D&D 5E Grind-out fights, unconscious heroes, and retreat

[transtemporal]

I think part of the sense of doom was because I wasn't hiding the fact that I thought the situation had gotten out of hand, but also wasn't making any indications that I planned to put it back "in hand." I figured I was watching myself administer a TPK. And I may have cackled every time a villain landed a hit on the 19 AC plate dwarf. In a cartoonish french accent.

I think a little dose of reality is good for players from time to time. Pen & Paper is good for exactly the reason that the story can develop in context, e.g. if it seems appropriate that every guard in the castle is alerted because you launched a frontal attack so be it. They can always get their invincible, level-appropriate fix from an MMO if they need it.

I do like the possibility of a mechanic where some sort of mechanic is earned after a certain amount of time in combat. Might be too gamist for some, but, if done right, it could be fun. On the other hand, since I'm trying to get my players to cut their losses more often, the idea of introducing a special combo that opens up after 5 rounds of combat seems a little antithetical.

Ah yes, that's true. I'm amazed they didn't run. We ran from the dragon in the second round (after it feared 3 of us, then breathed on and killed the monk outright) although I guess in that fight, we were absolutely sure we couldn't beat it.

 

[Gnarl45]

@OP: Your players are either brave or reckless and that's why they don't run away. Normally when one PC goes down, they're that much closer to a TPK and that should make your players nervous. Isn't that kind of the point of a hard fight?

I was wondering about something. Do your players know that some fights aren't meant to be won without trickery or traps and that others might not be winnable at all? Both of those are really fun but it can get frustrating for your players if they started in the 3e/4e era where every fight is supposed to be "level-appropriate" (according to the DM guidelines anyways).

 

[werecorpse]

The problem you want to solve is that the party either suffers TPK / near TPK or is fine the next day.

I suggest the solution is a type of lingering injuries - based on failed death saves. That imposes a longer lasting result for getting whipped but won't result in death. It seems to me to be the middle ground you are looking for.

 

[FormerlyHemlock]

I think intelligent monsters are actually the other way around. Unless they're zealots of chaos or completely feral, they are reluctant to kill because they want to show enough strength to drive off the invaders, but not give the invaders an incentive for revenge or back them into a corner where mutually assured destruction is the only course.

I think a "no quarter given or asked for" situation would be uncommon, since everyone wants to live and theres always the chance you'll be on the losing side...

I've actually got a formalized dueling code for encounters between mutually hostile-but-not-mindlessly-hating parties, like githyanki and githzerai, or drow and githyanki, or duergar and yuan-ti. The idea is that there has to be a way of conducting limited exchanges without mutually annihilating each other--or else why wouldn't they have annihilated each other already? And it's both natural and game-appropriate for that method to be based off of single combat.

The rules:

* When you meet a hostile force, you offer "kresim" (life-price, like a bag of gold) and indicate who your champion is, usually not your leader but a second-in-command. The other party does the same.
* The two champions fight until one of them is down or surrenders.
* The loser's life now belongs to the winner, but if his life is spared, the winner takes kresim instead. (Conveniently explains why githyanki and drow carry around treasure per the DMG treasure tables.) Then the loser can be stabilized/revived as normal (healing or rest).
* If one side decides that they can take the other side in open combat, ignore all the above and just launch an attack.

This way, evil nations can defend their borders against other nations while gathering intelligence on each other's relative capabilities, and it doesn't cost them 100% of their population every year. It also lets good fighters gain status, which has secondary effects on society.

My players are dimly aware of this convention since a githyanki offered them 'kresim' (the wizard dominated the githyanki champion and made him kill himself with a sword, which was awesome, unexpected, and quite intimidating--so the githyanki acted rather friendly afterwards) but they don't know all the details.

 

[FormerlyHemlock]

Being knocked unconscious, in of itself, just isn't that scary. Sure, there's something like a 3% chance that you'll crit-fail one death save and then fail another one, killing you in 2 rounds. But, on average, an unconscious character has 4 or 5 rounds of death saves, and a 55% chance that they will stabilize without any intervention whatsoever. There's even a 5% chance that they'll pull through on their own and be back up on their feet, swinging that greatsword. When you're a monster, repelling invaders, and that's the most you've got to scare your invaders with, you're not going to have much success.

