5E DMG's definition of "Deadly" is much less deadly than mine: Data Aggregation?

Plus, even then, the sharpshooter is only picking off at most a number of hobgoblins equal to their attacks/round. That gives them a fair bit more time to make the front line PCs' and the big flashy wizard's lives difficult.
Well, kind of. If the big flashy wizard is still doing big flashy wizard things instead of lying prone behind total cover and letting the Sharpshooter handle it, he is in my opinion engaging in incorrect counterplay.

Then again, players always do counterplay wrong. And their PCs usually survive anyway, because 5E's "Deadly" is calibrated to be not-too-deadly. :)
 

I'm A Banana

Potassium-Rich
Yes, in fact they do. Quoting from the Basic Rules (which are the same as the DMG but easier to copy/paste):



It then gives the daily XP table. That table is for adjusted XP, not raw XP. Poor writing on their part to refer to it as "XP", they should have called it "Difficulty value" or something to avoid confusion.
Thanks for the catch! It's interesting. I was reading this line:
DMG said:
For each character in the party, use the Adventuring Day XP table to estimate how much XP that character is expected to earn in a day.
(emphasis mine)

Since you don't earn the adjusted XP (DMG: "This adjusted value is not what the monsters are worth in terms of XP; the adjusted value's only purpose is to help you accurately assess the encounter's difficulty"), I figured the intent was raw XP. But the table definitely lists it as adjusted, as does the sentence before the table.
This provides a rough estimate of the adjusted XP value for encounters the party can handle before the characters will need to take a short or long rest.
Going by adjusted XP, one encounter with 4 carrion crawlers eats up the day's XP for six 2nd-level characters. If you kick it up to 5, you're close to your back-of-envolope calculation of 4x Deadly (4,500 adjusted XP, vs. 4,800 4x Deadly).

I mean, I'd go with 8, but maybe I'm mean. ;) At any rate, 5-8!
 
Thanks for the catch! It's interesting. I was reading this line:
I know! It's terrible writing.

I mean, I'd go with 8, but maybe I'm mean. ;) At any rate, 5-8!
As a player, I think eight is about the right amount to be fun. It's worthwhile to fall back and try tricky things like luring some of them out with decoys (monk!) so you can divide and conquer, whereas with a regular Medium-ish encounter like 2 Carrion Crawlers the whole thing will be over in eighteen seconds, and I don't find that fun.

To me, Medium/Hard = usually boring.
 
Last edited:
A lot of good advice of the "it depends" variety. Yes, I know, a thousand times over I know. Veteran DM, I get that CR is not a perfect system, blah, blah, blah. It's an important point, but it has been discussed to death online.

And ultimately, "it depends" doesn't help me. My mind, as a veteran DM, is already trained to keep an eye out for the "it depends" elements.

However, I'm a veteran DM pretty new to 5e still.

What I need is a metric that works more accurately than the DMG for encounter difficulty estimating. It don't expect it to be perfect, but I expect most of the time that I should be able to predict when an encounter will be easy/medium/hard/deadly using the definitions giving for those difficulties in the DMG (which, I will note, the DMG adjusted experience seems to produce easier encounters than these definitions).

Hemlock said:
In my experience, the point of actual 50%-chance-of-death-in-the-absence-of-really-smart-tactical-play comes when there are an equal number of N level PCs and CR N monsters, which works out at approximately 4x the official Deadly threshold. That's just at the raw mechanical level and there are a number of factors that can tilt the balance toward either side to make Deadly come sooner (e.g. exploit 120' drow darkvision and poison against a melee-centric party) or to make 10x Deadly encounters survivable (Necromancer undead skeleton archers), but as a general guideline I find that Quadruple-Deadly fights make you have to think hard about how you'll survive while anything less is a straightforward "kill the enemy and take their stuff."
Awesome, thanks for putting some hard numbers to it.

So you're saying 400% of the DMG Deadly threshold is ACTUALLY DEADLY?

