In Pathfinder 1e, solo boss enemies had great difficulty being relevant due to the number of ways in which such an enemy could be trivially eliminated by a typical four-PC party. 2e, from what I have observed, goes the opposite direction: solo boss enemies are absolutely terrifying due to the ways their numbers are stacked against the party, and due to how easy it is for such monsters to critically hit PCs. This effect is especially apparent when it comes to monsters who do little except for toss out physical attacks that do raw damage and nothing else; since they land critical hits often, that damage becomes frightening.
I think a good way to demonstrate this would be to take a generic physical attacker monster and use it as the centerpiece for an "extreme" difficulty encounter, under the bestiary's guidelines: "Extreme-threat encounters are so dangerous that they are likely to be an even match for the characters, particularly if the characters are low on resources due to prior encounters. This makes them too challenging for most uses. An extreme-threat encounter might be appropriate for a fully rested group of characters that can go all out, for an end-of-campaign encounter, or for a group of veteran players with powerful character teamwork."
In theory, such a battle should be 50/50 between the PCs and the monsters. However, in the case of generic physical attacker monsters whose attacks do raw damage and nothing else, I am convinced that the party's chances of victory are actually closer to 1% or so, short of the entire party being nothing but horse-mounted characters with the luxury to kite away indefinitely on a flat plain.
Take a four-PC, 2nd-level party, then. They have standard 2nd-level wealth as per page 348, and Suppose they are at full resources, they have a modicum of preparation ("Usually, you can let each character cast one spell or prepare in some similar way if they have the drop on their foes,"), and they confront an ettin or a weak hill giant in a relatively "neutral" environment. Does the party actually stand a 50/50 chance of victory, or is it much lower?
The highest possible attack bonus at 2nd-level is a fighter with ability modifier 4 + level 2 + expert 1 = +7. Make that +8 if they have a casting of magic weapon from an arcane, divine, or occult spellcaster. The highest possible non-shield AC at 2nd-level is base 10 + level 2 + Dexterity modifier 3 + breastplate 4 = 19. A 2nd-level human fighter with Constitution 12 and General Training (Toughness) has HP 34.
An ettin has Perception +14 for initiative, and two turns per round, though only two actions per turn. They have AC 20, TAC 18, HP 105, and speed 35. They have reach 10 feet, Attack of Opportunity, and an effective two Reactions per round. On a regular hit, they deal 2d6+5 damage (average 12). On a critical hit, they deal 4d6+10 damage (average 24). Remember those two turns per round, tracking multiple attack penalty separately for each, and remember those Attacks of Opportunity with two Reactions per round.
A weak hill giant has Perception +12 for initiative. They have AC 20, TAC 17, HP 115, and speed 35. They have reach 10 feet, but thankfully, no Attack of Opportunity. On a regular hit, they deal 3d10+3 damage (average 19.5). On a critical hit, they deal 6d10+5 damage (average 38 damage), enough to one-shot even a barbarian.
A magic weapon-boosted fighter's highest-attack-bonus attack hits with a regular hit 40% of the time, and a critical hit 5% of the time. The ettin and the weak hill giant's attack bonus is +15, so on their highest-attack-bonus attack hits a Dexterity 16 breastplated PC with a regular hit 50% of the time, and a critical hit 35% of the time.
I dont know the odds as I am not a math guy but I will throw in my 2 cents. I have seen a lot of posts about how powerful monsters are in the this edition.
For me this is a huge plus. I like my players to be worried about dying. Scary monsters make characters that much more heroic. If I want to run and easier game it is very easy to run lower level mobs against the players.
My only gripe about the current system (which appears to work very much liek the starfinder system) is that monsters of a particular CR are very, very similar. The monster creation system kind of makes it so there are just 3 monsters at ever CR ad caster, fighter, and skill monkey. I dislike that Pit Fiend and Balor Demons are basically identical because they are both CR 17.
IF they could work a little more variety into each CR level I would be happy.
