D&D 5E Level = Challenge Rating

[Yaarel]

@NotAYakk

Your criterion sounds about right, a standard combat should average three rounds while depleting a certain proportion of resources.

At the same time, the term "level" needs to remain consistent. Player character levels are highly calibrated for consistency and balance. Level is a measure to gauge the power of any creature − even when creatures have different features. Monsters need to use the same Level that heroes do, because Level is a unit of measurement that works.

Suppose every monster is playable, with a player character Level that is equivalent to other player characters.

There is a Level 9 monster. A standard combat would be something like two Level 9 heroes versus this monster. However, if there are four Level 9 heroes versus the monster, then the standard combat would need a more powerful monster at about Level 12.

• Two Level 9 heroes versus Level 9.
• Four Level 9 heroes versus Level 12.

Each monster is whatever Level it is, using the same metric that player characters use. The powers of the monster need to be appropriate for its Level.

The monsters need to use player Levels, because this unit of measurement is reliable and self-evident.


When comparing player Level with monster Challenge, the Proficiency is the same. Hence the combat accuracy is comparable between. The biggest difference between Level and Challenge is the amount of Hit Points.

• Player character gets about 5 Hit Points per Level (d8 or d10) plus a likely +2 Constitution = 7 Hit Points per Level.

• Monster gets double this, about 15 Hit Points per Challenge Rating!

The monster normally has way more Hit Points at the same level of Proficiency. Even a player Barbarian with a +5 Constitution is only getting 12 Hit Points per Level.

On the other hand, player character also have access to extreme Hit Point inflation. Consider the Druid Wild Shape feature that adds the Hit Points of a separate creature. Plus healing spells, Second Wind, healing potions, and other methods to inflate Hit Points.

The player Levels with the features appropriate to each can even include extreme Hit Points as part of the budget for the features of that level. Generally speaking, monsters are simple. They only need enough features to cover about 3 rounds or so. All the extra unspent budget for that level is converted into extra Hit Points.

It is possible to quantify every monster in terms of player character Levels. It helps because this gives the DM a clear understanding of how powerful the monster is.

 

[Cadence]

Tangent...

I'm looking over Goodman's Castle Amber, and there is a Caster with Polymorph that is clearly designed to be a 13th level Wizard, but has a CR of 5. So I go to look up Polymorph and it says "The new form can be any beast whose challenge rating is equal to or less than the target's (or the target's level, if it doesn't have a challenge rating)." It seems strange that if they were a PC they could be a CR 13 beast (if such a thing existed), but as an NPC only CR 5.

P.S. The CR of 5 looks low compared to what a typical level 13 caster would be from googling around, but the CR would still be well under 13 from what I can find (see also MotM casters).

P.P.S. Why aren't there any beasts above CR 8 (except the Traxigor). In any case, feel free to adjust it to where the target is obviously an 8th level caster at CR 4 or so.

P.P.P.S. If they target someone being possessed using Magic Jar, is the CR/Level that of the possessor or possessed?

 

[Yaarel]

Tangent...

I'm looking over Goodman's Castle Amber, and there is a Caster with Polymorph that is clearly designed to be a 13th level Wizard, but has a CR of 5. So I go to look up Polymorph and it says "The new form can be any beast whose challenge rating is equal to or less than the target's (or the target's level, if it doesn't have a challenge rating)." It seems strange that if they were a PC they could be a CR 13 beast (if such a thing existed), but as an NPC only CR 5.

P.S. The CR of 5 looks low compared to what a typical level 13 caster would be from googling around, but the CR would still be well under 13 from what I can find (see also MotM casters).

Yeah. The Mordenkainen statblocks for wizards are also a bit all over the place.

For example, the Apprentice Wizard statblock is clearly a Level 2 Wizard with three slot 1 spells and three cantrips including Arcane Burst (= Eldritch Blast cantrip). But it only rates as Challenge ¼.

It seems this low rating, which feels like an arbitrary guestimation, means the designers think the Wizard class sucks at combat. Heh, Gygax still haunts the low level Wizard!

Maybe if the statblock picked better slot-1 spells, the Apprentice Wizard would rate a higher Challenge? Heh, but in this case, the designers imply the spells are horribly unbalanced compared to each other.

 

[Yaarel]

In Mordenkainen, I am looking closely at the Hit Points per Challenge.

In sum:

monster Challenge ≈ player Level

The only difference is, a Challenge adds extra "bonus" Hit Points. Everything else is about the same: Proficiency, AC, Save, Attack, Damage.


I am focusing on the monsters at the entrance of each tier, at Challenges 1, 5, 9, 13, 17, and 21. Plus the highest Challenges 25 and 26, and the lowest "background levels" of 0, ⅛, ¼, and ½.

There are outliers, but most monsters are comparable to the expected average.

Challenges 0, ⅛, ¼, and ½, each add about 5 Hit Points. Thus by Challenge ½, the background totals 20 Hit Points.

Challenge 1 is still part of these modest 5 Hit Point increments, thus totals about 25 Hit Points, on average.

However, by the time of Challenges 5 thru 21, one can safely estimate the average Hit Points to be:

Challenge*12 + 20 Hit Points

This looks to be the norm for 5e 2024: 12 Hit Points per "Level" + 20, with a less steep on-ramp.

The Epic monsters at Challenges 25 and 26, such as Orcus, are almost by definition "Solo" bosses with their own monster-mash math.


Overall, the 2024 monster math is tighter than 2014, and the Hit Points are actually leaner. These extra Hit Points that monsters get in 2024 are intentional, and it is worth thinking about why it is this way.

Most of the time, it seems possible to take a normal player character, treat its Level number as the Challenge number, then add the extra Hit Points that the Challenge tends to get.

 

[NotAYakk]

No, that is completely not how it works? You are describing a system that isn't 5e and then pretending that is how 5e works then objecting that it isn't your system.

And a CR 7 monster may have a defensive CR of 1 through 13 - their HP could vary over more than a factor of 2. A monsters Challenge Rating is a very roughly a measure of roughly how much damage it will do before it is killed in a fight. So a monster with more HP needs less damage for the same CR, and a monster with less HP needs more damage for the same CR.

(Here, damage is a proxy for "threat or bother to PCs": a monster that petrifies might have that petrification "count as damage").

You cannot, and should not, map existing monster CR to an expected amount of HP or DPR. That is explicitly not how 5e existing monsters work.

And if you constrain your system for designing monsters to making something playable by a PC, you are throwing out huge chunks of monster design space and doing a bad thing.

You should split your measuring monsters and your designing monsters problems from each other. Having a system to design monsters is one thing, but unless you want to throw out all existing monsters, you will want a measuring monsters system that is more flexible (or flexible to handle existing monsters).

Existing monsters are not designed to be measured 1:1 with PCs, but meant to be measured against PC parties. This is why they (especially higher CR ones) sometimes have inflated HP compared to damage output: the HP they have determines the length of the fight, and if they don't have enough HP the fight ends after too few rounds to be interesting.

Meanwhile, lower CR monsters are often envisioned as being more than 1 foe. Using the 5e encounter balancing system this is mainly restricted to CR 3 and under outside of very high level PCs. Note that most 5e games end by level 10; so monsters of CR 4 and above are rarely in packs.

If you are creating a monster Level system meant to be measured 1:1 with PCs, existing monsters won't fit all that well. And that is ok. These monsters need not and should not be forced to emulate PC HP and damage output. They should be designed to make an encounter with an interesting level of challenge when used close to 1:1 with PCs.

This means their damage output should be calibrated to PC defences and HP to PC offence, not the other way around.

As noted, you can usually take a monster and half its HP and double its damage output to get a monster with a "similar" total threat. You do have to be careful that you don't end up with a monster that can one-shot a PC if it wins initiative, and gets one-shot if it loses initiative, as that isn't a super interesting encounter.

When comparing player Level with monster Challenge, the Proficiency is the same. Hence the combat accuracy is comparable between. The biggest difference between Level and Challenge is the amount of Hit Points.

Your accuracy is your attribute bonus plus your proficiency bonus plus other modifiers. As designers of monsters, we get to pick how they compare.

Your chance to hit factors in AC. As designers of monsters, we get to pick if their AC is higher or lower than equivalent PCs.

So yes, we have tools to control accuracy.

If you assume magic items on the side of the PCs (which most tables have) and not the monsters, we have even more accuracy tools.

What more, as "a single huge foe" is a fun thing to fight, you'll want math to handle a single monster suitable for 3-4 PCs of level X. In 5e that is a monster of CR X to 1.5X, as monsters are designed (mainly) as solo-against-a-party creatures. With a "pure" monster level system, it is likely to be level 2X-3X or something ridiculous: and that 3X monster will probably have too much single-target damage and not enough HP to make an interesting solo encounter. If you assume a linear progression of stats and proficiency it will also have an insanely high ATK modifier and save DCs.

OTOH if you permit an elite/solo sub-mechanic on top of level you can get a lower level foe which "counts as" more than one monster.

 

[Yaarel]

You cannot, and should not, map existing monster CR to an expected amount of HP or DPR. That is explicitly not how 5e existing monsters work.

The post refers to "average" monster hit points. There are "outliers" but most are reasonably comparable with average amounts of hit points. The distances mainly rely on concept.

Hit point outliers include very high where Solos conflate more than one creature, and very low where virtual hit points mitigate any damage, such as being incorporeal.

There are also tradeoffs between defense and offense.

That said, normal player characters also run the gambit from high hit points to low hit points.

Criteria for player characters can quantify the monsters in Mordenkainen to determine an equivalent player character level. Outliers can be accounted for.

Hit points are the "currency" to "purchase" powers. Any Level budget that remains unspent defaults to an amount of extra hit points. Accounting for hit points is the easiest part of determining a Level.

And if you constrain your system for designing monsters to making something playable by a PC, you are throwing out huge chunks of monster design space and doing a bad thing.

I agree, any Level system needs to apply to every monster in Mordenkainen, including the outliers. This is doable.

Outliers happen with player characters too. For example, I normally dont want to see them with unlimited flight during Levels 1 thru 4. But it sometimes happens, and there are ways to quantify this, in terms of game balance. I prefer limited forms of flight at low tier. But even full flight can be accommodated. Similar concerns arise with the early uses of magic items.

DMs can freeform a new monster however one wants, and a methodology can assess how much a benefit is worth, and approximate the overall player character Level.

You should split your measuring monsters and your designing monsters problems from each other. Having a system to design monsters is one thing, but unless you want to throw out all existing monsters, you will want a measuring monsters system that is more flexible (or flexible to handle existing monsters).

Outliers are important. For monsters and player characters.

At the same time, when there is an outlier, there needs to be a self-evident way to convey how powerful this will be in combat. In other words, a player character level.

Existing monsters are not designed to be measured 1:1 with PCs, but meant to be measured against PC parties. This is why they (especially higher CR ones) sometimes have inflated HP compared to damage output: the HP they have determines the length of the fight, and if they don't have enough HP the fight ends after too few rounds to be interesting.
Meanwhile, lower CR monsters are often envisioned as being more than 1 foe. Using the 5e encounter balancing system this is mainly restricted to CR 3 and under outside of very high level PCs. Note that most 5e games end by level 10; so monsters of CR 4 and above are rarely in packs.

If you are creating a monster Level system meant to be measured 1:1 with PCs, existing monsters won't fit all that well. And that is ok. These monsters need not and should not be forced to emulate PC HP and damage output. They should be designed to make an encounter with an interesting level of challenge when used close to 1:1 with PCs.

This means their damage output should be calibrated to PC defences and HP to PC offence, not the other way around.

As noted, you can usually take a monster and half its HP and double its damage output to get a monster with a "similar" total threat. You do have to be careful that you don't end up with a monster that can one-shot a PC if it wins initiative, and gets one-shot if it loses initiative, as that isn't a super interesting encounter. ... Your chance to hit factors in AC. As designers of monsters, we get to pick if their AC is higher or lower than equivalent PCs. So yes, we have tools to control accuracy.

Your accuracy is your attribute bonus plus your proficiency bonus plus other modifiers. As designers of monsters, we get to pick how they compare.

This trade-off between offense and defense can be true for player characters as well.

The budget for each Level can measure it.

If you assume magic items on the side of the PCs (which most tables have) and not the monsters, we have even more accuracy tools.

Regarding magic items, I prefer a high magic setting with abundant magic, but "attunement" mitigating how many items each character can utilize.

I dont like "magic loot", unless the monsters are first using the magic item against the players. Of course, this makes the combat more deadly for the players.

I feel magic items should come with a Level prereq. Certain items are only attunable at a certain level. A DM can override this (via an "artifact" that any one can use), but if a DM does this, at least the DM has a sense of the power imbalance one is getting oneself into. In terms of Level.

What more, as "a single huge foe" is a fun thing to fight, you'll want math to handle a single monster suitable for 3-4 PCs of level X. In 5e that is a monster of CR X to 1.5X, as monsters are designed (mainly) as solo-against-a-party creatures. With a "pure" monster level system, it is likely to be level 2X-3X or something ridiculous: and that 3X monster will probably have too much single-target damage and not enough HP to make an interesting solo encounter. If you assume a linear progression of stats and proficiency it will also have an insanely high ATK modifier and save DCs.

OTOH if you permit an elite/solo sub-mechanic on top of level you can get a lower level foe which "counts as" more than one monster.

A system of assigning playable character Levels to any monster, requires a mechanic for assessing and building a "Solo", where defacto mechanics for a group of creatures functions as if a single creature. The Solo has multiple times more hit points, extra attacks at multiple targets, and similar, all the while remaining within "bounded accuracy".

This situation seems unusual for a typical player character, but it does happen. A summoning spell allows a character to effectively be two or more creatures, including additional hit points to deplete and multiple attacks. It happens by other means as well.