This contrasts noticeably with [MENTION=158]Henry[/MENTION]'s suggestion of 120% of the Deadly threshold is DEADLY.

I wonder why such a large discrepancy.
 
The dmg guidelines are for 4 PCs I think. So... Yeah of course they stomped it
For crying out loud. Did you read what I posted? There are adjustments you make in the encounter XP "budget" (hate that term) - specifically outlined in the DMG - for increased party size. I obviously used those, and I even included a snapshot of the page from Kobold Fight Club where those calculations were done. Geez!
 
Thanks for the catch! It's interesting. I was reading this line:

(emphasis mine)

Since you don't earn the adjusted XP (DMG: "This adjusted value is not what the monsters are worth in terms of XP; the adjusted value's only purpose is to help you accurately assess the encounter's difficulty"), I figured the intent was raw XP. But the table definitely lists it as adjusted, as does the sentence before the table.


Going by adjusted XP, one encounter with 4 carrion crawlers eats up the day's XP for six 2nd-level characters. If you kick it up to 5, you're close to your back-of-envolope calculation of 4x Deadly (4,500 adjusted XP, vs. 4,800 4x Deadly).

I mean, I'd go with 8, but maybe I'm mean. ;) At any rate, 5-8!
To pin you down on this a little, it sounds like you're agreeing with [MENTION=6787650]Hemlock[/MENTION] that to present a fresh party a Deadly challenge, the DM needs to up the Deadly XP threshold by 400-500%. Is that correct?
 
So you're saying 400% of the DMG Deadly threshold is ACTUALLY DEADLY?

This contrasts noticeably with @Henry's suggestion of 120% of the Deadly threshold is DEADLY.

I wonder why such a large discrepancy.
Yes, that's what I'm saying. I wouldn't mind hearing Henry's perspective myself. Speculation: maybe Henry uses more gimmicky monsters like Ghouls and Banshees? My Quadruple Deadly observation is based around the basic meat-sack monsters in the MM but there are a handful of monsters with special abilities that can be deadly in smaller numbers. I alluded to this with the drow--due to sleep poison and Bounded Accuracy, plus drow darkvision, even relatively small numbers of drow can be deadly. (I took out a mid-level Necromancer and Shadow Monk, despite animated skeletons, using eight drow warriors in the dark, which I believe was officially an Easy encounter at that level if you don't count the Elite Warrior who didn't even do anything different from the regular mooks. That felt really good.)

It's possible that Henry's DM experience instinctively reaches for the tougher monsters within a given CR level.

P.S. The party doesn't necessarily have to be "fresh" to handle Quadruple-Deadly. I remember one encounter with vampires and zombies that occurred at the end of a long day (several other vampire fights), without the PCs even having taken the time for a short rest[1], and the XP total for that fight when I computed it afterward came out as 130% of the daily XP encounter budget all by itself, so I think it must have been at least in the Triple or Quadruple Deadly range. It was definitely a challenging fight for the players, which they won due to exploiting the cleric's Destroy Undead feature (which made him feel awesome, destroying 20 zombies all by himself over the course of four rounds) and grappling the vampires and dragging them out into the sunlight (which made me feel good about my encounter design).

[1] Note to self: I should be more generous with just telling the players, "After clearing three houses, you stopped for lunch. That counts as a short rest, go ahead and spend HD if you want." They shouldn't have to declare to me that they eat and sleep when they've already told me they're searching the city for vampire infestations.
 
Last edited:

Uller

Explorer
I sometimes think Pass Without Trace might be the single most powerful low-level spell in the game.
It's definitely up there, along with Faerie Fire. We have a shadow monk...she's used PwoT a few times to good effect. We've nerfed surprise a bit to make it more like older editions.