Fun fact: someone starting from 0 failure and 0 successes on death saves has a 41.4% chance of stabilizing at 0 HP, a 40.5% chance of dying, and an 18.1% chance of getting back on his feet with 1 HP before five rounds are up. (F# code follows.) All in all, he's got a 59.5% chance of not dying.

If he's starting at 2 failed death saves due to a melee crit, the chances are 12.5% chance of stabilizing, 78.8% chance of death, and 8.8% chance of getting back to 1 HP.

// Return list of fractions: Stable | Dead | Up 
let rec stabilize(fail, success) =
    if success >= 3 then
        [1.; 0.; 0.]
    elif fail >= 3 then
        [0.; 1.; 0.]
    else
        let vals = [0.; 0.; 0.05] // 5% chance of immediate up due to natural 20
        // recursion weighted by probability
        let failVals = stabilize(fail+1, success) |> List.map((*)0.4)
        let critFailVals = stabilize(fail + 2, success) |> List.map((*)0.05)
        let successVals = stabilize(fail, success + 1) |> List.map((*)0.5)
        vals |> List.mapi(fun i prob -> prob + failVals.[i] + critFailVals.[i] + successVals.[i])
stabilize(2, 0)

 

[Sunseeker]

Some thoughts:

Retreat is a result of two situations:
1: Poor party planning.
2: Poor DM planning.

There can be a lot of reasons for poor party planning: Maybe they didn't pay attention, maybe they're not high enough level, maybe they're just kinda dumb, maybe they utilized their class features and tactics poorly. But poor party planning is most often the quickest route to failure.
If the party was at the appropriate level for the challenge, prepared well enough, played well enough then the possible alternative is the DM may have messed up in their encounter design, as I think you OP, encountered in your game. Poor DM planning is thankfully adjustable on the fly, perhaps the sudden influx of new monsters results in a decrease in their tactical capabilities, or not all of the new monsters are on the same side or other reasons as necessary.

In my experience retreat comes in two forms:
1: Sensible party decisions.
2: DM intervention.
-Capture instead of kill by the opponents is a form of DM intervention.

In my experience, parties are rarely sensible. Even the smartest players and smartest characters can be overcome by a few bad decisions or the sight of treasure. And of course there's the age of axiom of: no plan survives contact with the enemy. So generally speaking parties will choose the option with the highest potential for reward, which is often the option with the lowest chance of success. Since the party usually chooses to take the high-risk-high-reward option first, this means that when retreat becomes the only viable solution to the situation, the party has risked all and lost. This tends not to set well with parties, so it doesn't happen.

DM intervention is a touchy subject. Some people are "victory or death!" sorts and they tend not to appreciate the DM stepping in, in whatever manner and saying "Sorry guys I'm going to take victory off the table for you because really it ain't gonna happen." Well, who is the DM to make that judgement call? Sometimes the party survives the worst of situations surprisingly better than you'd expect so it feels unfair for the DM to jump in and call the fight. Some folks are OK with the DM jumping in and saving the party's ass wen a fight goes sour, so long as it's pretty obvious that victory is such a far-fetched possibility that there's really no alternative.


Killing downed party members:
This really comes down to the sort of monsters you're running. I tend not to kill downed party members when I have the chance, and there's a variety of reasons for this: Perhaps the beasts you're fighting like live food. Perhaps the villains want some slaves or minions or to interrogate you. Perhaps I'm just a softie.

I think there are very few monsters that would kill you when you're down. Strategically its the best move but the problem is that very few monsters are that highly strategic. Unless the monster has specific orders to make sure you're dead, once you're down most of them are going to assume you'll stay down and move on to killing the rest of the party. Highly strategic monsters should not be played as genre-savvy. Genre-savvy villains suck. They're boring to run and they're even worse to fight because they'll always always go for the kill. Highly strategic monsters are likely going to want to use the party for strange rituals, for fun and for torture and potentially interrogate them, giving the party ample opportunity to escape thanks to stupid henchmen and faulty prisons.

