D&D 5E Encounter difficulty: how to fix it.

[Tormyr]

I've been comparing your system with the RAW system for the purposes of encounter design and noticed something along these lines. When it comes to large groups of same CR monsters, your system seems to underestimate the encounter challenge by one degree as compared to RAW.

For example, take a party of 4 characters of 8th level.

In your system, the PEL thresholds would be EASY (40%) = 12+ / MEDIUM (60%) = 19+ / HARD (80%) = 25+ / DEADLY (100%) = 32+

Whereas in the RAW system, the XP thresholds would be EASY =1,800+ / MEDIUM = 3,600+ / HARD = 5,600+ / DEADLY = 8,400+

First, they face a pack of 8 dire wolves. In your system this would be just an EASY encounter (PEL 16), whereas in the RAW system this would be a MEDIUM encounter (8 * 200 * 2.5 = 4,000 XP). Probably not a big deal, since the key difference between an EASY and MEDIUM encounter is the expenditure of healing resources.

Next, they face a pack of 14 dire wolves. In your system this would be a HARD encounter (PEL 28), whereas in the RAW system this would be a DEADLY encounter (14 * 200 * 3 = 8,400). And if it were 15 dire wolves it would clearly be a DEADLY encounter thanks to the monster multiplier. This is a more significant discrepancy because the difference between a HARD and a DEADLY encounter could be the difference between few / no PC deaths and multiple deaths / a TPK.

Which is the more accurate guideline? I don't know. I suspect for a well-equipped fully rested party of 4+ PCs with powerful area spells that your system is a bit more accurate. However, if the party is under-equipped, not fully rested, is of small size (thus with diminished force multiplier capacity), or lacks a caster with area spells, then I suspect the RAW system is a bit more accurate.

With the encounter difficulty changing in the new version of the Basic DMG, I am not sure the 40%, 60%, 80%, and 100% thresholds are still valid. There is also a differentiation between scaling the adjusted XP in both systems. So where you calculate the two systems matters, but you already pointed that out.

I use a spreadsheet, so I don't really care if the difficulty is calculated using PEL or RAW. The greatest strength of PEL in my opinion is that it provides a gradual, constant scaling of encounter difficulty with each new creature added. RAW has some nasty jumps in difficulty. Dire Wolves 3-6 add 400 XP each, but Dire Wolf 7 adds 1100 XP. A viable solution may be to introduce PEL's scaling into the normal XP budget system. But doing that on the fly would require a spreadsheet which is not something everyone is going to have when they are putting together an encounter. I have a feeling the PEL tables might still work, but they may need to be rebalanced.

 

[Tormyr]

Better but still a few problems. You need a "don't be a jerk" rule because 12 flying snakes (easy to medium) are going to annihilate a 2nd level party.

Since this lets you have multiple monsters more easily it lets a DM, intentionally or accidentally, mass a large number of under CRed creatures (and boy there are a lot of them in the MM).

Also without the multiplier I think you'll find larger parties, like 6, are going to be facing significantly tougher fights under your system.

You forgot to use the encounter difficulty multiplier for 12 creatures. Both systems rate that as a deadly encounter, not an easy to medium one.

The PEL accounts for adjustments to both the number of characters and the number of monsters. The math is buried in there, but it is there.

 

[Pawnmower]

Just saw this post. Has anyone been using this system since the DMG has come out? Are difficulty % still valid?

 

[doodlidoo]

PEL vs XP Budget spreadsheet

Hello all,

Reactivating this thread, as I think a lot of DMs new to 5th edition (I am one of those) will happen to ask themselves if the XP Budget method found in the DMG is accurate for multi CR encounters.

Here's my 2 coppers : a spreadsheet that allows you to compare the resulting thresholds of an encounter given by both methods (PEL vs XP Budget). As I was a little doubtful about the numbers given by the OP for the breaking points (40%, 60%, 80%, 100%), I added an option that lets you select the average D&D experience of your players, changing the results accordingly. In that way, for example, the same encounter could be tagged "easy" for high experienced players, and "hard" for novices.

Feel free to send any question, comment or suggestion.

 

[Tony Vargas]

That's a good effort, but even with the OP's method saving the multiplier step (which is really nice), it still seems more cumbersome than 3e's CR and less dependable than 4e's EL.

Back in the old AD&D days, where monsters had a level scaling from I to X (in roman numbering, mind you!), there were no encounter building rules. You had to follow your gut feeling. Use intuition. This will of course yield some easy fights and some tough fights. And some of those will be unexpectedly easy or difficult.

Best solution then, and, IMHO, best solution, now. I ran the fights in HotDQ straight, they were disasters. I re-created some using the guidelines - not much better.
I started just running encounters the way I did in the old days, by feel, adjusting on the fly: much better results. Of course, it's a very personal approach, YMMV in a big way.

 

[Pickles III]

I'm sure someone has mentioned this already, if you were using the OP system, why not just multiply the XP as given by the difficulty rating as given by the OP.
For instance 40% of the total monster XPs for easy encounters, 60% of the total monster XPs for medium encounters, 80% of the total monster XPs for hard encounters and 100% of the total monster XP for deadly. I'm actually going to try out the system tonight and offer some feedback of its success/failures.

This exercise highlights the whole mess of the encounter building system which is predicate on what I consider s design error in the game (not one that effects game play directly so it's not a biggy).