Anyone who has ever had any kind of military training will tell you...the side that usually wins a fight in the real world is the side that gets the first _effective_ shot (meaning, the first side to remove an enemy from permanently from the fight)
 

I'm A Banana

Potassium-Rich
To pin you down on this a little, it sounds like you're agreeing with [MENTION=6787650]Hemlock[/MENTION] that to present a fresh party a Deadly challenge, the DM needs to up the Deadly XP threshold by 400-500%. Is that correct?
Yeah, you'll want to go well beyond the Deadly threshold in the DMG for one encounter if that encounter is meant to be your only encounter. 4x-5x would fit with the idea of doing 4 to 5 deadly encounters in a row, which would be more in line with upping the value to be worth 4 or 5 encounters for a whole day.
 

Green1

Villager
I see what Oakleaf is trying to get at here. But, I am not sure any better system for determining what is "deadly" is really possible.

I can also see "appealing to the lowest common denominator" as not really a bad thing. Not every player or DM is going to be very skilled at play or be familiar with all their character can do until they get a lot of games under the belt. So, I have no issue with CR being somewhat skewed downward.

I personally just eyeball things. I also run a sandboxy type campaign. These are the things in the area, and the PCs cope. CR will never be balanced and a science. Even if you assign a point system to each ability of monsters, PCs, ect there are too many intangibles like terrain/ battlemat set up, the skill of the players, and tactics of the DM.
 
Yeah, you'll want to go well beyond the Deadly threshold in the DMG for one encounter if that encounter is meant to be your only encounter. 4x-5x would fit with the idea of doing 4 to 5 deadly encounters in a row, which would be more in line with upping the value to be worth 4 or 5 encounters for a whole day.
Well, according to Kobold Fight Club, my 3 carrion crawler anecdote was worth 2,025 adjusted XP, versus a DMG-calculated Deadly lower end threshold of 1,200 XP.

By the way, looking at the amount of monsters (in terms of sheer XP) that you need to throw at the PCs to get a Deadly fight and (at least at level 2) it's like practically leveling up the party. Does this persist throughout higher levels too? Or is it exclusive to low levels (e.g. 1-3 or 4)?
 

I'm A Banana

Potassium-Rich
Well, according to Kobold Fight Club, my 3 carrion crawler anecdote was worth 2,025 adjusted XP, versus a DMG-calculated Deadly lower end threshold of 1,200 XP.

By the way, looking at the amount of monsters (in terms of sheer XP) that you need to throw at the PCs to get a Deadly fight and (at least at level 2) it's like practically leveling up the party. Does this persist throughout higher levels too? Or is it exclusive to low levels (e.g. 1-3 or 4)?
D&D for levels 1-4 is basically "a day or two per level." Starting at 5, it starts opening up and slowing down significantly.
 
Well, according to Kobold Fight Club, my 3 carrion crawler anecdote was worth 2,025 adjusted XP, versus a DMG-calculated Deadly lower end threshold of 1,200 XP.

By the way, looking at the amount of monsters (in terms of sheer XP) that you need to throw at the PCs to get a Deadly fight and (at least at level 2) it's like practically leveling up the party. Does this persist throughout higher levels too? Or is it exclusive to low levels (e.g. 1-3 or 4)?
There's a player at my table with two 14th level PCs right now (other players have rotated more PCs in and out so are lower level but he sticks mainly with two) and I've seen the pattern hold true at all levels up to this point. I've seen it hold in a party of four 11th-level PCs. I have not seen whether it continues to hold at 17th level when you're dealing with multiple adult red dragons (but then, I don't have vanilla red dragons in my game).

The combination of my willingness to expose my players to uber-Deadly encounters, and my policy of using kill XP instead of milestones, is a major factor in the accelerated levelling which makes it both possible and fun to have heterogenous party levels (e.g. a level 2 PC in a level 10 fight won't stay level 2 for long; he'll soon have enough XP to be level 6 although house rule says you don't gain more than one level per play session so it takes a while for his level to catch up with his XP total).

But there are some sessions where all the combat XP my players get to split is 1750 XP from one hobgoblin and a hobgoblin captain, or even less. Combat isn't a core element of my game.
 
