A Demon Haunted World

[jt]

If your mass combat rules and skirmish rules are compatible, and you have rules for how an individual character affects a skirmish (say AOE becomes single target and small AOEs get reduced damage), and rules for the reverse (a squad of skeletons does AOE damage with their melee attack, extra if they outnumber the opponent), then maybe you can rig it so awkward middle sizes work out. Which would let you handle five skeletons more naturally; they'd act more like a swarm in the next scale down.

I think you have to throw away fine grained position for that to work. The system I'm working on uses a fixed sequence of combat stages (ambush, ranged, melee, close) instead of position. Something like that could do it.

[nockermensch]

How hard is to come with a correct enough formula for resolving a number of equal attack rolls with a single die? This could make not just five skeletons but hydras less cumbersome.

I'm talking about something like (assuming you're rolling a d20):

roll target number -3 = 25% of the attacks are hits
roll the target number = half the attacks are hits
roll target number +3 = 75% of the attacks are hits

But you know, statistically sound, something I'm pretty sure my attempt isn't.

[Foxwarrior]

Well, the bell curves are different depending on your chance to hit, so the correct enough formula would output a 20x20 chart (or 19x20, since 1s always miss).

[jt]

The statisticaly sound version is a binomial distribution, which is a hot mess. You could approximate it with a pile of dice with the same mean and variance, but you'd need a look up table, and the result would be some fractional number it dice per attacker.

I'd try something like rolling an 18 for 5 attacks is equivalent to rolling an 18 17, 16, 15, and 14. So you multiply the damage by min(margin of success, number of attacks). This is more swingy than the correct answer but that might be what you wanted anyway.

(But really I think converting to AOE makes more sense than 20 skeletons attacking one guy out of your party. Both in terms of vague realism, and in terms of what you'd want to have happen in a game.)

[DrPraetor]

[*] Is too small a number to be reasonably aggregated into a swarm unit.[/list]
-Username17

I disagree. For goblins, indeed for any monster that is expected to occasionally outnumber the players, *two* is a large enough number for a swarm unit.

Furthermore...

what if we place an inherent limitation on summoners that they can only give and maintain one command per turn,

This is also bad, because it encourages summoners to spam commands.

Borrowing ideas from the sibling WoD thread, there should be swarm attack options, which swarm monsters should furthermore be incentivized to use at least against higher-level opposition.

So if you have five skeletons, they make a cooperative attack against some foe as their single collective action (with one die roll) - unless for tactical reasons you need them to cover five different orc raiders while you escape with the villagers, in which case you do that instead.

Partially I argue from consistency - if being fast doesn't involve making multiple actions (and it shouldn't), but gives bonuses instead - then, being three goblins doesn't have to involve three spear-stabs and can be bonuses instead.

Frank also commented that extra attack rolls are to be avoided where possible - I was handing them out too much in the monsters I wrote up above, so I will change that. Being able to take an attack action and also a buff, or something like that though, is I think a pretty good mechanic.

Gotta go
--DrP

[Username17]

How hard is to come with a correct enough formula for resolving a number of equal attack rolls with a single die?

For goblins, indeed for any monster that is expected to occasionally outnumber the players, *two* is a large enough number for a swarm unit.

Creating mathematical formulas to aggregate and disaggregate groups of any size is not particularly hard. Actually using those forumlas in play is a fucking nightmare. We're talking about multiplying by percentages and converting numbers into Logs and then adding them. At the point where you say "A slide rule would make this operation easier to perform at the table" you pretty much have to rethink your life's priorities.

Now there are concessions to playability you can make and you can do a lot of the log conversions ahead of time. You can reduce it so that adding or subtracting Goblins from the pile is a relatively simple addition or subtraction operation. But it's difficult for me to imagine getting such things down to the level that they can be collapsed and expanded at the table faster than you could just roll attacks for five Goblins. There is definitely a point where doing the math operations or even looking up a chart in the DMG is faster than rolling separate attacks for every Goblin or Skeleton Warrior, and the breakpoint is going to vary substantially depending on a host of factors from how long your game takes to resolve attacks (parry rolls and knockdown tests and shit make individual attacks take longer to resolve and make aggregation look more attractive) and even how mathematically inclined the people at your particular table are.

