D&D 5E Dwarven Stag Party In The Caves of Chaos?

[slobo777]

Intoxicated is still in the play test. I wish I knew where it was coming from as an idea. Can't say I'm a fan, really.

However, I've been adjusting my code for the new monster data, and encounter building rules. Just for fun I decide to see whether stacking up damage reduction is worth the payment in taking disadvantage . . .

Our three dwarves, all first level fighters with Survivor speciality, and the Protect ability, face a challenging series of caves, containing violent enemies:

twelve Fire Beetles, 120xp, an easy encounter
five Hobgoblins, 200xp, an average encounter
two Bugbears, 280xp, a tough encounter
an Ogre and two Bugbears, 520xp, a killer encounter (i.e. one that breaks the rules on how much to throw at the PCs)

This is beyond a normal day, so we expect some heroic casualties. But will the drunken dwarves pay for their tomfoolery?

First, how would the dwarves do without a visit to the inn before the adventure . . .

Adventure results for 10000 teams of brave dwarves:
 TPK:                  19%          

 Defeated:              6%
      two dwarves died      0%   {1 events}
        one dwarf died      1%   {58 events}
         badly injured      1%   {100 events}
     moderate injuries      4%
        light injuries      0%   {3 events}

 Completed adventure:  75%
      two dwarves died      0%   {13 events}
        one dwarf died      3%
         badly injured      5%
     moderate injuries     47%
        light injuries     20%

Now here are some intoxicated Dwarves, they met in the tavern, had a drinking contest, and rather than sleeping it off, went straight into the caves (probably a drinking game dare!) . . .

Adventure results for 10000 teams of brave intoxicated dwarves:
  TPK:                  16%

  Defeated:              1%     {145 events}
       two dwarves died      0%   {2 events}
         one dwarf died      1%   {61 events}
          badly injured      0%   {4 events}
      moderate injuries      0%   {75 events}
         light injuries      0%   {3 events}

  Completed adventure:  83%
       two dwarves died      1%   {39 events}
         one dwarf died      7%
          badly injured      1%   {130 events}
      moderate injuries     36%
         light injuries     38%
          not a scratch      0%   {29 events}

. . . conclusion? Unfortunately, that intoxication can be significantly beneficial to a defensive fighting style based on Protect and Parry abilities. Applying disadvantage to all D20 rolls is not enough of a penalty to offset the extra damage absorption.

 

[slobo777]

Without using Protect (which is a little niche to have three protectors in an optimised team), there is still a difference, but things get more complex and interesting!

I've optimised used of Deadly Strike here. It is now preferred on a hit, but not used wastefully on creatures with low hit points (such as the Fire Beetles, or a Hobgoblin that has already been hit)

Same dwarves, same adventure, first lot without performance enhancing substances:

Adventure results for 10000 teams of brave dwarves:
  TPK:                  27%

  Defeated:             30%
       two dwarves died      0%   {10 events}
         one dwarf died      2%   {234 events}
          badly injured      8%
      moderate injuries     20%
         light injuries      0%   {2 events}

  Completed adventure:  43%
       two dwarves died      0%   {31 events}
         one dwarf died      4%
          badly injured     12%
      moderate injuries     24%
         light injuries      3%

Back down the pub, and try again . . .

Adventure results for 10000 teams of brave intoxicated dwarves:
  TPK:                  35%

  Defeated:             11%
       two dwarves died      0%   {23 events}
         one dwarf died      4%
          badly injured      1%   {112 events}
      moderate injuries      6%
         light injuries      0%   {5 events}

  Completed adventure:  54%
       two dwarves died      1%   {110 events}
         one dwarf died     10%
          badly injured      6%
      moderate injuries     30%
         light injuries      7%

. . . although the drunken dwarves actually succeed more often than their sober friends, and if they do tend to suffer less injury, there's a lot more party deaths in the second group. This is due to disadvantage on Death Checks. Once a drunken dwarf is taken down, all that extra damage previously absorbed is causing system shock*, and they tend to die rather than stabilise. In addition, their colleagues take longer to mop up the battle because they miss more often, so don't always get to use the healer's kit in time.

* just fluff of course, but I kind of like how there is payback for the earlier resistance to damage .

 

[GreyICE]

From these results it doesn't appear that three protectors are too niche to be an optimized team, it looks like three protectors ARE an optimized team.

