v1. A double attack can be optimal when ganging up on someone who already spent all their combat dice, or when you have super good combat skill and know that you have a chance to land 2 attacks or definitely 1 even when splitting dice. Unless there's some benefit to overshooting on hit accuracy?

This combat system doesn't seem to resolve the issue of focus fire always being optimal, instead of a situational choice.

[owlassociate]

This combat system doesn't seem to resolve the issue of focus fire always being optimal, instead of a situational choice.

That might be more of a feature than a bug. Realistically, spreading your attacks between multiple oppponents is kind of a bad tactical choice. And if focus fire is important, then battlefield control also becomes more important.

[Orca]

If you're now dividing about 4 dice between 2 attack and 2 defence phases then I'll double down on my claim that a lot of attacks will miss. Yes it adds a strategy component, but it also means that whiffing will be happening an awful lot. This could take a long time to resolve.

If you don't believe me try getting someone you know in RL to run thru a couple of sample combats with you. Right to the end, not to first hit. Some people need to see the results of what they're suggesting to appreciate them.

[brized]

Realistically, spreading your attacks between multiple oppponents is kind of a bad tactical choice.

Oh? Based on what? And why does it matter if we're talking about game design, where a player choice that's always good makes all other options for that choice into trap options?

[OgreBattle]

On the ganging up issue and possible solutions

Ronin- The beginning of the melee phase requires players to separate combat into as many 1 on 1 duels as possible, only afterwards is there a one vs many. That can feel too 'gamey' for an RPG though.

I'd like to do something like that in a Movement ("Maneuver" phase, ranged attacks also happen here) phase where the game is about setting up melee in a way that's advantageous for your party and not for your enemy. But if that doesn't work out the most straightforward thing is "if a character declares a melee engagement against an enemy already engaged then an unengaged model within range can declare an interception"

So if it's 2 on 2, A1 engages B1, when it's B2's turn they can get intercepted by A2. But if it's 3 vs 2 then then A3 can choose which 1 on 1 fight to dogpile into.

[owlassociate]

Realistically, spreading your attacks between multiple oppponents is kind of a bad tactical choice.

Oh? Based on what? And why does it matter if we're talking about game design, where a player choice that's always good makes all other options for that choice into trap options?

I'm not going to explain why, IRL, efficiently reducing the amount of opponents you're fighting is, in almost all situations, better than not doing that . Can you explain to me a common situation where that would not be beneficial?

On to your next point. It matters in the context of this post because OgreBattle cares about having "tactical realism" in his game. Seriously, just read his posts. He is constantly talking about how he wants combat in his hypothetical games to feel like IRL martial arts and skirmishes. Now, he didn't make this explicit in this post, but I'm willing to bet that's his goal.

Finally, I disagree with your notion that any interesting or fun RPG would completely remove the ability to make bad decisions during play.

[angelfromanotherpin]

I'm not going to explain why, IRL, efficiently reducing the amount of opponents you're fighting is, in almost all situations, better than not doing that . Can you explain to me a common situation where that would not be beneficial?

What very few games recognize is that the cost of focus fire is that you have opponents who aren't being attacked. Those enemies can then focus their efforts on offense and become much more threatening. Simulating that is both more realistic and creates an interesting tactical trade off between efficient removal and cautious suppression.

But beyond that, if your combat paradigm is dominated by focus fire, then PCs are going to get focus-fired down as well, which usually feels pretty bad.

[OgreBattle]

Maybe focus firing is a feature, as the target can turtle if they have dice left over.

Say we have A&B vs Y&Z in a melee

AB
XZ

Team AB wins initiative and decides A will attack first, A attacks Z, they bot allot combat dice.

Team YZ now decides who to activate, X is activated.

If A had gone all-in offense: now A has no dice to defend with so X can decide to throw a few attack dice at A with a high chance of hitting while still having dice left over to deal with B.

If Z had gone all in and has no dice left over, X might want to engage B so B doesn't have a max dice attack against Z.

[owlassociate]

What very few games recognize is that the cost of focus fire is that you have opponents who aren't being attacked. Those enemies can then focus their efforts on offense and become much more threatening. Simulating that is both more realistic and creates an interesting tactical trade off between efficient removal and cautious suppression.

I agree, and I think that a combat pool sort of intrinsically models this point. I was considering focus fire in the context of direct damage abilities, which would encourage the use of things like battlefield control AoEs to single out opponents and keep yourself from getting surrounded.

But beyond that, if your combat paradigm is dominated by focus fire, then PCs are going to get focus-fired down as well, which usually feels pretty bad.

If we're doing group initiative anyway, you could say that mobs of like opponents are single entities and turn their attacks into AoEs. You could still model focus fire with a special ability for highly disciplined or organized opponents but say that most of the time a mob of enemies just isn't trained enough to focus all attacks on a single target. Do you think that would be a worthwhile way to counteract this?

