Hypothetical party:

1st level human wizard (PC)
3x guard NPC
1x acolyte NPC

This is the player's first time, and it's a solo campaign. The first mini-adventure involves recovering the cargo of a supply caravan that didn't make it to her magic academy due to bandit hijacking (as a sort of graduate-test). The guards and healer are on contract from the school to assist her. The acolyte is essentially just to hang back and tend to wounds and probably won't engage in combat unless someone goes down or the PC is seriously wounded.

So, with that "party" makeup, how do I determine the XP budget for a "medium" or "hard" encounter since there's no clear CR->level mapping in 5e for those NPCs (who will certainly factor into any combat as meatshields and melee damage)? I was thinking about reverse-engineering them as crumby lvl2 fighters (given the 2 HD worth of HP they have), but they certainly aren't on par with a lvl2 fighter, given the lack of action surge, second wind, etc., so the encounters would end up being too hard compared to PCs or DMPCs.

Could they be lvl1-equivalents, essentially trading any abilities a PC would have for extra HP? How would this work at later levels with higher CR NPCs (veteran, priest, mage, etc.)? I'd rather not have to build a bunch of DMPCs as hirelings, because I want the NPCs to be a bit vanilla compared to the versatility of my player. I'm already introducing 1-2 DMPCs to the game for her to take control over as she gets a feel for the flow of the game (for now, she'll handle them out of combat, and I'll effectively show her how to utilize them until she's ready).

Just to be clear, I'm comfortable with the monster-generation rules for crafting NPCs as enemies to determine their CR, but I'm confused on how to use them in the adventuring party when budgeting encounter difficulty XP.


6 Answers 6


Here's what I've been trying, and it seems to work out ok, although I am only a few sessions in:

First, balance the encounter considering only the player characters. Then for each allied monster, add a same-CR monster to the other side. So, if you have a CR2 monster as an ally, the opposing side gets a CR2 monster added in.

Otherwise, it's very hard to do the math because the CR of a monster doesn't translate very well to its "level" as a character.

  • \$\begingroup\$ Does this work if the other side gets 2 CR 1 creatures instead? Or 4 CR 1/2? \$\endgroup\$
    – daze413
    Commented Oct 5, 2015 at 11:01
  • \$\begingroup\$ @daze413 In general no, because the number of creatures impacts the action economy. If you want to change the number of creatures, you might try the encounter multipliers (considering the "extra creatures" separate from the party and the original creatures you balanced out for them), but I haven't tried that yet so I can't say how well it would work. \$\endgroup\$ Commented Oct 5, 2015 at 15:58
  • 2
    \$\begingroup\$ I've tried using the "adjusted XP values" for an uneven number of creatures added on both sides, but nothing really wild ( no 8 CR 1/4 against 1 CR4, for example) it works out pretty well! \$\endgroup\$
    – daze413
    Commented Jul 12, 2017 at 2:12

The easiest thing here is to count the hit dice and subtract 1 to determine the NPC's rough character level. Though that is far from perfect.

The reason you subtract an HD is that for regular PCs, they get double their first HD to start (it's max value), thus a 2HD monster actually has L1 equivalent HP. (This is not official, just a quick and dirty check to get a starting idea of level).

Acolyte's are L1 using this math. This is supported by the fact that they have 2 HD and are listed as L1 spell casters. Guards are L1 by this logic as well.

Both of these NPCs have +2 proficiency bonuses. The only marked difference from L1 characters is that their stats are significantly lower than a typical L1 PC's would be (which is actually pretty significant). This makes them likely L0 based on starting stats, which is pretty meaningless in terms of setting encounter budgets.

I would go ahead and plan for fights to be harder than if you had 5 L1 PCs, as the significantly lower stats of these NPCs is probably going to be a significant barrier to their effectiveness. Perhaps count each NPC as half a L1 PC and see how that works?

Instead, you should probably consider designing one or two (if not all, though that's probably extreme) of these NPCs using PC rules instead and only count those 3 (or using 4 L1 characters for your calculations) when determining your XP budgets. Factoring them based on 5 L1 PCs with this party is going to end in TPKs.

The takeaway here would be that if you care enough about an NPC to be concerned about their equivalent character level, you probably want to go ahead and make them using PC rules. It's a bit more work than pulling a pregenerated NPC out, but it will be worth it.

  • \$\begingroup\$ I see. So, would you generally always use DM-PCs for everything from escorts to hirelings to make the encounter budget math easier? \$\endgroup\$ Commented Jan 6, 2015 at 21:40
  • 1
    \$\begingroup\$ @Andy_Vulhop not necessarily, there is certainly a balance that you'll have to develop through trial and error, but I think you'd find that 1 NPC as PC is worth several of these NPCs at similar character levels and might actually end up reducing complexity. If these are one off escort mission NPCs then just treat them as level 1/2 characters and move on, if they will stick around a few sessions put a few more minutes into them and build them as PCs \$\endgroup\$
    – wax eagle
    Commented Jan 6, 2015 at 21:42

At the moment you are trying to find equivalent PC levels of NPCs and use this number to find the proper XP budget. Maybe you should try the opposite, more direct way: take the NPCs and calculate their XP budget as if they were a group of enemies, and use this number to create an encounter. Such an encounter should have an approximately 50% chance of winning for NPCs (and should be probably considered as "hard").