 

[slobo777]

From these results it doesn't appear that three protectors are too niche to be an optimized team, it looks like three protectors ARE an optimized team.

My opinion on that is that Parry is nearly always better option than Deadly Strike. And Intoxicated is plain silly - perhaps drop the damage soak to 1 point . . .

Anyway, this IMO is why the strategy of pouring alcohol into Guardian dwarves works so well:

Stacking Protect, Parry and Intoxicated, to reduce incoming damage on a hit by 3d6 at first level. That's great against small numbers of heavy-hitting enemies (couple of bugbears)

Any fighter can Parry one attack. Stacking it with Intoxicated gives 2d6 damage soak, pretty good versus middling damage where you might be facing ~1 monster per PC (hobgoblins)

Intoxicated has the added advantage of working well on it's own against multiple small attacks. Mobbing the PCs with Fire Beetles, Kobolds or Zombies won't be too dangerous either.

So the team has all basic combat encounters covered - at least those without major special abilities in play. Such a team will have a weak spot on saving throws and exploration though . . . my code isn't sophisticated enough (and won't be for a while) to show that.

 

[GreyICE]

But the guardians look better even without the booze. I mean cutting the booze from the equation, they're STILL succeeding 75% of the time while the "balanced and optimized" team is succeeding 43% of the time.

Is there a flaw in the code, or is it just better (at the moment, very early in the playtest cycle) to build Fighters 'ard as rok? I can't see their utility from being 'ard diminishing in any way in a team situation.

 

[Dausuul]

I also am not a fan of Intoxicated. Unless you're playing some kind of drunken master character, it seems like a joke mechanic, more appropriate to Munchkin-type games. I'm not sure why it's still in the playtest.

 

[Blackwarder]

Intoxicated should have some kind of disadvantage, maybe upon the end of being intoxicated the character need to pass a con saving throw or faint.

I like the idea of playing a drunken fighter, and we all know that dwarves are practically functioning alcoholics so having three drunken dwarves wipe the floor is thematically pleasing to me.

Warder

 

[slobo777]

But the guardians look better even without the booze. I mean cutting the booze from the equation, they're STILL succeeding 75% of the time while the "balanced and optimized" team is succeeding 43% of the time.

Is there a flaw in the code, or is it just better (at the moment, very early in the playtest cycle) to build Fighters 'ard as rok? I can't see their utility from being 'ard diminishing in any way in a team situation.

Always possible there is a flaw in the code, but if so, it's not on basic stuff such as rolls to hit, or dice used to Parry etc. I can set it to dump a full, detailed log of adventure (and filter by category, e.g. see what happens in a "one dwarf dies, team wins" scenario). Most likely flaw at the moment is that I have misinterpreted a rule or forgotten something (I forgot to apply disadvantage to Death Checks initially, and spotted it in one of the dumps).

There are plenty of weaknesses in the model, though. Stuff I have modelled very basically or not at all. Combat is just a random whirlwhind, no choices of action (it's always melee attack) or movement. No allowance for spells and missile weapons. That means there are plenty of options I cannot compare.

Despite this, I honestly think that Parry and Protect need a slight adjustment down in power level. Just to make Deadly Strike and other uses of ED more equally viable.

Although a gang of three Guardians standing shoulder-to-shoulder can face off ogres at level 1, I'm not too concerned about that. It's getting into territory that won't be realistically played, like 10 Wizards all casting Magic Missile.

 

[Blackwarder]

The geek in me find your models interesting but I wouldn't change stuff in the playtest like parry or protect based on it.

Warder

 

[Patryn of Elvenshae]

The geek in me find your models interesting but I wouldn't change stuff in the playtest like parry or protect based on it.

... why not?

 

[Blackwarder]

Because the model is to simplistic and doesn't reflect decision making process during the game.

The only thing that is not dependent upon the situation or the actions of the players and DM on these models is the intoxicating condition which is a passive condition.

Warder

 

[slobo777]

Things a computer model cannot easily tell you about a particular option

• whether it is fun to play (irrespective of how it is balanced)
• whether it is easy and simple to use during the game (because computers can do a lot of maths very easily)
• whether it gives a nice "feel", related to the fluff

one concern about reducing power level of Parry might be that it could make using ED more complex. As it stands you either use an Expertise Die as-is, or just write off the whole die. It's simple, and keeps things moving.


Anyway, my OP is basically about Intoxicated, don't want to get too sidetracked . . .