[deaddmwalking]

Focus fire is the smart action if you're aware that the opponent is still available on your turn.

Assume that each member of your party of 4 has a 25% chance of eliminating the most dangerous enemy. If you declare your target before attacks are resolved, each player has the opportunity to decide whether to attack or not.

If they all attack, there is a roughly 70% chance that the target dies that round. However, there is also a real chance that the first attack neutralizes the enemy and the remaining 3 attacks would have been more efficiently used targeting another enemy.

If the event that all targets are equally valuable, it makes sense for each player to target a unique target; odds are only one of them is neutralized, but there isn't much sense to risk the potential of wasting attacks; if they all target someone different there is a chance that more than one opponent is neutralized...

If a target cannot be neutralized with a single hit and the other players are assured that their attacks will not be wasted, it makes sense that they would concentrate fire.

From the player perspective, having a chance of being removed from play with a single attack isn't always a good thing... You may want to have a formalized 'plot armor' that lets players 'dive out of the way' when they otherwise might face elimination.

[OgreBattle]

If a target cannot be neutralized with a single hit and the other players are assured that their attacks will not be wasted, it makes sense that they would concentrate fire.

From the player perspective, having a chance of being removed from play with a single attack isn't always a good thing... You may want to have a formalized 'plot armor' that lets players 'dive out of the way' when they otherwise might face elimination.

In my dice pool system though the one being focus fired can dedicate all of their dice to defending, and then their ally can presumably hit with all dice set to attack against an opponent with no or few dice left over to defend.

It's kinda like a soft tanking mechanic, the one getting focus fired is pitting defense dice against attack dice.

[deaddmwalking]

It's kinda like a soft tanking mechanic, the one getting focus fired is pitting defense dice against attack dice.

I think I'm missing something. If each player has 4 dice, including the one being focus-fired on, even if each player went with a balanced approach (2 attack/2 defense), the target will only have 4 defense dice against 8 attack dice.

[OgreBattle]

It's kinda like a soft tanking mechanic, the one getting focus fired is pitting defense dice against attack dice.

I think I'm missing something. If each player has 4 dice, including the one being focus-fired on, even if each player went with a balanced approach (2 attack/2 defense), the target will only have 4 defense dice against 8 attack dice.

Yeah you got it. The people throwing all attack dice against one target going full defense now have no dice left to defend with, so the defending guy's partner can dedicate all 4 attack dice too.

Initiative also passes between teams so it would be A1 throws 4 attack dice against B1 (who presumably defends), then B2 can actiate and engage A1 who has no dice to defend with, or engage A2 who hasn't acted yet.

----

More considerations....

I'm thinking there should be some "Warlord" "Chrono Trigger Tech attack" way of getting two allies to activate at the same time too, maybe at cost of dice. So if both discard one die they can both attack for a total of 6 dice in one activation. So tradiing potential dice in a whole round for immediate pressure at that moment.

Another consideration is how Osprey Ronin does it, allies involved in the same melee can share dice pools. So if a swordsman has a spearman ally behind them, 1 of the spearman's dice can be used for attack or defense.

[brized]

I'm not going to explain why, IRL, efficiently reducing the amount of opponents you're fighting is, in almost all situations, better than not doing that . Can you explain to me a common situation where that would not be beneficial?

You made an assertion. The burden of proof is on you to back it up. If you didn't have anything to back you up when you made the assertion, why did you make it?

Finally, I disagree with your notion that any interesting or fun RPG would completely remove the ability to make bad decisions during play.

Having a tactical decision be good or bad depending on the situation is fine. The more decisions that are always good, or always bad, leads to a pure strategy, or degenerate play.

[owlassociate]

Brized, what I' trying to get across to you is that I really don't care about this argument. The thing you are asking me to prove seems like it's only unobvious to you, which is why I said that it's your problem and not mine. Personally, I think you're being dense for the sake of being argumentative, but my opinion doesn't really matter. I posted to stimulate a constructive discussion and I think that went more or less okay. I did not post to get into some sort of shit-flinging pseudo-intellectual argument about game design theory that will ultimately go nowhere and change nothing. And if you go crying about "bad faith arguments" or "ad hominem" or some shit, I seriously don't care. Because that's fucking goofy. It's a fucking internet forum about RPGs and I'm just here to have a good time and share ideas.

[deaddmwalking]

I'm not going to explain why, IRL, efficiently reducing the amount of opponents you're fighting is, in almost all situations, better than not doing that . Can you explain to me a common situation where that would not be beneficial?

You made an assertion. The burden of proof is on you to back it up. If you didn't have anything to back you up when you made the assertion, why did you make it?

I'll take a stab at it.

In real life, if you have 10 people shooting at you, and you don't shoot any of them, you still have 10 people shooting at you. If you shoot one of them, you only have 9 people shooting at you.

9<10.

If you shoot 9 of the 10 people shooting at you, there is only 1 person left to shoot at you. 1<10.

