D&D 5E Fixing Challenge Rating

[EzekielRaiden]

At the very least, I respect the courage to not only admit that a previous effort fell short, but to try to make it better.

 

[Quickleaf]

Just wanted to say- this is brilliant, and is extremely good advice. This should be in the DMG!

Cheers James! I appreciate the kind words. It's just what has worked for me with my eccentric intuitive approach. Hope it's helpful to you.

 

[bedir than]

What if the real secret is that encounters don't need to be balanced

 

[Umbran]

What if the real secret is that encounters don't need to be balanced

What if the real secret is that some folks want a solid gauge of how tough an encounter is before they put it in front of players, and if you don't, maybe this discussion isn't really relevant to you?

 

[Clint_L]

Ooh, I'm interested to see where this ends up going in the future.

I'm curious, by any chance, have you watched this video, and if so, what are your thoughts on it? I made it last year, and it was very positively received, and it's also about trying new approaches to this little math problem.

A) That video is really, really well done. I recommend.

B) I use the encounter builder on DnDBeyond, but as far as I can tell it just generates CR according to the basic method or, as you aptly describe it, "hopes and vibes." More or less, anyway. Since it's a digital resource and we don't have to do the math ourselves, I hope they adopt something like your much more accurate, simulation-based method.

Right now, CR is fine in the sense that it gives me a very basic idea of an encounter budget, and I have decades of experience to contribute, so that occasionally my encounters work out as intended. Vaguely.

Your method, and what Mearls is working on, will be invaluable, but again, especially if incorporated into DDB and the encounter builder.

 

[J-H]

Your method, and what Mearls is working on, will be invaluable, but again, especially if incorporated into DDB and the encounter builder.

This comment caused another thought to percolate up: Not everyone uses D&D Beyond or online encounter builders. Whatever replaces CR needs to be something that can be used without a computer/phone or calculator to be truly useful.

The current way to use CR, per the DMG, is:
1. Look up your player's levels and XP budget per level
2. Add up their total XP budget.
3. Check the CRs of the monster you want to use.
4. Convert the CRs into a total XP value.
5. Add multipliers if there are a lot of monsters.
6. Check the two XP values against each other.
7. Decide.
There may be a way to compare CR vs party's average level directly, but if so, it's not in the DMG explanation and I don't know that it will hold up. For example, an Ogre is CR 2. It's a reasonable challenge for a 2nd level party, and can knock 1-2 of them to 0 hp. A Vampire is CR 13. It's a speedbump for a 13th level party.

I find myself unsure that reducing MM/NPC entries in all their varieties down to a single reductive yet also meaningful number is going to be achievable. That said, a Battle Value number (to steal Battletech terms) does seem to be the way to go versus something tied to "levels." Battletech covers a variety of ranges, damage types, movement types, and protection/durability levels, with BV, and it seems to be generally accepted as reasonably accurate.

Rating player characters by BV may cause some heartburn, as there will be differences!

I'm interested in seeing where this goes.

 

[tomedunn]

A) That video is really, really well done. I recommend.

B) I use the encounter builder on DnDBeyond, but as far as I can tell it just generates CR according to the basic method or, as you aptly describe it, "hopes and vibes." More or less, anyway. Since it's a digital resource and we don't have to do the math ourselves, I hope they adopt something like your much more accurate, simulation-based method.

Right now, CR is fine in the sense that it gives me a very basic idea of an encounter budget, and I have decades of experience to contribute, so that occasionally my encounters work out as intended. Vaguely.

Your method, and what Mearls is working on, will be invaluable, but again, especially if incorporated into DDB and the encounter builder.

One of the major problems that online encounter calculators run into is that they assume every monsters contributes to the encounter's XP multiplier. If you are only ever using monsters of similar CR then this shouldn't pose much of a problem, but if you try to build an encounter with say a CR 9 monster and two CR 3 monsters then it's going to massively overestimate the adjusted XP total. Until online calculators add in some method for intelligently picking which monsters contribute to the XP multiplier and which don't this will continue to be a problem.

A while back I put together a encounter calculator spreadsheet that tried to correct for this by giving a few alternative calculations based on different assumptions about the party's AoE capabilities and what tactics they might use. Instead of giving a single adjusted XP value, it gives you one for each of those scenarios to help form a more complete picture of how the encounter might play out.

Also, while the encounter XP multiplier might feel like "hopes and vibes", there is some solid math underlying it. I wrote a series of articles that go into detail on how it all works and the math behind it (part 1, part 2, and part 3) .

 

[Clint_L]

One of the major problems that online encounter calculators run into is that they assume every monsters contributes to the encounter's XP multiplier. If you are only ever using monsters of similar CR then this shouldn't pose much of a problem, but if you try to build an encounter with say a CR 9 monster and two CR 3 monsters then it's going to massively overestimate the adjusted XP total. Until online calculators add in some method for intelligently picking which monsters contribute to the XP multiplier and which don't this will continue to be a problem.

Agreed! That's why I think the simulation-based method that Trekiros has built is so good.

 

[mearls]

Definitely easier to use than the current system and I think it's a good start. However, I am definitely skeptical of the numbers at the upper CR range. CR + 5 = deadly encounter for a party of 6? We just know that this isn't the case. Perhaps the number of points per player could be adjusted a bit per level. a single CR 6 monster might be deadly for 6 people at level 1, but it's a near certaintly that a single CR 16 is not a deadly encounter for a level 11 party.

Yeah, I'm pretty sure this breaks as we go up in level. However, hopefully the act of breaking it reveals a pattern of where monsters should be at high levels. Right now we don't have a great benchmark IMO.

 

[mearls]

Are you looking for any assistance with any aspect of this? I'm a software engineer with a math degree so working on a software tool to help fix one of my biggest frustrations as both a DM and a math nerd would be a ton of fun!

I will soon! Doing this in part to re-activate my programming skills, but as I post stuff I definitely hope people review the work and find ways to improve it. A point scheme like this is a useful community resource. Having it locked up in one person or company's hands wouldn't help.