But there's another thing going on, which is that smashing individual skeletons is actually pretty fun. The 4e Minion rules were intellectually insulting, and got old super fast, and did not interact properly with levelling mechanics, but at introduction they were one of the most popular things about 4th edition. Taking an action and removing an enemy model is pretty satisfying, especially when that means that there's demonstrably less attacks coming back the other way. I don't think people would get the same visceral thrill from inflicting a mathematically derived debuff onto a skeleton mob unit.

Another issue of course is that RPG characters tend to attract followers and acquire weaponry in a piecemeal fashion. One day they start getting followed by a pixie, another day they liberate some hobgoblin slaves, and so on. And further, the party are going to be using the 5 best weapons they have and some of those might be pretty weird - but their hobgoblin ex-slave follower soldiers are going to be equipped with their 6th through 9th best weapons - and those are unlikely to be a homogenous group. When the players start having their own banner carriers, those armsgobblins are going to be equipped with castoff adventurer equipment. Even if we could aggregate them effectively into a single unit, we probably wouldn't want to. One of those Hobgoblins is carrying the pincer staff you took from the Kuotoa monitor you fought five adventures ago, another is using the weeping glaive you took from the Durzagon iron lord you murderstabbed the adventure before that. And so on.

-Username17

[erik]

If having a mob of skeletons attack something in melee then you can probably cap their effectiveness at like 5 since any more than that and the next skeleton is just waiting in line for its turn to stab. So you can have a magic number of boost for numbers in offense up to 5 and then the virtue of greater numbers beyond that is just extra hp.

[DrPraetor]

Swarm units don't need to be mathematically identical to having the five skeletons stab a guy in turn. In fact, it's better if they aren't, because then there is a tactical choice to be made in whether the skeletons should do a coordinated attack, or whether they should stab five times sequentially.

If the coordinated attack is usually the better choice, or even usually the better choice against superior opposition, then you will save time at table as an emergent property. You can use the same skeleton stat block you used at level 1, when you get to level 5.

Simple proposal:
Swarm-suitable units have the ability "coordinated attack" in which they make one attack roll, with +2X to hit where X is the number of extra units.

If we assume 3D6, then +2 more-than-doubles your chance to hit when you need a 14, +4 more-than-triples your chance to hit when you need a... 14, and +6 more-than-quadruples your chance to hit when you need a... 14! So the rule for monsters is simple - if they need a 14+ to hit, they swarm attack, otherwise they attack individually. In order to play nice with the RNG, the stack limit is 4.

Players can take more time to think about it if they want because they may be more concerned with tactics and ZOC and shit, as they should be because they're supposed to be enjoying playing their small-combat tactics game.

This is not intended to be mathematically identical to having 2-4 goblins stab a fool, but it:
[*] Keeps the goblins as independent units, which become as an aggregate less dangerous as you stab them.
[*] Enables a stack of four goblins to hit an armored knight slightly more often while at the same time not enabling 20 peasants to easily grind down a pit fiend.

[Thaluikhain]

How hard is to come with a correct enough formula for resolving a number of equal attack rolls with a single die? This could make not just five skeletons but hydras less cumbersome.

I might have missed something somewhere, but if the attacks are identical, can't you just roll 5 dice simultaneously? Ok, yes, you'd have to own 5 D20s, and in systems that use multiple dice (say, 2d6 for an attack), you'd need pairs of different coloured or sized or shaped dice. Or have an app that rolls 5 d20s for you.

[jt]

You can, but I think the hope was that there'd be a substitute formula that's faster. As a few posters have covered now: nope.

[Username17]

If having a mob of skeletons attack something in melee then you can probably cap their effectiveness at like 5 since any more than that and the next skeleton is just waiting in line for its turn to stab. So you can have a magic number of boost for numbers in offense up to 5 and then the virtue of greater numbers beyond that is just extra hp.

Certainly Shadowrun capped the outnumber bonuses at +4 and no one called bullshit on them for doing so. 3e D&D caps most engagements at 5-8 adjacent enemies and everyone seems OK with it. The issue isn't that we're struggling to find reasons to aggregate when enemies come in groups of 6+, it's that there's a pretty significant range where we do not want to aggregate and it nonetheless slows things down if we don't.