Last edited:
So... from your experience with 5e what does it take to have an ACTUALLY DEADLY fight?
Running a 1st level game did the trick for me on a number of occasions.

Repeated TPKs does count as 'deadly,' yes?

I've told this story about HotDQ a few times, but what the heck: So I had run the playtest at Encounters as much as I could, and was very much still in the playtest mind-set when 5e Encounters started. So I ran the first few sessions of HotDQ 'straight.' Not pretty. I tried it again at a convention - not so much because I wanted to, but because the organizers settled on that scenario for the time slot - 3 sessions, 2 TPKs.

Anyway, I snapped out of it and started running the game the way I'd done back in the day, very behind-the-screen, adjusted-on-the-fly, DM-directed ("DM force" they call it sometimes) and it was awesome.
(I'll admit, running in that mode is starting to get a little old, but I'm taking a break until a project at work is done, so hopefully I'll be back to it with enthusiasm when that's over.)

That was first level, though. Sometime after that - 2nd, it appears, in your case - it reverses and the game becomes 'too easy.' If you play it 'straight,' that is. Run it to be fun, rules & dice notwithstanding, and it will be.
 

Chaosmancer

Villager
One thing I know the CR system is supposed to account for, that I can't tell if it does is the various abilities of the monsters. A CR X "beatstick" (no specials, just damage and hp) is less challenging than a CR X caster. Heck, even a CR X-2 caster I'd think.

I haven't been keeping close track but a few things I noticed from my last campaign and a convention game.

-Vampires with their ability to regen and deal damage at the same time are much harder to beat than their CR suggests
-Groups of Flameskulls are a massive force multiplier, because each one gets 3 fireballs per day, and even high level groups get dusted by 3-5 of them .
-A monster without the ability to disengage quickly is essentially dead if it can't drop a pc. Alternatively, Goblins with their stealth and bonus actions are very annoying even at higher levels if the terrain favors their guerilla warfare style. It is all about who pins who into getting hit.
-Intelligent enemies who prepare for the PC's tactics and abilities are harder than stupid enemies who don't, but weaker if they don't actually prepare anything
-Hordes are great if they don't use fireballs for whatever reason, because it spreads out the damage from classes like the fighter or paladin.
-Frakking assassins


The last one comes from a player in that campaign, he'd solo encounters. I think by the end he was a lv 6 assassin rogue lv 2 ranger (he wanted longbows). That first critical sneak attack wrecked encounters, to the point where I remember once or twice putting a big massive enemy into the fight, simply to absorb that initial attack. Solo enemies never worked.

Not sure how much it helps, but I built a deadly encounter with a vampire lord and flameskulls for a con. I wrecked the pcs so hard I still feel bad about it to this day. It was 6 lv 10 pcs in the game, though at least one left the dungeon before the fight began.
 
Running a 1st level game did the trick for me on a number of occasions.

Repeated TPKs does count as 'deadly,' yes?

I've told this story about HotDQ a few times, but what the heck: So I had run the playtest at Encounters as much as I could, and was very much still in the playtest mind-set when 5e Encounters started. So I ran the first few sessions of HotDQ 'straight.' Not pretty. I tried it again at a convention - not so much because I wanted to, but because the organizers settled on that scenario for the time slot - 3 sessions, 2 TPKs.

Anyway, I snapped out of it and started running the game the way I'd done back in the day, very behind-the-screen, adjusted-on-the-fly, DM-directed ("DM force" they call it sometimes) and it was awesome.
(I'll admit, running in that mode is starting to get a little old, but I'm taking a break until a project at work is done, so hopefully I'll be back to it with enthusiasm when that's over.)

That was first level, though. Sometime after that - 2nd, it appears, in your case - it reverses and the game becomes 'too easy.' If you play it 'straight,' that is. Run it to be fun, rules & dice notwithstanding, and it will be.
I think I'd seen you share this particular experience before, but thanks for sharing it again!