----

In the game I've been playing there's been a lot of retreat. We're not a very strategic party, several players blatantly don't know their class and we're easily swayed by the lure of big treasure. The DM, however is fairly generous with us and gives us ample opportunity to retreat. We don't always take it, and tonight he had to force us to retreat ala: powerful NPC intervention. Which was fine in several ways, the DM got to point us in the direction we should be headed in order to level and gear up and raise our reputation with the world, etc... It also answered some unanswered story questions and generally the party was okay with not being dead.

The DM almost never kills downed players unless the enemy has a specific reason to want to be sure they're dead.

It's important to remember that in 5E, being "downed" or "at 0" doesn't mean you're easy to kill. You maintain your AC, but lose your dex bonus, and by the rules you either have to: deal damage equal to that person's maximum hit points (-100% health) OR deal damage 3 times (each time you take damage counts as a failed death save). There's no auto-kill rules for someone to walk over, slice your throat open and boom, you're dead. While there's nothing stopping this from happening, it's worth keeping in mind that doing so often comes off less as "You're severely wounded, this guy wants to make sure you don't get back up." as the DM saying "I don't want you to live."

 

[FormerlyHemlock]

Some thoughts:

Retreat is a result of two situations:
1: Poor party planning.
2: Poor DM planning.

There can be a lot of reasons for poor party planning: Maybe they didn't pay attention, maybe they're not high enough level, maybe they're just kinda dumb, maybe they utilized their class features and tactics poorly. But poor party planning is most often the quickest route to failure.
If the party was at the appropriate level for the challenge, prepared well enough, played well enough then the possible alternative is the DM may have messed up in their encounter design, as I think you OP, encountered in your game.

I don't think party failure equates to poor planning the way you suppose here. A disastrous retreat that goes wrong in an avoidable way is likely due to poor planning, no doubt about that--but good planning also builds in SOP for retreats when unexpected obstacles are encountered. "Whoa, illithids?!? I thought we were here to retrieve a widget from the Valley of the Giant Ape! Nobody said anything about illithids! Time to pull back and re-evaluate the mission parameters. Yo, PCs, let's go to fighting retreat configuration and pull back to point Eta." That's good sense, born of good planning including planning when and how to abort. It may include the bard casting Longstrider on everyone while the wizard casts Invisiblity on everyone and the warlock holds his action to Repelling Blast anyone who comes to close, while all party members begin moving at high velocity towards the valley exit to the north. Without that planning, you may discover that half your party charges and the other half retreats--take your pick which half of the disaster you want to end up on.

Unless the monster has specific orders to make sure you're dead, once you're down most of them are going to assume you'll stay down and move on to killing the rest of the party.

That may not be a good assumption. When I ran the math earlier in this thread, I was surprised to find that there's nearly a one in five chance of a downed creature popping back up if you don't give it an extra whack or two to make sure it stays down. For a hunting monster, "popping back up" could mean "losing your meal", so it might be realistic for them to learn to give it an extra swipe. If you need a justification, there it is.

It's important to remember that in 5E, being "downed" or "at 0" doesn't mean you're easy to kill. You maintain your AC, but lose your dex bonus

This must be a house rule at your table. By PHB rules, you keep your Dex bonus (very odd I know) but any attackers get advantage to hit you, unless they're ranged attackers in which case the disadvantage from prone will cancel out advantage to hit. Any hits in melee range count as crits due to the Unconscious condition, so it only takes two hits to kill you for sure, or one hit to put you at two death save failures (80% chance of dying before stabilizing).

 

[Libramarian]

Isn't any discussion of house rules to make death saves more difficult kind of academic when 5e Clerics have Spare the Dying and Healing Word?