 

[Blackwarder]

Anyway, my OP is basically about Intoxicated, don't want to get too sidetracked . . .

Which I totally agree can be tested on a computer model and even without it is easy to see that it's not working as intended (I think...)

Warder

 

[Mattachine]

In my playtests, dwarves can't get intoxicated, since they are immune to poison.

 

[slobo777]

In my playtests, dwarves can't get intoxicated, since they are immune to poison.

Poor dwarves! The numbers are almost identical for humans.

 

[Patryn of Elvenshae]

Because the model is to simplistic and doesn't reflect decision making process during the game.

And yet, if the model is telling us that constantly reserving your Expertise Dice for Parry / Protect is far-and-away stronger than reasonably intelligent use of, say, Deadly Strike, then the model is providing a useful service.

 

[ZombieRoboNinja]

And yet, if the model is telling us that constantly reserving your Expertise Dice for Parry / Protect is far-and-away stronger than reasonably intelligent use of, say, Deadly Strike, then the model is providing a useful service.

The model tells us that when you're in a group full of fighters, built with high HP, facing melee opponents, using Parry is a good strategy.

However, if you're facing off against a wizard guarded by a few orcs, the better strategy would probably be to pick off the wizard with Deadly Strike first.

If you're in a mixed party with a wizard and cleric, you may be better off as a slayer using Cleave and (a fixed) Glancing Blow to take down melee units threatening your squishy friends, rather than just fighting defensively while they get swarmed.

If you're fighting a bunch of archers hiding behind battlements, those archer powers will start to look pretty tempting. And so on.

I agree that Parry is powerful. In fact it's probably very often a good strategy to parry EXCEPT when you have an opportunity to take down a dangerous opponent more quickly.

And of course all these equations change more based on attributes, equipment, race, specialties and level.

 

[Patryn of Elvenshae]

The model tells us that when you're in a group full of fighters, built with high HP, facing melee opponents, using Parry is a good strategy.

Amended:

The model tells us that when you're a fighter, facing melee opponents, using Parry or Protect is the best strategy.*

And, given the rules on Parry / Protect, it'll probably work on ranged opponents, too.

So, I don't think the test is quite as limited as you're proposing.

Moreover, I'm responding to the notion that no changes should be made to the rules based on the outcome of these models, ever, as [MENTION=6688285]Blackwarder[/MENTION] was proposing.

This is exactly the sort of modeling that I hope WotC is doing, and if they aren't, then they should absolutely look at [MENTION=6694877]slobo777[/MENTION]'s results and say, "Hey - is this really working the way we want it to?"

* So far. Maybe additional work will suss out the usefulness of Deadly Strike or something else, but it's really hard to argue with always-available, stacking DR.

 

[Connorsrpg]

[MENTION=6678226]Mattachine[/MENTION]

You beat me to it. AS soon as I saw the OP I was thinking the same The rule about not only immune to damage but to effects made me think a LOT about dwarves and drinking contests. I truly am not a fan at all of straight immunity.

I am fine with dwarves being able to drink a lot, but to never suffer the effects sucks. How do 2 dwarves go about a drinking contest? Why drink if you don't suffer the effects? I thought about all this when writing up games for our Swallowtail festival in the Rise of the Runelords modules.

We have since changed to Advantage.

 

[Blackwarder]

Amended:

The model tells us that when you're a fighter, facing melee opponents, using Parry or Protect is the best strategy.*

And, given the rules on Parry / Protect, it'll probably work on ranged opponents, too.

So, I don't think the test is quite as limited as you're proposing.

Moreover, I'm responding to the notion that no changes should be made to the rules based on the outcome of these models, ever, as [MENTION=6688285]Blackwarder[/MENTION] was proposing.

This is exactly the sort of modeling that I hope WotC is doing, and if they aren't, then they should absolutely look at [MENTION=6694877]slobo777[/MENTION]'s results and say, "Hey - is this really working the way we want it to?"

* So far. Maybe additional work will suss out the usefulness of Deadly Strike or something else, but it's really hard to argue with always-available, stacking DR.

I'm not saying don't change anything, I clearly said before that this model clearly show that intoxicated is a huge bonus and probably not working as intended.

I sincerely hope that WotC does not make any decisions based solely on computer models (and if the do use computer models I hope they don't use simplistic one such as this), a computer model might be used to validate the conclusion from the surveys and play testing but they shouldn't be the sole arbitrator.

Warder