I haven't done any in-depth analysis, but here's my initial analysis of the scenario:
The Monster's side: (Ettin only)
The party of 4 level 2 PCs will have an incredibly difficult time winning these fights. The Ettin hits on a 4 vs. a PC with AC 19, so its accuracy is 85%/65% and 85%/65% for its maximum 4 attacks per round, and 75% for each of its 2 possible Attacks of Opportunity.
So on each of its turns, the Ettin is expected to deal 22.4 damage to a PC, or 2/3rd of the Fighter's hp in the example (34).
Expected damage for the AoO is:
0.5 * (12) + 0.25 * (24) = 12
Which is a bit more than 1/3rd of the Fighter's hp.
At low levels, PCs have fewer options to throw around, including fewer buffs, items etc. With how much damage the Ettin is expected to throw out (downing the Fighter in 3 attacks, on average), there is no good way to brute force this thing down. Even if the Ettin doesn't attack in round 1 and allows the PCs to surround it, once round 2 begins, it can start downing 1 person per round.
That's assuming the PCs are all Fighters with 19 AC and 34 hp.
The Party's side:
In reality, if the party is optimized for an encounter of 4 levels higher, they will eschew with spells and abilities that rely on a successful check or save to work, because the monster's saves will probably be too high for those to have an effect. For this exercise, I will pick a party of Bard/Cleric and 2 Shield Fighters.
The PCs start the fight with both Fighters having Magic Weapon cast on their weapons. Then the Ettin will probably have 2 turns in a row because its Perception check is so high (+14 vs. +6 for the Cleric, and +3-5 for the others). It'll spend its first turn moving up to melee range, and then its second turn almost killing a Fighter.
Then, the battle begins. Having the Ettin take 2 turns in a row may be a blessing in disguise, because it can't use its 2 reactions effectively.
If it closes in to melee range, the PCs don't have to waste movement getting to it. With Delay, the PCs can sculpt a good turn order, like Bard/Cleric/Fighter/Fighter.
The Bard will start with Inspire Courage and try to use Lingering Composition to lengthen the duration, then he will cast Bane. From this point on, each turn where he needs to, he will re-cast Inspire Courage, concentrate on Bane and shoot at the Ettin from range.
The Cleric will stay out of range as much as possible, spamming Heals with Healing Hands to keep the Fighters topped up (3 actions for the ranged Heal, for 2d8+4 healing, or 13 per turn). He has Bit of Luck as his domain power, ready just in case.
The Fighter who got hit will Step in range and Strike and Raise his Shield. His comrade will follow suit, Stepping twice if need be to flank and avoid AoOs. If that's not possible, he will eat up an AoO and get in flanking position and raise his Shield.
The Ettin now has -2 AC, -1 attack from Bane and the Fighters have +2 AC from raising their shields. It could try to move to attack the casters, but then it will eat 2 AoOs from the Fighters. It can stay and try to down the Fighters using its 4 attacks. If it focuses on 1 target, the expected damage is 32.6 using all 4 attacks. On an already wounded Fighter, it's likely to down him, unless he uses Shield Block to prevent a hit or gets luck, or both.
The Fighter's expected damage on the Yeti is as follows:
To hit: +2 (lvl) + 1 (expert) + 1 (item) + 4 (strength) + 1 (Conditional) = +9 vs. AC 17 (from Flat-Footed). He hits on an 8, or 65% accuracy, to deal an average of 2d8+5 damage (+1 conditional damage).
Attack sequence will be Strike-Strike-Raise a Shield:
Expected damage is:
11.2 + 6.3 = 17.5 per turn.
With both Fighters going at it, the Ettin is expected to die in 3 rounds.
Meanwhile, the Cleric is spending all his actions healing, the Bard is Inspiring Courage and plinking away with not so accurate Shortbow shots (2 lvl + 3 Dex + 1 Expert shortbow + 1 conditional = +7 vs. 19 because the Bard isn't flanking). Screening may also be involved. He's contributing about 2~ damage per round this way. It's probably better for the Bard to whip out the Demoralize action with Intimidating Glare to try and Frighten the Ettin.