There are two factors why having more monsters makes for a harder fight. One is "Lanchester" - fighting 5 monsters then 5 monsters is easier than fighting 10 monsters all at once. The other is that the XP value of lower CR monsters is disproportionately high. Wizards managed to conflate these two issues & come up with their overly cumbersome system. Ie XP value of monsters rises faster than their threat.

The true XP values of the monsters should be proportionate to their PEL, so if a CR 1 monster is the baseline worth 200 xp then a CR 1/4 should be 80 not 50 & CR 3 should be 320 not 450. If these values were used you could just add up the XP budget & get the right results.
Somehow, given that this is what 4e did correctly & well, WotC managed to completely screw this up & got the complex mess now cleaned up by #Gobelure.

The next step would be to disassociate PEL from CR so we can have elites solos & especially minions back *

The DM encounter building guidelines could then address how swarms of monsters are rather vulnerable to AOE at higher level & how higher than party CR monsters are especially dangerous.

Whether you should get more XP for fighting tougher fights or the same XP for seeking out & ganking individual orcs has always been an issue with D&D. It is not an especially big one but the wonky maths of the encounter building guidelines drew attention to it (what they were really highlighting is the bad XP values for low CR critters.

As as side effect of this maths conjure animals & its ilk are pretty broken** - you get 2 CRs worth of critters adding up the CR literally. Clearly you get twice the bang for your buck with 8 PEL of CR 1/4 wolves than with 4 PEL of polar bear.Not helped by the fact that wolves are very good CR 1/4. Also not a problem for their use as monsters - bad when the PCs will cherry pick the best summons they can.

I'm not yet to the point where I may advocate how to attribute XP to players according to encounter difficulty. I leave this for the DMG as I will probably use the milestone system for myself.

Equivalent Monster CR	Total PEL	Encounter XP
0	1/3	10
1/8	2/3	25
1/4	1	50
1/2	1.5	100
1	2.5	200
2	4	450
3	6	700
4	8	1,100
5	11	1,800
6	13	2,300
7	15	2,900
8	18	3,900
9	21	5,000
10	24	5,900
11	28	7,200
12	32	8,400
13	36	10,000
14	40	11,500
15	44	13,000
16	48	15,000
17	52	18,000
18	56	20,000
19	60	22,000
20	64	25,000
21	76	33,000
22	88	41,000
23	104	50,000
24	120	62,000
25	136	75,000
26	152	90,000
27	168	105,000
28	184	120,000
29	200	135,000
30	216	155,000

Thanks for the table referenced above! I use milestones too.

* Jk bounded accuracy largely fixes this. Monsters with lots of attacks they cannot focus on individuals work as solos (lair actions?) & CR<1 work as minions.

** I was going to say "broken but not game breaking" but since it was a major factor in me stopping DMing the game I ran maybe not. YMMV in fact almost certainly will.

 

[Pickles III]

That's a good effort, but even with the OP's method saving the multiplier step (which is really nice), it still seems more cumbersome than 3e's CR and less dependable than 4e's EL.

It's less cumbersome than CR which added logarithmically. The fact it's worse than the 4e EL is due to the vagaries of the CRs themselves.

Granularity is an issue especially at low value. I am fine with a bit of eyeballing & fudging - it's a guideline explicitly afterall.

 

[Tony Vargas]

It's less cumbersome than CR which added logarithmically.

What makes this system seem cumbersome to me is that nothing quite equates to actual level. CR equated to level and EL was level. Doubling is a very easy operation, so figuring out how many ogres theoretically present a challenge to a high-level party isn't that cumbersome, since at least you're starting with something like level on both sides of the calculation. You know you want an EL 10 to modestly challenge a level 10 party. One CR 10 equals 2 CR 8, 4 CR 6, 8 CR 4, and, boom 16 CR 2 ogres. Probably won't actually turn out to be much of a challenge (between UnBounded Accuracy and Tier 1 casters), but the math wasn't hard or unintuitive.

The fact it's worse than the 4e EL is due to the vagaries of the CRs themselves.

Granularity is an issue especially at low value. I am fine with a bit of eyeballing & fudging - it's a guideline explicitly afterall.

It's also just the looser 'balance' of 5e, in general. You can't count on a party (not) having certain general capabilities let alone synergies or even ballpark overall effectiveness depending on party composition, as well as the monster's CR's not being exactly dependable, as well as Bounded Accuracy making being outnumbered such an issue. It's always been a guideline, just a more helpful one and/or easier to use one in other modern eds - but, yes, guidelines need always be used with caution.

5e's saving grace is that the DM has more than enough latitude to fix any problems as they occur, via said eyeballing and fudging.

 

[Athinar]

Search for "Excel spreadsheet that does the calculations CR

Search for "Excel spreadsheet that does the calculations CR", Created it back in January for both "Level Threat" and "CR Calc"

 

[Gobelure]

FYI, the numbers haven't changed between the preview and the DMG. So the system is still valid !

Just a word about the thresholds : they work fine in practice for lower levels. At higher level, thresholds get shifted a bit and become 35% for easy, 55% for medium and 75% for hard.

Not such a big shift anyway, I seriously doubt you can predict the outcome of a fight at 5% level

IMHO, feeling should always prevail on the math when designing encounters : the encounter should look like something you want. Math can still help not getting too far outbounds.