Basically there exists an anti-sweet spot in which characters get more complex and have minions and activatable equipment and shit and the enemies have weird resistances and puzzle monsters have multi-part solutions and shit but people don't have entourages so big that they be conveniently abstracted off the skirmish battlefield entirely. Pressures exist which make combat turns take longer and effective remedies aren't immediately obvious and might be universally worse than simply accepting that skirmish-scale level appropriate combats gradually take more physical table time as player characters become more powerful and have more options and abilities.

Swarm units don't need to be mathematically identical to having the five skeletons stab a guy in turn. In fact, it's better if they aren't, because then there is a tactical choice to be made in whether the skeletons should do a coordinated attack, or whether they should stab five times sequentially.

If the coordinated attack is usually the better choice, or even usually the better choice against superior opposition, then you will save time at table as an emergent property. You can use the same skeleton stat block you used at level 1, when you get to level 5.

This is the kind of thing you're going to want to do in order to seamlessly move from giving a shit about individual giant rats at level 1 to having a writhing mischief of giant rats pouring out of a tunnel at level 7. But 5 is a very problematic number, because that is about the number of player characters there are. You definitely do not want any of the player characters to emergently want to abstract themselves into a passive bonus on the Berserker's attacks. It invalidates their agency as a co-author in the collaborative fiction that is a role playing game.

-Username17

[Chamomile]

You can get around that by making coordinated attack a specific ability that only certain units have, and none of them PCs, which I believe was DrPraetor's original suggestion. This does raise potential verisimilitude problems where the Fighter, the Barbarian, and the Paladin want to coordinate their attacks because they're fighting a baddy with high enough AC that it's worth the trade-off, and by all logic those three should be just as capable of coordinating with one another as three random hobgoblins. Is there a mathematical sweet spot where players technically can coordinate their attacks, but are rarely incentivized to because they're already very likely to hit their opponents relative to the trash mooks intended to use the feature? Particularly with the 3d6 it feels like there comes a point where squeezing an extra 2.5% success rate out of the Barbarian's rage attack isn't worth giving up a 90% chance that each of the Paladin's and Fighter's attacks will land and do damage separately.

[Pariah Dog]

Something like aid another?

Coordinated attack: A group of characters can work together to try and create an opening to strike a much tougher opponent. Each person participating in this must be able to reach the target creature (possibly restrict to melee only? or only purely melee or purely range for the whole "volley of arrows" gimmick of a team of archers firing at, say a dragon?) for each participant the the attacker gets a +1 (2?) bonus to hit. This uses up each participant's action as well as the attackers.

[jt]

the Fighter, the Barbarian, and the Paladin want to coordinate their attacks

This makes me think:


Which is a silly answer, but Chrono Trigger style combined attacks actually do reduce the number of actions taken and scale right on up to arbitrary sizes. It's pretty similar to aid another, with the difference being that the result is a distinct ability owned by the group, instead of a pumped up version of an ability owned by one character.

[Omegonthesane]

the Fighter, the Barbarian, and the Paladin want to coordinate their attacks

This makes me think:


Which is a silly answer, but Chrono Trigger style combined attacks actually do reduce the number of actions taken and scale right on up to arbitrary sizes. It's pretty similar to aid another, with the difference being that the result is a distinct ability owned by the group, instead of a pumped up version of an ability owned by one character.

Chrono Trigger combined attacks don't scale up at all, there is a single combined attack designed for each combination of 2 or 3 characters that the designers wanted to have a combined attack.

Although the more general idea of having explicit abilities that interact with combined attacks sounds like it would reduce the pain for players who decide their best option is to coordinate their attacks.

Or you could look at what SIFRP does (and then cry in a corner at the sight of roll & keep RNG, but w/e) where Coordinate is an action you can take where you make a Will roll and add an amount to the attack roll of the person you're coordinating with based on how high you yourself rolled. (And then rule it out in the context of D&D where there aren't any noncombatants, and then cry in a corner when your local group bans it from future SIFRP campaigns for being hilariously broken due to the boring number output).

[jt]

Chrono Trigger combined attacks don't scale up at all, there is a single combined attack designed for each combination of 2 or 3 characters that the designers wanted to have a combined attack.

You could just... not do it that way. Make a specific combined fighter/wizard attack with very specific requirements, sure, but also a combined 3-8 mooks attack that just cares everyone has a sword. Make a 1000-ish-mook attack and there's your mass combat system.

[OgreBattle]

So in that triple tech attack, would each PC roll attack too? It would be cleaner to have 1 roll, but it seems more playful to have everyone involved roll.