His bonus is 4 (Cha) + 2 (level) = +6 vs. 21 Will DC. It's a tall order, but the Bard has a 25% chance to reduce the Ettin's AC by 1, and a 5% chance to reduce it by 2, as well as further reduce its Attack rolls by 1. This is better than firing a Shortbow, so the Bard keeps doing this.
Now it's a question of if the party can survive for 3 rounds until the Ettin dies.
The probability of that happening is... unclear, since there are many variables involved. If the Ettin can't down a Fighter, whether through unlucky rolls, good healing, bad tactics, then it will die in 3 rounds. However, if it gets lucky and downs a Fighter early, the fight quickly goes its way, and there is little left to control the damn thing.
Another tactic the PCs might try is for the Cleric to cast Bless and maintain the attack bonus while the Bard tries something different. He uses Inspire Competence targeting one of the Fighters, who tries to Trip the Ettin. Using a Trip weapon like the Flail (and thus lowering his expected damage a tiny bit), he can try to land a trip with a bonus of +9/+11 depending on if the Bard critically succeeds with his Aid Reaction. This gives a 55-65% chance to hit the Ettin's 19 Reflex DC.
If the Ettin is Prone, the fight becomes a lot easier. It can stay prone and take a further -2 penalty to its attacks, thus allowing the Fighters to survive longer, or it can stand up, waste an attack (and lose out on about 8 damage) and provoke 2 AoOs from the Fighters (Standing up is a move action).
Note that the Ettin can also employ this tactic, sacrificing 8 damage in 1 turn to try and Trip the Fighters (with a MUCH higher success rate, as it's rolling +13 Athletics vs. 16 Reflex DC), and then getting a free AoO when the Fighter tries to stand up. In fact, this is probably what it'll do each turn, because the AoO's accuracy is higher than the 2nd Strike, and it also eats up an action from the Fighter.
Conclusion:
This is a dire fight. The PCs have few tools in their arsenal, and their enemy can down them in few hits. If it employs proper tactics, the fight is nigh impossible. However, if the PCs get lucky and the monster uses poor tactics, it is possible to win in about 3 rounds.
My prediction at a glance: the PCs have less than 30% chance of winning the fight, some of it because they're low level and don't have many options yet.
Their actual odds aren't great. The encounter guidelines need a little asterix warning "Be very careful when pitting a low level party against a single monster whose cr is equal to double the party level or higher.
Sure, it says a party of four level 1s should be able to take out a cr5 enemy, but it will be ugly.
Now, as soon as you get past level 4 when PCs have a magic weapon, more HP etc, those severe and extreme guidelines settle down quite a bit and it gets manageable. Don't get me wrong, they won't be easy fights, but the rules just have way too low a margin of error at low levels.
I haven't done any in-depth analysis, but here's my initial analysis of the scenario:
The Monster's side: (Ettin only)
The party of 4 level 2 PCs will have an incredibly difficult time winning these fights. The Ettin hits on a 4 vs. a PC with AC 19, so its accuracy is 85%/65% and 85%/65% for its maximum 4 attacks per round, and 75% for each of its 2 possible Attacks of Opportunity.
So on each of its turns, the Ettin is expected to deal 22.4 damage to a PC, or 2/3rd of the Fighter's hp in the example (34).
Expected damage for the AoO is:
0.5 * (12) + 0.25 * (24) = 12
Which is a bit more than 1/3rd of the Fighter's hp.
At low levels, PCs have fewer options to throw around, including fewer buffs, items etc. With how much damage the Ettin is expected to throw out (downing the Fighter in 3 attacks, on average), there is no good way to brute force this thing down. Even if the Ettin doesn't attack in round 1 and allows the PCs to surround it, once round 2 begins, it can start downing 1 person per round.
That's assuming the PCs are all Fighters with 19 AC and 34 hp.
The Party's side:
In reality, if the party is optimized for an encounter of 4 levels higher, they will eschew with spells and abilities that rely on a successful check or save to work, because the monster's saves will probably be too high for those to have an effect. For this exercise, I will pick a party of Bard/Cleric and 2 Shield Fighters.
