D&D 5E lowest level to fight a marilith

[Tales and Chronicles]

Ok, then I'll try with this, to make a more interesting battle encounter since 5th ed is not the best for solo boss monster.

I'll take:
1 Warlock of the Fiend (CR7 caster, Volo's p 219), and add some yuan-ti abilities to it (Magic resistance, constrict, Invoke nightmare)
6 flying swords (CR 1/4 MM p. 20)
1 Priest (yuan-ti) (CR 2, MM p. 348)
2 Accolyte (yuan-ti) (CR 1/4 MM p. 342)

Hope this will be interesting. As some have said, maybe they wont even fight her, instead trying to bargain with her (Warlock pact RP possibilities?) or flee, making her a recuring threat in the game.

 

[hawkeyefan]

I had my PCs run into a Marilith at level 4. It was meant as a skirmish...they offended her in Sigil and later on bumped into her in the gate town of Plague-Mort. She toyed with them a bit before they managed to hit her for a little damage. At that point, she got furious and was going to destroy them...but luckily for the PCs the Great Modron March arrived and they were all swarmed by Modrons, separating the PCs from the Marilith. She'll be showing up later in my campaign.

I think that the party needs to be of sufficient level to have many PCs with multiple attacks, otherwise her party ability is going to shut them down pretty effectively. I think that level 10 is probably a solid bet for how low the party could be and still have a good shot. If the players are particularly effective, maybe as low as 8. Though I wouldn't expect a win to be certain at that level.

 

[MostlyDm]

A quick sim (https://repl.it/EnFq/11) with very non-strategic, non-optimized characters gives me the following results:

Two 5th level Champion tanks with nonmagical longsword + shield, plus two 5th level Champion archers with nonmagical longbows, win 0 out of 100 combats against a Marilith.

If you double the number of PCs (4 5th level tanks and 4 5th level archers), they win 13 out of 100 combats.

If you bump them up to 9th level (2 9th level tanks and 2 9th level archers), they win 2 out of 100 combats.

If you bump them up to 9th level and give them all +1 magic weapons (+1 longswords and +1 longbows), they win 84 out of 100 combats, with 2.42 members still above 0 HP on average.

If you bump them up to 9th level and add another archer (2 9th level tanks and 3 9th level archers) and give them all +1 weapons, they win 100 out of 100 combats with 3.92 members still above 0 HP on average.

Code is here: https://repl.it/EnFq/11

Conclusion: if I haven't made any major mistakes, then it looks like you could do a 9th level fight with four or five PCs and get what you want, especially if you give them magic weapons.

Awesome stuff. I think your Champion metric is pretty clearly useful for this sort of encounter calibration.

Did the 9th level make a significant difference? If you dropped them to 7th or 8th but let them keep the +1 weapons, do their survival chances plummet or just drop a few points?


Edited because my posts keep getting messed up formatting. Adding quote breaks in odd places.

 

[FormerlyHemlock]

Awesome stuff. I think your Champion metric is pretty clearly useful for this sort of encounter calibration.

Did the 9th level make a significant difference? If you dropped them to 7th or 8th but let them keep the +1 weapons, do their survival chances plummet or just drop a few points?


Edited because my posts keep getting messed up formatting. Adding quote breaks in odd places.

Losing a level costs 1 point of proficiency bonus, 8 HP, and a point of second wind:

// champion8b was generated using PHB standard array and variant human Fighter 5 with Dueling Style and plate armor + shield + longsword +1 and Heavy Armor Master feat
let champion8b() = Combatant(nameOf humanNames "Champion", (20, 14, 14, 10, 11, 8, 68), AC=18,
                            Traits=[HeavyArmorMaster; ActionSurge; AthleticsProficient; ImprovedCritical], Prof = +3,
                            Actions = [Action.Create("Melee attack",
                                            Attack [
                                                Attack.CreateMagic "slashes" 9 [DieRoll.Create(1, 8, 6)]
                                                Attack.CreateMagic "slashes" 9 [DieRoll.Create(1, 8, 6)]
                                                ])
                                       ],
                            BonusActions = [
                                Action.Create("Second Wind", Healing (DieRoll.Create(1, 10, 8)), 1)
                            ])
// archer8b was generated using PHB standard array and variant human Fighter 5 with Archery Style and studded leather and longbow +1 and Sharpshooter feat
let archer8b() = Combatant(nameOf humanNames "Archer", (14, 20, 14, 10, 12, 8, 68), AC=17, Prof = +3,
                            Traits=[ActionSurge; AthleticsProficient; ImprovedCritical],
                            Actions = [Action.Create("Shoot",
                                            Attack [
                                                Attack.BestOf(Attack.CreateMagic "shoots" 11 [DieRoll.Create(1, 8, 6)], Attack.CreateMagic "headshots" 6 [DieRoll.Create(1, 8, 16)])
                                                Attack.BestOf(Attack.CreateMagic "shoots" 11 [DieRoll.Create(1, 8, 6)], Attack.CreateMagic "headshots" 6 [DieRoll.Create(1, 8, 16)])
                                                ])
                                        ],
                            BonusActions = [
                                Action.Create("Second Wind", Healing (DieRoll.Create(1, 10, 8)), 1)
                            ])