I think any house rules that nerf natural stabilization/healing will only make Clerics (and other healing classes, to a somewhat lesser extent) more necessary, without affecting parties that do have a Cleric much.

 

[S'mon]

What I do:

1. Group init roll - single d20 roll by side, but *add individual init mods*. Gives the best of both worlds; fast PCs still have an advantage but it resolves very quickly and the PCs will usually be acting en masse as a team, not as dissociated individuals. This makes tactical coordination, including retreat, far far easier. Going forward I'm planning to use this in all future campaigns, it is possibly the best house rule I've ever come up with.

2. Negative hit points, not death saves - Your hp can go negative and you die at negative hp = your max hp. No healing from 0. Attacks vs helpless are auto-crit, but no death saves. In practice this MASSIVELY ENHANCES PC survivability, because (a) monsters chewing on helpless PCs will often take several rounds to kill them and (b) without pop-up PCs or fast kills, intelligent monsters have no incentive to target the fallen, so they focus on PCs still standing. And w/out death saves, in case of a lost battle PCs will often be captured, not killed. I do use a death save after 1 hour unconscious - pass and you have short rested & can spend hit dice to recover, fail & we roll again in 1 hour, you only die after 3 consecutive fails. Also if you were taken to 0 by a crit from a piercing or slashing weapon you are bleeding out & I'll use the RAW.

edit: auto-crits from slashing/piercing while helpless also impose a death save fail, or 2 death save fails with a dagger.

 

[redrick]

Some thoughts:

Retreat is a result of two situations:
1: Poor party planning.
2: Poor DM planning.

There can be a lot of reasons for poor party planning: Maybe they didn't pay attention, maybe they're not high enough level, maybe they're just kinda dumb, maybe they utilized their class features and tactics poorly. But poor party planning is most often the quickest route to failure.
If the party was at the appropriate level for the challenge, prepared well enough, played well enough then the possible alternative is the DM may have messed up in their encounter design, as I think you OP, encountered in your game. Poor DM planning is thankfully adjustable on the fly, perhaps the sudden influx of new monsters results in a decrease in their tactical capabilities, or not all of the new monsters are on the same side or other reasons as necessary.

In my experience retreat comes in two forms:
1: Sensible party decisions.
2: DM intervention.
-Capture instead of kill by the opponents is a form of DM intervention.

In my experience, parties are rarely sensible. Even the smartest players and smartest characters can be overcome by a few bad decisions or the sight of treasure. And of course there's the age of axiom of: no plan survives contact with the enemy. So generally speaking parties will choose the option with the highest potential for reward, which is often the option with the lowest chance of success. Since the party usually chooses to take the high-risk-high-reward option first, this means that when retreat becomes the only viable solution to the situation, the party has risked all and lost. This tends not to set well with parties, so it doesn't happen.

DM intervention is a touchy subject. Some people are "victory or death!" sorts and they tend not to appreciate the DM stepping in, in whatever manner and saying "Sorry guys I'm going to take victory off the table for you because really it ain't gonna happen." Well, who is the DM to make that judgement call? Sometimes the party survives the worst of situations surprisingly better than you'd expect so it feels unfair for the DM to jump in and call the fight. Some folks are OK with the DM jumping in and saving the party's ass wen a fight goes sour, so long as it's pretty obvious that victory is such a far-fetched possibility that there's really no alternative.


Killing downed party members:
This really comes down to the sort of monsters you're running. I tend not to kill downed party members when I have the chance, and there's a variety of reasons for this: Perhaps the beasts you're fighting like live food. Perhaps the villains want some slaves or minions or to interrogate you. Perhaps I'm just a softie.

I think there are very few monsters that would kill you when you're down. Strategically its the best move but the problem is that very few monsters are that highly strategic. Unless the monster has specific orders to make sure you're dead, once you're down most of them are going to assume you'll stay down and move on to killing the rest of the party. Highly strategic monsters should not be played as genre-savvy. Genre-savvy villains suck. They're boring to run and they're even worse to fight because they'll always always go for the kill. Highly strategic monsters are likely going to want to use the party for strange rituals, for fun and for torture and potentially interrogate them, giving the party ample opportunity to escape thanks to stupid henchmen and faulty prisons.