The PCs start the fight with both Fighters having Magic Weapon cast on their weapons. Then the Ettin will probably have 2 turns in a row because its Perception check is so high (+14 vs. +6 for the Cleric, and +3-5 for the others). It'll spend its first turn moving up to melee range, and then its second turn almost killing a Fighter.
Then, the battle begins. Having the Ettin take 2 turns in a row may be a...
Not sure how much it will change the expected damage but the ettin is listed as having 2 flails, and each head only controls 1 arm which means each head has to choose to use only flail or only fist.
I do not see what is stopping each of the ettin's heads from controlling a single flail.
I probably worded that badly, what I meant was both heads can have a flail but they then lose the ability to use their fist for an agile weapon on second attacks. Basically the decision is +14/+9 at 2d6+5 or +14/+10 at 2d4+5.
I think the wording in the "extreme" entry should be changed. It says and "even" match for the players but that'd obviously not the case. As you keep reading it says that it is "too challenging to use in most cases". How is that even?
'Even' in the context of an encounter is a 50% chance of the party losing and likely all dying.
That said, I must agree that single huge monsters are a bigger threat than usual, though the Ettin's particular action economy makes it specifically and unusually deadly vs. lower level characters. A manticore is also level 6 and likely a somewhat less severe threat vs. a PC group geared up to take one down.
This is an example where "expected damage" falls down.
The outcome of the ettin opening up on the AC19 fighter with 34 HP on average is less important than the answer to the question of how often the fighter will actually have a turn rather than being knocked out.
A monte-carlo simulation of turn one: (ettin moves up, flails, then flails and fists), shows that just under half the time, the fighter won't get a turn alive.
Out of 300000 simulations, we have:
Fighter is knocked out after the second flail attack: 84082 times
Fighter is knocked out after the fist attack: 56933 times
Fighter survives: 158985 times.
If the ettin instead went after that AC17, 30HP bard (or Nethys forbid, the cleric), things go south fast:
Bard is knocked out after the first flail attack: 3217 times
Bard is knocked out after the second flail attack: 156349 times
Bard is knocked out after the fist attack: 70433 times
Bard survives: 70001 times.
As another way to look at it, supposedly, to a four-PC, 2nd-level party, a group of eight goblin warriors would be as deadly an enemy force as an ettin or a weak hill giant. Is that really the case though? I have no doubt that the goblins could prove vicious, especially with massed fire from their deadly d10 shortbows, but would the goblins truly be as terrifying as the ettin or the weak hill giant?
As yet another way to look at it, allegedly, a four-PC, 4th-level party would consider sixteen goblin warriors to be an extreme encounter, with only a 50/50 chance of success for the party. Meanwhile, facing off against a stone giant or the Sandpoint Devil, both 8th-level melee attackers with little but damage, would also give 50/50 odds. Yet I am doubtful that the goblin warriors would be anywhere as terrifying as the ever-critting melee beatstick.
Well let's see using some very rough math. Here's how it went.
First round of 16 against average AC 20, 3 attacks
6 hits (average 18 damage)
1 crit (average 11 damage)
Assume players kill 1 goblin per character so 4 per round, you could argue more or less.
Second round of now 12 against AC 20
7 hits (average 21 damage)
no crits
Third round of now 8 against AC 20
no hits
no crits
Fourth round of now 4 against AC 20
2 hits (average 6 damage)
Fifth round they are all dead.
That's a grand total of a possible 112 damage but an average damage of 56-ish. So I mean in this instance they may drop a party member or 2. Although more Crits could very quickly have made this a much bigger issue, mainly due to the shortbows. Also to be fair they did get a lot of 18s and 19s in my rolling so there could have been at least a couple more hits if they were against lower AC party members like Spellcasters.
I wonder if Orc Brutes or Kobold Warriors would fair as well though.
This is an example where "expected damage" falls down.
The outcome of the ettin opening up on the AC19 fighter with 34 HP on average is less important than the answer to the question of how often the fighter will actually have a turn rather than being knocked out.
A monte-carlo simulation of turn one: (ettin moves up, flails, then flails and fists), shows that just under half the time, the fighter won't get a turn alive.
Out of 300000 simulations, we have:
Fighter is knocked out after the second flail attack: 84082 times
Fighter is knocked out after the fist attack: 56933 times
Fighter survives: 158985 times.