[Pariah Dog]

That's what I was suggesting. The group designates 1 hitter (ideally their hardest hitter) and they do the attack roll applying bonuses for teammates declaring they're assisting.

[deaddmwalking]

The hardest hittter will nearly always be the same player. You want to rotate primary so something like initiative (which is determined each battle) would be better or some other form of rotation. Since combat might (sometimes) be a single round, if A always leads the attack, A is going to end up being the one that gets the most attacks inevitably. Being sure that sometimes B and sometimes C goes first is important.

[Chamomile]

How did we come to the conclusion that boiling down most or all party actions to one roll by one guy who just adds the bonuses of his teammates would be a good thing? This conversation started as "this is great for identical mooks but we need to find some reason why it's not optimal for unique PCs" and somehow jumped tracks to "how can we make this work for unique PCs" without any discussion of why we would actually want that.

[Pariah Dog]

For mooks yeah it works great because no one gives a shit about the difference between orc 1 and orc 5. Or even as a boss fight gimmick (The weaker orcs all do the swarm tactics providing damage bonuses for the orc warboss instead) For players to be using this it strikes me more of a desperation tactic when they've picked a fight with something much bigger than them (levelwise) and their odds of hitting it are slim to none rather than a Crono Triggeresque "Spam X Slash with Crono and Frog while Marle casts Ice" tactic. Also it started as a way to cut down on the amount of time Bill the Necromancer spent ordering around his 6 skeletons.

[jt]

I think the sweet spot is that the optimal actions for a regular party are (mostly?) individual moves, the best way to use added henchmen is to combine them with someone more important, the best way to use lots of minions is to combine them into squads, and combined attacks between squads are equivalent to a reasonable mass combat system. That may or may not be actually possible.

I brought up the Chrono Trigger method because it does the same thing as Aid Another in a more interesting way. The part of that worth stealing is "named group actions with different prerequisites," not the rest. Any system that asks a two player characters to do the equivalent of spamming x-strike is definitely a failure. (Asking the PC knight and their NPC squire to, though, may be fine.)

[jadagul]

If we want to make this advantageous for mooks but suboptimal for PC characters, one option is to say it only works for basic attacks, whatever that means in the given system.

Skeletons can't do anything other than a basic attack, so the group attack is an upgrade. PCs and other "interesting" characters sometimes could upgrade their basic attack by doing a swarm attack, but they can upgrade it even more by using their actual abilities.

This breaks down if you have mooks with actual abilities, but on some level you really shouldn't have that anyway because mooks are supposed to be kinda boring.

[jt]

A proposal:

Everyone has a mook rating (which might just be their Level/CR). When you use some sort of group action, you spend your turn and add your mook rating to it, and the action determines what you do with that. There's a certain rough standard for what one mook point is worth; adventurers and interesting monsters have actions more effective than their weight in mook points, while soldiers and minions don't. Group actions that require 20+ participants are more efficient per mook point, so the solo/mook bar gets raised at that scale.

Everyone knows the standard angry mob action, which take 4+ folks with melee weapons and turns them into an AOE attack with +mook to hit and an area based on the mook rating. There's also a ranged variant.

Billy Necromancer has a mook rating of 5 but has better stuff to do than to use it. He has six acolytes with a mook rating of 1 each, which they want to use instead of their weak necromancy cantrip. Billy can cast Animate Dead to summon five skeletons at a time, up to a max of ten, and the skeletons are melee characters mostly worth their mook rating of 1. Animate Dead can also be cast as a magic circle, which gives you just one higher-quality undead at a time, which depends on how many mook points your circle adds. Billy's five acolytes are enough to help him summon a wraith.

This quickly escalates to 18 participants, but Billy, the accolytes, the wraith, and the ten skeletons use just 4-5 actions by grouping, so this still is a reasonable boss fight for a party of four.

In a later plot twist where Billy joins the party, Billy Necromancer adds his skeletons to Sally Paladin's militia to make an army big enough to use mass combat rules, and we fold the acolytes and a bunch of retainers into it. A few levels later Billy can summon enough skeletons to make a pure skeleton army, so he splits them out to use the better Army Of The Dead group actions instead of the generic army ones.

[Pariah Dog]

Until the militia start taking morale penalties for undead in the army. Sorry been playing Heroes of Might and Magic lately.