----

In the game I've been playing there's been a lot of retreat. We're not a very strategic party, several players blatantly don't know their class and we're easily swayed by the lure of big treasure. The DM, however is fairly generous with us and gives us ample opportunity to retreat. We don't always take it, and tonight he had to force us to retreat ala: powerful NPC intervention. Which was fine in several ways, the DM got to point us in the direction we should be headed in order to level and gear up and raise our reputation with the world, etc... It also answered some unanswered story questions and generally the party was okay with not being dead.

The DM almost never kills downed players unless the enemy has a specific reason to want to be sure they're dead.

It's important to remember that in 5E, being "downed" or "at 0" doesn't mean you're easy to kill. You maintain your AC, but lose your dex bonus, and by the rules you either have to: deal damage equal to that person's maximum hit points (-100% health) OR deal damage 3 times (each time you take damage counts as a failed death save). There's no auto-kill rules for someone to walk over, slice your throat open and boom, you're dead. While there's nothing stopping this from happening, it's worth keeping in mind that doing so often comes off less as "You're severely wounded, this guy wants to make sure you don't get back up." as the DM saying "I don't want you to live."

First of all, I whole-heartedly disagree with the idea that retreat must be the fault of either poor party planning or poor DM planning. Retreat need not be the result of player or DM failure. Furthermore, as a DM, I make no pretense whatsoever that the "encounters" in my campaign are level appropriate, and I usually don't plan them strictly as encounters at all. I use some discretion and math in determining how many hostiles of which types to put in a location, but once all the PCs and NPCs are hanging out, things just kind of run their course.

Generally speaking, I am a far worse tactician than all but your most unintelligent of monsters, so the sudden influx of creatures does decrease my tactical capabilities (as I'm now more likely to forget about special combat abilities from at least one of the mobs). Beyond that, I don't go out of my way to adjust-down a fight just because it's over the heads of the party. I might go out of my way to tell the party just how dangerous their opponents are.

I've forced a retreat once and, honestly, it was a terrible session and I wish I hadn't done it. I don't think, as the DM, that I should force the players to keep their characters alive. However, once the players decide to get out of Dodge, I'm happy to work with them on it. Retreat opens up so many more options for creative problem solving than your standard combat.

I also disagree with you that attacking downed PCs is a genre-savvy or meta-gaming action on the part of monsters. A PC who has gone to 0 hp is almost like a boxer who has been knocked down. They're briefly out of it, but also clearly still alive and breathing. A monster would see this, and understand instinctively, "this guy's still a potential threat." Even without the possibility of magical healing. (Which monsters are also well aware of, since they live in the same world of magical healing that the player characters do.) Genre-savvy is understanding the unique motivations of PCs as controlled by detached otherworldly beings stuffing pizza and beer down their faces and taking several minutes to plan out every 6-second action. Basic awareness is understanding the general mechanical underpinnings of the world you live in.

I think there are plenty of reasons why a monster wouldn't attack a downed PC, but I think many of us, my earlier self included, tend to rationalize away the reasons why a normal opponent would attack a downed opponent. I'm re-evaluating those reasons. Will the monsters stand over the unconscious PC hacking and slashing until that PC fails their final death save? In the middle of combat? Probably not. But will they use their second attack as part of multi-attack to give another hack at that PC as he falls to the ground? Absolutely. Happens in medieval war movies all the time. The character might still survive that follow-up hack, but it makes the stakes a lot higher.

What I like about what I'm describing is that it's not a house rule. It uses the death save mechanic exactly as written. Even the unconscious PC still has some defenses against monster attacks, and that's as it should be. Falling unconscious in combat should be really dangerous. Characters should be able to die in a combat that isn't a total party kill. This will never happen in 5e once you get past 2nd level if the DM applies some sort of Red Cross/Geneva Convention to monsters.

What's the point of a "victory or death!" approach if there is actually no risk of death?