If the ettin instead went after that AC17, 30HP bard (or Nethys forbid, the cleric), things go south fast:
Bard is knocked out after the first flail attack: 3217 times
Bard is knocked out after the second flail attack: 156349 times
Bard is knocked out after the fist attack: 70433 times
Bard survives: 70001 times.
141015 fighters and 229999 bards were killed in the making of this post.
I think your computation is off; I obtain your results using the following attack sequence: flail +14 2d6+5, flail +9 2d6+5, fist +6 2d4+5.
This is not how the ettin works. Thanks to his independant brain ability, he ettin is actually 2 creatures with two action each; his attack sequence is something like flail +14 2d6+5, flail +9 2d6+5, fist +14 2d4+5, fist +10 2d4+5 (or we can replace the fist attacks by other flail attacks, since the ettin has two flails; this doesn't change the result very much). Now the fighter has 81.3% chance of being dead at the end of turn 1. We can tweak the situation: the ettin has to move, so we remove the last fist attack. Now the fighter has only 67.8% chance of being dead at the end of turn 1. In the other hand, the ettin has 2 flails, so I guess each head may use flails attacks. In the other hand, the fighter has to approach the ettin - otherwise the ettin can attack the bard - , so we can use the 4-attack sequence and add an AO; now the fighter has 91% of being dead, but at least he had some time to attack once (since he was the one closing to melee) (all my computations assume he raise his shield - hence the fighter has only one attack when he close the distance with his shield raised).
Now, let's assume the 2 fighter attack the ettin, shield raised, each of them suffer an AO (since the ettin has 2 AO per turn), then attack once, and then the two part of the ettin attack. The tool I use for my computation can't give an exact results (I can't specify that both ao have to go on a different fighter) (actually, I'm not sure. As i understand the independant brains ability, each head can make an AO when a fighter approach), but it seems there's more than 1/2 chance that both fighter are dead at the end of the ettin's turn.
I think the ettin is a bit more powerful than 8 goblin warriors.
Deadmanwalking wrote:
That said, I must agree that single huge monsters are a bigger threat than usual, though the Ettin's particular action economy makes it specifically and unusually deadly vs. lower level characters. A manticore is also level 6 and likely a somewhat less severe threat vs. a PC group geared up to take one down.
Err... No.
The manticore has a higher attack bonus (+15) and higher damages. And an agile secondary weapon. In the end, she has 66.8% chance of killing a level 2 fighter in 1 full attack. She doesn't have AO, the fighters don't take damages while charging into melee, in the other hand she has a ranged attack so she doesn't need to move and it's unlikely the fighter have their full health in melee.
The manticore has a higher attack bonus (+15) and higher damages. And an agile secondary weapon. In the end, she has 66.8% chance of killing a level 2 fighter in 1 full attack. She doesn't have AO, the fighters don't take damages while charging into melee, in the other hand she has a ranged attack so she doesn't need to move and it's unlikely the fighter have their full health in melee.
Not to mention, she flies.
Given that, by your math, the Ettin apparently has an 81.3% chance of killing them on a full attack and the Manticore apparently has a 66.8% chance of doing the same, I'm not seeing how that proves me wrong. In fact, that seems to prove me almost precisely correct in my eyeball analysis.
I wasn't saying it was easy to take down a Manticore, I was just saying that Ettins were even worse than usual vs. a low level party. Which your math seems to back up (an almost 15% greater chance of taking out a Fighter in the first round is a pretty big difference).
The manticore has a higher attack bonus (+15) and higher damages. And an agile secondary weapon. In the end, she has 66.8% chance of killing a level 2 fighter in 1 full attack. She doesn't have AO, the fighters don't take damages while charging into melee, in the other hand she has a ranged attack so she doesn't need to move and it's unlikely the fighter have their full health in melee.
Not to mention, she flies.
Given that, by your math, the Ettin apparently has an 81.3% chance of killing them on a full attack and the Manticore apparently has a 66.8% chance of doing the same, I'm not seeing how that proves me wrong. In fact, that seems to prove me almost precisely correct in my eyeball analysis.
I wasn't saying it was easy to take down a Manticore, I was just saying that Ettins were even worse than usual vs. a low level party. Which your math seems to back up (an almost 15% greater chance of taking out a Fighter in the first round is a pretty big difference).
You're right, I had in head the computation I did a few days ago against level 6 PC (where the damages of the Ettin end the Manticore are almost the same); and the number stored in my head was the probability the ettin kills a fighter with 3 attack (66%). I re-saw the numbers afterward, and it was too late to edit my post. :/
Spoiler:
A few days ago I did the same kind of computation (ettin/manticore), but against level 6 PCs (AC 24), for someone else on another forum. And it appears the ettin deals a few more damages in average, but not that much, and with fewer ouliers. That's because the manticore deals more base damages and her first attack crits on 19-20 against AC 24.
Hence, at level 6 the monk or the rogue should go against the ettin while the fighter go against the manticore - the ettin has less chance to OTK a character with "meh" defenses. And that's very strange when you consider the ettin as a melee brute while the manticore is a ranged harasser (ranged attack + an ability to stay out-of-melee): usually, the role of the monk is to go for the ranged harasser thanks to his superior mobility, while the fighter goes for the melee brute because no one else can stand such a punishment.
But but but, you're right, for level 2 character, melee with a manticore is far less dangerous than melee with an ettin.
Anyway, this is something I don't like in Path 2: it doesn't define monsters roles (and derive base stats from level and role). Every level 6 monsters has the same stats as any other level 6 monster - and the melee brute = the ranged harasser. There are some very conservative tweaks, like "the ettin attacks at -1/-1/-6/-6 (I consider the ettin as having a -1 penalty on every attack, since he's the only one with +14 instead of +15 at level 6) instead of 0/-5 or -4/-10 or -8", but in the end the ettin ends up with almost the same melee capability as a flying ranged monster.
141015 fighters and 229999 bards were killed in the making of this post.
I think your computation is off; I obtain your results using the following attack sequence: flail +14 2d6+5, flail +9 2d6+5, fist +6 2d4+5.
This is not how the ettin works. Thanks to his independant brain ability, he ettin is actually 2 creatures with two action each; his attack sequence is something like flail +14 2d6+5, flail +9 2d6+5, fist +14 2d4+5, fist +10 2d4+5 (or we can replace the fist attacks by other flail attacks, since the ettin has two flails; this doesn't change the result very much).
My underlying assumption was that the first head needed to use an action to move up to the fighter/bard, then would attack with a flail at +14. The second head would then attack with flail at +14, followed by fist at +10.
Feel free to issue a pull request if these assumptions have changed.
From the above I guess the first question is, is there a precast spell/alchemical item used in advance/etc. which can delay the ettin? You need to avoid that first full attack. Caltrops might help in some situations.
If you can somehow outrun the ettin then persistent damage is nasty - acid flasks do 1d4 persistent damage and an ettin isn't especially good at getting rid of it. There are some spells which might inflict persistent damage too.
Going toe-to-toe with an ettin at level 2 looks like a terrible idea but maybe a 2nd-level party mounted on horses in the open, with some prep, could take one.
Edit: no, wait, riding something costs you an action each round so forget the horses. Elves with nimble for a start I guess. The principle to keep in mind is "A strange game. The only winning move is not to play."
My underlying assumption was that the first head needed to use an action to move up to the fighter/bard, then would attack with a flail at +14. The second head would then attack with flail at +14, followed by fist at +10.
Feel free to issue a pull request if these assumptions have changed.
ARG!
I think I've made a mistake when computing the crit chance. My computation are wrong.
So, let's use the following sequence for the ettin: Fail +14/+14/+9/+9 (2d6+5) (I assume if the ettin has one flail in each hand, he can't use his fist; I assume the two independant brains control a different arm) (the last attack being done with the fist or the flail doesn't change the results that much anyway).
Ettin moves ( I remove his last attack): 50.6% to kill a fighter in one round (0.004% chance to kill two).