This had a lot bigger effect than I was expecting.

With four PCs:

evalGroup [marilith] [champion8b;champion8b;archer8b;archer8b]

Grizzabella the Champion and Elias the Champion and Katie the Archer and Ruprecht the Archer win 50 out of 100 matches against Raitor the Marilith, with 2.16 members still alive on average

I ran the sim multiple times and that's not a fluke outcome.

With five:

evalGroup [marilith] [champion8b;champion8b;archer8b;archer8b;archer8b]

Robin Hood the Champion and John the Champion and Rupert Grint the Archer and Robin Hood the Archer and Katie the Archer win 90 out of 100 matches against Kmorduk the Marilith, with 3.27 members still alive on average

So yes, that extra point of proficiency makes a big difference in the victory rate.

Note BTW that there's a mistake in the sim: the Marilith gets her parry bonus even against the archers' attacks. I haven't bothered to fix it though because I would house-rule the Mariliths to do exactly this at the table, because parrying an arrow is awesome.

 

[Caliban]

Now sub one of the melee Champions with a Paladin (vengeance oath) - assume the paladins have sword+shield & full plate, either defensive or duelist style, and casts Protection from Evil and Good on himself and declares the Marileth his enemy via Oath of Vengeance. If you have the feat to spare, take Resilient Con.

 

[Bigsta]

My Dark Sun campaign PCs are level 5.

Last night they were fighting a Hezrou (sp?) demon that killed the spell-less Rangers pet Jakar during the fight.

The fight was taking place next to a demonic portal. During the fight, the Bard tied a rope around the dead Jakar and tosses it into the portal.

In my mind, a nearby sucubus saw the Jakar body and intended to resurrect it as a demon pet, so I told the Bard he felt a tug on the rope. On the next turn the Bard told me, "I pull on the rope."

It was then I realized my fatal mistake, a sucubus is a devil, not a demon. Being the quick thinking DM I am, I decided that a Merilith made the most sense for what I had pictures in my mind.

As such, I had the Merilith emerge from the portal and say, "are you trying to take my puppy!!!!" The Bard, cowering in fear, dropped the rope and said "no ma'am." The Merilith then went back through the portal with her soon to be resurrected purse demon Jakar to show off to all her friends a demonic tea time.

So to answer your question, not level 5.

 

[FormerlyHemlock]

Now sub one of the melee Champions with a Paladin (vengeance oath) - assume the paladins have sword+shield & full plate, either defensive or duelist style, and casts Protection from Evil and Good on himself and declares the Marileth his enemy via Oath of Vengeance. If you have the feat to spare, take Resilient Con.

Perhaps later. I haven't added spells yet, and right now it's looking like I'm probably not going to until I've got a web UI up and running for building combats. I made some good progress yesterday, but there's still a ways to go. It won't be less than several days.

 

[Caliban]

Perhaps later. I haven't added spells yet, and right now it's looking like I'm probably not going to until I've got a web UI up and running for building combats. I made some good progress yesterday, but there's still a ways to go. It won't be less than several days.

No worries. I didn't mean to word it like a command. I suspect that having a paladin able to tank the marileth will have a significant impact, assuming he can keep her attention for a round or two.

 

[FormerlyHemlock]

Agreed. Tactics make a big difference, and imposing disadvantage is huge. (Cancelling out the Marilith's advantage from restraining is less huge but still useful.)

 

[Tales and Chronicles]

Another question: are solo boss monsters as static/unimpressive as many people say? I fear that since Marilith dont get lair actions/legendary resistance, it may be a short uninteresting combat. BUT, if I add some lair action to make the combat less static, I end up with an even higher CR.