Assuming that each shooter had roughly the same odds of successfully attacking you, you are guaranteed fewer attacks and virtually guaranteed fewer hits.

Fighting one person is hard enough. Fighting against two people is harder still. Real-life doesn't obey the ninja principle where more attackers automatically makes each of them significantly weaker than when they attack alone.

[owlassociate]

Thank you deaddm for doing what I would not. I apppreciate your sense of ethics. I'll admit I've gotten really jaded against this kind of "I've read the list of logical fallacies and now everything is a philosophical debate" guy. I tend to just react by telling them to fuck off in so many words.

Anyway, back to the sharing ideas thing I was talking about...

It seems like simply ganging up against a single target would be more effective than the "tech attack", especially with the side-based initiative because it'd be essentially the same result but you'd have more dice. Unless I'm missing something. The only ideas I've come up with to change that are to either add unique rider effects to "tech attacks" (like "bounce the lightning bolt off my shield to double the area of effect") or just abolish the idea.

[MGuy]

If multiple.people are spending their actions to work in tandem you probably want the value of that action to be greater than if the two people had simply acted separately. You could let tech attacks do more damage to a single target, therefore making them valuable a for use against single high priority or tough targets while making it not as useful versus groups of people where spreading damage among a wider group of lesser threats is better. You could also do the opposite where tech attacks allow you to most efficiently spread damage among a wider group which would then let ganging up against single targets get done by acting separately.

As far as the players getting ganged up on you probably want to approach that by allowing players to be able to establish and hold battle lines or 'hate'. You should want a game where positioning matters enough to be able to control territory so it encourages the players to utilize tactics that reduce the enemy's ability to gang up on them. You could also give players the ability to control where the enemy damage goes. Maybe through 'hate' effects or something similar.

[OgreBattle]

On team attacks having a penalty, my thought process is trading power for speed. Say the dragon only needs to be wounded one more time to die. FighterA can attack with 4 dice, FighterB can also attack with 4 dice, but if FighterA's attack fails then the dragon gets in a round before FighterB attacks. So then FighterA & B can declare a team attack of combined 6-7dice, trading sheer power for speed. It also means that team attacks are more of a finishing move so you don't just do them all the time. I haven't thought too much about this though just jamming ideas.

---

As for more ways to prevent dogpiling (or get the dogpile in your favor)... there's a movement/maneuver phase before the melee phase that also follows the alternating activations. So the first guy to move can then get counter-charged by the first guy from the enemy team to move... which means you can activate your big brute first to take up space and absorb charges for later movement dudes, and your fastest party member activates later to try and dance around the brute to hit the wizard. This is based on what I've seen of Fantasy Flight xwing where you want armored ships to move first and block then agile ships attack the enemies you blocked.

In Mordheim, movement is in a straight line and there's 'intercept zones' as shown below:


So your heavy armor dude doesn't have to be standing right in front of the wizard, they can move in to intercept.

I'm also thinking this maneuver phase is where you attempt to push, shift enemies, maybe knockdown to penalize them in the combat phase. Like in Bloodbowl & Dreadball. I haven't thought exactly on how to disengage and the balancing point between "stuck in a tarpit with no escape" vs "Mongol archer keeps on kiting my slow melee guy"

In Dreadball you declare reactions to how you're being tackled, you either slam back (chance to push them back and move into their space) or dodge (chance for them to move into your space as you get a free move). I'd like to do that too so this game has fancy footwork and "my agile thief sidesteps the charging brute, the brute is now flanked by thief and his barbarian pal" baked into the core movement mechanics.

[deaddmwalking]

In standard D&D, a wizard has some abilities that work on ranges of 400 ft + 40 ft/level (iirc for long range). A typical battle grid might only be 20 x 20 squares (100 feet by 100 feet). If you want to ensure that wizards are 'in the fight' you probably want to constrain those types of abilities. Otherwise, in the 'movement phase' the wizard will always move AWAY from the enemy, and nobody will be able to get close enough to engage...

I still think you need to consider whether there is an optimal strategy - if every player always puts all dice into attack, then what's the point in having a mechanic where you COULD shift some to defense?

If my opponent is a much better attacker, even if I put all of my dice into defense, I'll still be hit - therefore I might as well go full attack.

If my opponent is much weaker than me, I should drop them as quickly as possible - therefore I might as well go full attack.

This might be reduced if there are 'degrees of hit' - ie, in D&D you roll an attack; if you hit you roll damage. If you're using something where you roll an attack and roll 4 hits you do 4x as much damage as you do 1 hit, then being able to negate some number of hits might make sense... Also, you could have limitations on number of hits... Ie, if I need a 4+ to hit on a d6 but the maximum number of hits I can have is 2, it probably doesn't make sense to roll 8 dice - having 4 hits but only being able to use 2 is a waste...

Of course, it doesn't necessarily follow that 'shifting dice' is the right way to go, either. You could probably just have fixed number of attack dice and fixed number of defense dice.