Now, add the wizard and play out the battle (taking averages for all the damage rolls to speed things up), you will immediately see whether you need to tune the difficulty and which way. In fact "generate an encounter, play it out, and make conclusions" is the only reliable answer to this question since XP budget is an estimation that doesn't take into account many things. It is time-consuming though.

While this is not an answer for the stated question, it is a solution for the problem as a whole.

  • 6
    \$\begingroup\$ Solving the problem is always the aim. Sometimes that means answering a different, more useful question than was literally asked. \$\endgroup\$ Commented Mar 19, 2015 at 21:23

To be perfectly honest, I think using the "monster" type NPC stats in a party is probably not supported by the RAW in general. If these were temporary allies who were going to hang around for a session or two, you could easily fudge it, but if they're meant to replace PCs because of a player shortage, you're going to be far better off creating the "stand-in" NPCs as if they were PCs.

The two biggest components to a "monster's" challenge are its hit points and damage output per round, and to keep that up, "monster" stat blocks get things like multiattack and gang-up bonuses at pretty low level (like the Thug, with its challenge of 1/2). Stuff like that is what's going to make it particularly hard to approximate, because PC levels are made up of things like the breadth and depth of their abilities and their capacity to handle noncombat tasks on top of combat, and that's what your player is really going to be missing out on.

There may be a lot of ways you could simulate, run some kind of statistical regression or whatever to estimate what level equivalent those NPC stats should be, but none of that is going to be less work than rolling up four 1st-level characters to be the co-stars in your one player's story.

  • \$\begingroup\$ side note: in my experience, spell caster NPCs don't suffer from this issue as much, as they often get utility spells as well, and PC casters (especially wizards) don't get that much besides badass spells. There are of course a few benefits granted to PC casters besides spells, but some of them are sometimes replicated in the NPCs statblock under a different name, and even without that, the difference shouldn't be too significant. In fact, giving your "DMPCs" too many abilities beyond a maybe long, but at least overseeable, sorted spell list is just going to make your life harder as the DM. \$\endgroup\$ Commented May 12, 2020 at 23:45

Wax Eagle's post provides an excellent way to calculate the outcome theoretically. But you can also calculate it empirically, by simulating the combat.

Open up excel, and start a row for each character in the fight. PCs, NPC helpers, and monsters/opponents. Then use each column to show the outcome of that round.

Decide who would attack whom, and subtract the appropriate damage from their HP. Once someone is down, stop subtracting their damage from their opponents HP.

In the end, whoever is standing will win statistically.

However, if you don't want to have to fudge dice later, or risk TPKing your new player (should the enemy crit or just tend to roll high on damage), make sure his team has 1/4-1/2 of their max hp left. This will help account for the swing of the damage. If you would rather, multiply the friendly damage by .75 and the enemy damage by 1.25 to make sure that you won't TPK your party. Then as long as your PC's party has someone standing at the end, they should survive.

The formula for average damage for a rolled attack is as follows (\$N_\text{Damage Dice}\$ is the number of damage dice):

$$\text{Average Damage} = \frac{20 - \text{Target AC} + \text{Attack Bonus}}{20} \times \left( N_\text{Damage Dice} \times \frac{\text{Die size} + 1}{2} \right)$$

          Note: this does not take into account crits.

And for ST based attacks:

$$\text{Average_Damage} = \left(1 - \frac{\text{DC} - \text{Save Bonus}}{20}\right) \times \left( N_\text{Damage Dice} \times \frac{\text{Die size} + 1}{2} \right)$$


In 3.5 and other version of DnD, there were cohort XP rules. Given that I haven't found such rules for D&D 5th ed, I think it's time to abstract a tiny bit.

Rather then try and calculate the ECL of all of these NPC individually, consider them as a fighting force. You have three guards, offering three attacks per round, and offering tactical options for the PC if he commands them properly. Additionally, you have an acolyte who is capable of in and out of combat healing, and knows not to get into melee. If you consider two of the guards to be the combined equivalent of a first level fighter and the guard and the acolyte to be the combined equivalent of a first level cleric, then what you're left with is a party of 3 level 1 PCs, or it's equivalent.

You may also consider leveling them gradually, and less often then the PC, if the game last long enough. Either with their first level up being to PC classes as opposed to their NPC build, or by gradually incrementing their stats and ability between one NPC write up to the next, so that they doing have sudden jumps in competency more dramatic then one PC level.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .