Just wanted to get some fact-checking of some concepts I've been considering, going beyond the (pick your flavor of "You'll learn it with time").
This builds on the Tactics-based approach, emphasizing the DM's creativity and strategic use of monsters to create a challenging and enjoyable experience for players.
Adding: Numbers-based approach, focusing on statistics, PC party and monster composition, to create or fine-tuning balanced encounters.
It starts with:
0. Using CR as a starting point: Consider the CR of monsters as a general guideline for difficulty. Use your preferred encounter builder; DND Beyond Encounter Builder, Don Jon, etc
Consider Monster Abilities and Action Economy: Adjust difficulty based on these factors.
Be prepared to adjust on the fly: These are just estimations. The actual difficulty depends on how the encounter unfolds.
It's with a large grain of salt and acknowledging the adage that this will not be able to count for dice roll and shenanigans.
Gather player data:
Calculate Attack damage per round (DPR) and HP for each PC. (Credit to MakingLemonade, who I believe is the calculator author)
Calculate party's Total DPR: Calculate by adding DPR for each PC.
Determine monster damage: Calculate average player HP to dictates the number of attacks the monster can make and the desired encounter difficulty or survivability factor:
Formula:Monster Damage per Attack = Average Player HP / (Attacks per Round * Encounter Difficulty Modifier)
Encounter Difficulty Modifier: This is a value you choose, based on difficulty: Easy: 1.5Medium: 2Hard: 2.5 Deadly: 3
Combat Duration: This result represents the estimated number of rounds the encounter will last based on the party's total damage output and the total HP of all monsters.
Formula:Effective Rounds = Total Monster HP / Party Total DPR * (1 + Average Non-Damage Actions per Monster Round)
Total Monster HP Pool: The sum of the HP of all monsters in the encounter.
Party Total DPR: See #2
Compare this to your target rounds from #4; if it's close, the encounter is likely balanced. A higher number suggests a harder encounter, and a lower number suggests an easier one.
Multiple Monster HP Distribution or Proportionally Adjust HP: This method ensures all monsters contribute roughly equally to your intended difficulty.
Adjusted Monster HP per Monster = Original Monster HP from 0. CR Encounter Builder - (Target Monster HP from #4 - Total Monster Group HP) / Number of Monsters
For example, imagine you have a party with a total DPR of 40 and you want a medium encounter (difficulty modifier 1.25) that lasts 4 rounds. This would give you a monster HP pool of 200 (40 DPR * 4 rounds * 1.25 difficulty). If you have 3 monsters from your encounter builder, each with 100 HP originally, you would end up with:
Adjusted Monster HP per Monster = 100 HP - (200 target HP - 300 total HP) / 3 monsters = 83.3 (83/84) HP
This would make each monster slightly weaker, bringing the total monster HP down to 250, 249.9 rounded up (which is still more than the target HP of 200, but accounts for the fact that the party might not focus fire and could take down some monsters more quickly).
Focus on Diverse Roles: This method is more nuanced and allows you to tailor the encounter based on roles. Here, you would distribute the difference in HP strategically:
Boss Monster: Give the boss monster a larger share of the HP, making it a more significant threat and the focus of the combat.
Minions: Minion monsters can have lower HP, serving as fodder for the party to clear out while they focus on the boss.
Areas for improvement:
Oversimplification: This approach is trying to bottle something that doesn't exactly fit in a bottle. It's been really fun compiling it though.
Party Composition Dependency: This doesn't account exactly for party composition. Strong spellcasters could breeze through encounters designed for a melee-focused group.
This approach can create more interesting and dynamic encounters with more statistical support, by helping build understanding party and monster composition.
Wow. So much is wrong with this. The first major issue is step 1. I don't have any idea where MakingLemonade got the numbers from, but with any method I use to calculate the numbers, they come up significantly lower at mid-level. I suspect that he is assuming a character can take its most potent combination of actions each round, which is simply false.
Once you find or calculate numbers that have some level of accuracy, the next thing I notice is that your method explicitly states a 70% hit rate, but calculates damage as if the damage output were 100%.
The next obvious thing is when you calculate monster damage by dividing by 1.5 for easy, and 3 for deadly. This means that monsters in an easy encounter hit twice as hard as ones in a deadly encounter.
I haven't worked through it step by step, but there is a strong likelihood that other issues exist. DMs wishing to adjust encounters would be well advised to find a different method than this one.
Hey Randy, thanks for point that out, I'll revise it.
That said, you’re critiquing modeling assumptions without identifying which assumptions you’re using instead. Any predictive encounter system requires a baseline , whether that’s average DPR, sustainable DPR, or nova DPR. Without defining your own baseline, saying “the numbers are wrong” isn’t analysis, it’s preference.
The purpose of this framework isn’t perfection. It’s predictability. It gives DMs a way to approximate: • How many rounds an encounter might last • Which side is statistically favored if outcomes trend average • What a ±1 AC adjustment actually does to hit probability • How much a subclass could probably output and getting comfortable asking players "How are you going to play this (class), combat-wise?" or just observing the character sheet for data.
That kind of probabilistic awareness is valuable, especially when running for subclasses you haven’t seen at the table before.
Is the system perfect? No.
Is it more rigorous than relying purely on CR assumptions and intuition? Yes.
If there are flaws, I’m open to refining them, but constructive critique includes demonstrating alternative calculations or proposing improvements. Simply labeling something “obviously wrong” without walking through the math doesn’t meaningfully move the discussion forward.
Looking back at this, almost 2 years later, I wish I learned it alot sooner, I genuinely believe understanding encounter pacing and these types of probability should be baseline knowledge for DMs.
5e is such a power fantasy that I don't agree with RandyllTraeton on this at all. You definitely should be assuming optimal builds with full alpha strike assumption when building every encounter and when calculating encounter strength. If you don't push the envelope on the CR system, there is no challenge in the game, even among inept players who make sub-optimal characters. If they actually make optimal characters and know what they are doing, forget about it. They will crush the CR math.
The only way I have ever managed to make the current structure of the CR math work and actually be challenging is if, during character creation, you do 3d6 down the chain and force players to have highly sub-optimal ability scores and even then, unless it's at least a hard encounter, it takes zero tactical or strategic thinking to overcome combat encounters.
As a whole 5e is not even in the vicinity of a challenging game, combat is effectively a walk-through, especially in the 2024 edition
The basic math for a challenging encounter is that you take the average HP of the group and make sure every monster can one-shot that amount of HP. Then the game gets reasonably scary.
Just wanted to get some fact-checking of some concepts I've been considering, going beyond the (pick your flavor of "You'll learn it with time").
This builds on the Tactics-based approach, emphasizing the DM's creativity and strategic use of monsters to create a challenging and enjoyable experience for players.
Adding: Numbers-based approach, focusing on statistics, PC party and monster composition, to create or fine-tuning balanced encounters.
It starts with:
It's with a large grain of salt and acknowledging the adage that this will not be able to count for dice roll and shenanigans.
Combat Duration: This result represents the estimated number of rounds the encounter will last based on the party's total damage output and the total HP of all monsters.
Compare this to your target rounds from #4; if it's close, the encounter is likely balanced. A higher number suggests a harder encounter, and a lower number suggests an easier one.
Multiple Monster HP Distribution or Proportionally Adjust HP: This method ensures all monsters contribute roughly equally to your intended difficulty.
For example, imagine you have a party with a total DPR of 40 and you want a medium encounter (difficulty modifier 1.25) that lasts 4 rounds. This would give you a monster HP pool of 200 (40 DPR * 4 rounds * 1.25 difficulty). If you have 3 monsters from your encounter builder, each with 100 HP originally, you would end up with:
This would make each monster slightly weaker, bringing the total monster HP down to 250, 249.9 rounded up (which is still more than the target HP of 200, but accounts for the fact that the party might not focus fire and could take down some monsters more quickly).
Focus on Diverse Roles: This method is more nuanced and allows you to tailor the encounter based on roles. Here, you would distribute the difference in HP strategically:Areas for improvement:
This approach can create more interesting and dynamic encounters with more statistical support, by helping build understanding party and monster composition.
Wow. So much is wrong with this. The first major issue is step 1. I don't have any idea where MakingLemonade got the numbers from, but with any method I use to calculate the numbers, they come up significantly lower at mid-level. I suspect that he is assuming a character can take its most potent combination of actions each round, which is simply false.
Once you find or calculate numbers that have some level of accuracy, the next thing I notice is that your method explicitly states a 70% hit rate, but calculates damage as if the damage output were 100%.
The next obvious thing is when you calculate monster damage by dividing by 1.5 for easy, and 3 for deadly. This means that monsters in an easy encounter hit twice as hard as ones in a deadly encounter.
I haven't worked through it step by step, but there is a strong likelihood that other issues exist. DMs wishing to adjust encounters would be well advised to find a different method than this one.
Playing tabletop RPGs since 1978.
Hey Randy, thanks for point that out, I'll revise it.
That said, you’re critiquing modeling assumptions without identifying which assumptions you’re using instead. Any predictive encounter system requires a baseline , whether that’s average DPR, sustainable DPR, or nova DPR. Without defining your own baseline, saying “the numbers are wrong” isn’t analysis, it’s preference.
The purpose of this framework isn’t perfection. It’s predictability. It gives DMs a way to approximate:
• How many rounds an encounter might last
• Which side is statistically favored if outcomes trend average
• What a ±1 AC adjustment actually does to hit probability
• How much a subclass could probably output and getting comfortable asking players "How are you going to play this (class), combat-wise?" or just observing the character sheet for data.
That kind of probabilistic awareness is valuable, especially when running for subclasses you haven’t seen at the table before.
Is the system perfect? No.
Is it more rigorous than relying purely on CR assumptions and intuition? Yes.
If there are flaws, I’m open to refining them, but constructive critique includes demonstrating alternative calculations or proposing improvements. Simply labeling something “obviously wrong” without walking through the math doesn’t meaningfully move the discussion forward.
Looking back at this, almost 2 years later, I wish I learned it alot sooner, I genuinely believe understanding encounter pacing and these types of probability should be baseline knowledge for DMs.
5e is such a power fantasy that I don't agree with RandyllTraeton on this at all. You definitely should be assuming optimal builds with full alpha strike assumption when building every encounter and when calculating encounter strength. If you don't push the envelope on the CR system, there is no challenge in the game, even among inept players who make sub-optimal characters. If they actually make optimal characters and know what they are doing, forget about it. They will crush the CR math.
The only way I have ever managed to make the current structure of the CR math work and actually be challenging is if, during character creation, you do 3d6 down the chain and force players to have highly sub-optimal ability scores and even then, unless it's at least a hard encounter, it takes zero tactical or strategic thinking to overcome combat encounters.
As a whole 5e is not even in the vicinity of a challenging game, combat is effectively a walk-through, especially in the 2024 edition
The basic math for a challenging encounter is that you take the average HP of the group and make sure every monster can one-shot that amount of HP. Then the game gets reasonably scary.