Same, but the fighters raise their shield: 33.7% to kill a fighter in one round (0.001% chance to kill two).
Ettin full attack: 66.9% to kill a fighter in one round (0.67% chance to kill two).
Same, but the fighters raise their shield: 48.5% to kill a fighter in one round (0.27% chance to kill two).
Ettin waits the fighter and use AO (assuming each head can make an ao on the same fighter) and then full attack: 92.5% to kill a fighter in one round (24.4% chance to kill two: now we're talking).
Same, but the fighters raise their shield: 81.1% to kill a fighter in one round (9.7% chance to kill two).
Hence it's highly dependent on the situation: if the fighters have to charge the ettin to prevent him from attacking the healers, I think they're screwed. Otherwise... I think they screwed as well, but the fighters may resist several turn.
The most "realistic" situation is, I think, a situation I can't really compute with my tool: the ettin engage the first fighter, but the second fighter has to come at range and suffers 2 ao (or loses several action to avoid AO). Or the fighters may use reach weapons, in which case they don't have shields...
OK, so that's the first impact. Now, two fighter are in melee, I assume they are healed of 16 HP/round (soothe + heal) (the cleric can't use 1-action-heal because of AO) (my tool doesn't allow me to set a random value for heal - it was conceived to handle fast healing); once a fighter falls, they stop healing him (since he'll lose 1 round to stand again and then he'll drop again). Not counting the damage from the initial impact (my tool doesn't allows for different attack sequence each round; yes, I know, it's a limited tool, but computing the exact probabilities was hard enough):
full attack from the ettin, regen 16 HP/round for fighters: round 1 66.9% chance the first fighter falls. Round 2 93.7% the first fighter falls (or have fallen at first round), 48.2% both fall. Round 3 98.9% the first fighter falls (or has already fallen), 82.9% both fall (bla same precision). Round 4 99.8%/95.6%.
the same, but the fighters have shields: round 1 48.5% chance the first fighter falls. Round 2 81.1% the first fighter falls (bla), 25.4% both fall. Round 3 93.6%/57.3%. Round 4 97.9%/79.3%. Round 5 99.3%/90.9%.
I don't think we need more to evaluate if the group has more or less than 1/2 chance to win; now we need to evaluate the number of round they need to defeat the ettin (and to evaluate if the fighter should have shield or a greatsword or a reach weapon).
OK, a fast evaluation: I assume the first fighter stands for 2 rounds and the second one stands for 3 rounds or 4 rounds (depending if they are using shield or not) (those are the avarages values of the number of rounds before they fall, rounded up). That's a total of 5 or 6 round of full attack from the fighters. The cleric casted two magic weapons, the bard sings, the fighters attack with +9 (2d12+5 with greatswords or 2d8+5 with longswords).
With greatswords, they have 24.3% chance to kill the ettin before they both fall ; there's a 50% chance they have inflicted 80+ damages: maybe the cleric and the bard can finish him? (and I didn't use the last action of the cleric; a sling maybe?)
With longswords, they have 12.9% chance to kill the ettin before they fall (with 50% chance they have inflicted 70+ damages).
Of course, if the cleric didn't prepare magic weapon, they're doomed...
That's more close than what I though; anyway, it seems it confirm the intuition I had when looking at the different Level+2 fight in the playtest (the manticore, the weakened golem...): many monsters seems to be just 1 Level too low. So those Level+2 fights correspond to Level+3 fights (weaker than 4 PCs, but strong chance of TPK if anything goes wrong), and this ettin should be Level+5 against a level 2 party (stronger than the 4 PCs - you need planning and luck to win).
Anyway, it doesn't remove the base problem: Path 2 is Optimization Hell Edition. We used the best possible group to challenge the Ettin, and it seems we don't have 50% chance to beat him. Now, let's assume one player wants to play a Alchemist instead of a Bard, and another one wants to play a Barbarian with high Con/med Dex instead of a fighter: there's no way they can beat the campaign end boss (and probably no way they can attain him)...
Spoiler:
let me recompute for the manticore, with the right number for crits.:
full attack vs the AC 19 HP 34 fighter: 55.9% OTK.