I have no trouble making the game fun for my players. And I get a lot of enjoyment from seeing what they come up with. However, like I said, sometimes their actions and/or narrative/foreshadowing call for a DEADLY fight.

Was it fun? Yes, always, that's why we play.

But could it have been more fun and lived up to what they (and I) perceived was a life-threatening encounter? Definitely, yes.

This is not an exercise in pedantry or numbers crunching for some theoretical "mental masturbation". Nor is it an obsession with needing to have every encounter conform to some GM-imposed difficulty level. It's about the players being able to say "F it, we're going into that umber hulk (or whatever) lair" knowing full well that it's going to be deadly...and it actually BEING DEADLY. Or maybe they trick their enemies into that umber hulk lair! It's about a player choosing to mouth off to, and picking a fight with, a drow priestess who has them surrounded and delivering on the implied DEADLY threat. It's about foreshadowing a dangerous villain and having that encounter actually be DANGEROUS.

I really want to know what metric I can use to predict and deliver challenges of a certain difficulty to my group going forward. So far my experience with the DMG metric is that it produces less challenging fights than advertised, though I'm not sure yet how much less difficult.

I don't know what's unreasonable or hard to grasp about that.
 
Last edited:
Anyhow, for anyone who's interested in such things, I'm tracking predicted vs. actual encounter difficulty here:

https://docs.google.com/spreadsheets/d/1AAIBDemfxoX8DTxAg2q7AzkZVZxwwONvhsxQ2eHTxXY/edit?usp=sharing

If you want to, you can use this spread sheet to track your group as well.
How do you expect this to be used? Is "actual difficulty in play" the DM's retroactive expectation of the probabilities, or is it outcome? E.g. if I think that 24 Umber Hulks (staged in groups of 1d4 per round) and a neogi wizard is a probable TPK (75% confidence), but all the PCs end up surviving[1] due to tactics, consumable items, use of terrain, and a minor rules error, does that mean it wasn't really Deadly because no one died? Or is it still Deadly because even in retrospect I think there was a high chance of PC death and they got lucky with dice and the rules error[2]?

[1] Surviving = the neogi were the ones whose morale broke, and they disengaged and ran away rather than fight to the death. About 50% of the umber hulks went down over the course of the engagement, and many others were low on health, so probably around 10-20% of them died.

[2] Rule error: a player told me that Evard's Black Tentacles had the same area as a Fireball and I believe him instead of checking. It's actually 20' on a side, about the same size as Confusion, Web, or Shatter. This had a largish impact on the battle. Honest mistake on his part and mine, retconned subsequently with karma points and a wild magic handwave.

Also, I don't seem to have edit permissions to the doc.
 
Last edited:
How do you expect this to be used?
Simple answer: I expect this to be used as a tool to help inform my encounter creation (whether prepped or impromptu) as a 5e DM moreso than the DMG encounter difficulty guidelines. Potentially it could be helpful for others too, hence why I'm sharing.

Is "actual difficulty in play" the DM's retroactive expectation of the probabilities, or is it outcome? E.g. if I think that 24 Umber Hulks (staged in groups of 1d4 per round) and a neogi wizard is a probable TPK (75% confidence), but all the PCs end up surviving[1] due to tactics, consumable items, use of terrain, and a minor rules error, does that mean it wasn't really Deadly because no one died? Or is it still Deadly because even in retrospect I think there was a high chance of PC death and they got lucky with dice and the rules error[2]?
I'd say it's more a retrospective of how everything panned out. Often this is from the DM's perspective, but just as often it can come from talking with your players about it after the game.

"ACTUAL" encounter difficulty is purely based on perception. It trusts DMs and players to use their good judgment, camaraderie, and communication skills to arrive at an accurate assessment of how a given fight went.

And, for the record, there are going to be some fights which so entirely break the mold that they're not useful for data collection purposes. And that's totally cool. Run those weird wild organic fights. For my part, I'll be restricting recording more "standard fights" that I feel confident in my/our difficulty assessment of.
 

Advertisement

Top