With regard to the "budget" for a particular Level, an Elite is like a player playing two separate characters. A Solo is like playing three or more characters. A subtler version of this is stats approximating the accompaniment of a summoning spell.

 

[Yaarel]

I am tweaking the Hit Point formula to account for every Challenge:

(Challenge − 1)*13 + 25 Hit Points


The "background" Challenges (0, ⅛, ¼, ½) and Challenge 1 are all increments of about 5 hit points on average, totaling about 25 hit points at Challenge 1.

However, after that, every increment is about 13 hit points.

Challenge 21 ≈ 285 HP
Challenge 17 ≈ 233 HP
Challenge 13 ≈ 181 HP
Challenge 9 ≈ 129 HP
Challenge 5 ≈ 77 HP
Challenge 1 ≈ 25 HP


Here is a close up of the lowest tiers

Challenge 4 ≈ 64 HP
Challenge 3 ≈ 51 HP
Challenge 2 ≈ 38 HP
Challenge 1 ≈ 25 HP

Challenge ½ ≈ 20 HP
Challenge ¼ ≈ 15 HP
Challenge ⅛ ≈ 10 HP
Challenge 0 ≈ 5 HP


There is some flexibility for concept.

 

[dmar]

@NotAYakk

Your criterion sounds about right, a standard combat should average three rounds while depleting a certain proportion of resources.

This is an interesting idea.

Encounter difficulty expressed as depletion of a % of daily resources of an average party.
Would solve one or two encounter per day situaciones, which I find ver often.

Of course this is not everything. You would have to compound it with some measure of encounter threat, not as a result of resource depletion, but of the encounter killing PCs before they can spend said resources.

 

[Yaarel]

I stopped using Experience Points a while ago.

Instead, I count the number of encounters until the next level.

The benefits of counting encounters are many. I drew up this chart to show what I mean. For example, during Level 4, players must complete 12 encounters in order to reach Level 5.

These numbers of encounters for each level derive from the 5e math of experience points until the next level and the experience points of each standard encounter. But it is so much better to simply count the encounters.

I modify Level 1. Intentionally, Level 1 is a lookaround to see what the fantasy world is like, and to get a feel for the player characters. There are three meaningful, nonlethal, encounters that sample the themes of the setting. Level 2 starts the possibility of lethal encounters. I also stretch out the middle tier (Levels 9 thru 12). This tier is significant for us, because it represents the most powerful characters of the Lord of the Rings genre (such as Legolas), and the least powerful characters of the superhero genre (such as versions of Batman).

ALL RESTS are SHORT RESTS. Whether player characters take an hour lunch or an eight hour sleep, the benefit is the same. However twice per Level, a player can declare that their own is "deep rest" that gains the benefit of a Long Rest. Also every new Level starts fresh with all resources at max. Level ups happen between gaming sessions.

The number of encounters might all happen during a single day, or stretch out across weeks. The story determines the passage of time. Any kind of story is possible. There can be a dungeon crawl with combats room after room. There can be an ocean voyage with sporadic encounters during a year. Whatever makes sense for the story. The math stays the same.

A standard encounter is 1 encounter. When counting an encounter, if it turns out to be easy, it is only worth ½ an encounter. If it turns out to be a near TPK it worth 2 encounters. If it is "Hard", somewhere between standard and TPK it is worth 1½ encounters.

It doesnt matter how difficult an encounter is supposed to be. The only thing that matters is how difficult it turns out to be in hindsight. If what is supposed to be "boss" fight turns out to be trivial, then it is only worth ½ an encounter. Oppositely, if a "cakewalk" somehow ends up a near TPK, it is worth 2 encounters. Sometimes players are ingenious, and resolve a challenge in a surprising way. If so, this tends to be worth 1½ encounters even if the idea successfully avoids personal risk.

Counting encounters is accurate and simple. The beauty is, it can also be a noncombat encounter. A social encounter might be Easy, Standard, or Hard, thus count as ½, 1, or 1½ encounters towards leveling up. A "Deadly" social encounter might mean exile or whatever. We like combat, but we also enjoy noncombat encounters. It is possible to play a strictly nonlethal that is meaningful and effective.

 

[dmar]

ALL RESTS are SHORT RESTS. Whether player characters take an hour lunch or an eight hour sleep, the benefit is the same. However twice per Level, a player can declare that their own is "deep rest" that gains the benefit of a Long Rest. Also every new Level starts fresh with all resources at max. Level ups happen between gaming sessions.

This is so interesting...
Shouldn't the number of rests vary with the number of encounters per level?
Also, did you notice any imbalances between classes as a consequence of this?

It is a good idea to decouple resource management from passage of time, for story purposes.
I was toying with the idea of short rest = 8 hrs., long rest = 24 hrs. But I believe yours is better.