Determining Challenge

[I'm A Banana]

So over in the thread about determining treasure and monsters, I was reading along, and I wanted to tease something out.

See, it's kind of impossible to measure the "difficulty" of monsters in a vacuum. No generic CR or XP value is going to capture the true challenge of a monster. There's too many variables.

For instance, quantity. A lone minion in 4e is probably not worth the XP the party gets for it, regardless of what level they are. A group of dozens and dozens of minions is probably worth a good deal MORE XP than the party gets for it.

There's also class powers. In 3e, a party of rogues fighting a pairing of a golem and his zombie buddies is going to have a much bigger challenge than CR would indicate. Meanwhile, that same enemy group, if the party is fighters and clerics, would wipe the floor with the opposition.

This isn't limited to combat, either. An investigation with a divination spell (or a sky-high Insight or Perception), or a social encounter with a charm effect (or a sky-high Diplomacy skill), or an exploration with characters who can either magically create food and protections, or who can reliably get it with great Endurance, Nature, or whatever. On the other side, an investigation with characters who all dumped Int, or a social encounter with characters who all dumped Cha, or an exploration with characters who all dumped Con. A group of angry wolves with a druid, or a group of angry wolves without a druid. Those are all remarkably different challenges.

4e probably approached the closest, but 4e's careful balance is loaded with assumptions that aren't necessarily palatable for the majority of D&D players. It's a good starting point, but it needs to be able to accommodate what people want out of the game, too.

So, given that it's hardly an exact science to determine individual monster threats, how do you give a DM guidelines? You need to give -- especially newbie DMs -- something to help them navigate the complexities of choosing challenges, and you need to give your designers some broad numbers that your monsters and skill DC's can sit in.

Tell me how you think it should be done. I'm listening.

Me?
[sblock=idea mongering]
I tend to think we might be looking at challenge too closely if we're determining the challenge for each individual monster. Perhaps if we're examining adventure-level design, we need to examine encounter-level challenges. Instead of just having "a wolf = 25 xp", we can have "a wolf encounter", complete with monsters, terrain, and guidelines, and give THAT a total target level. Then, we can hand out XP based on the level of the challenge as a whole.

Maybe? Clearly, I'm early on in this thought process.
[/sblock]

You now!

 

[GreyICE]

Well this is why we have a DM to create encounters.

For instance, there's many ways to make a challenge that significantly outstrips the Experience level. For instance, reflex targeters. Most (non-swordmage) defenders are very weak in the reflex defense. Almost all of them are Fort/Will for their Primary and Secondary stats.

The result is that if you nail most parties with nothing but mobs that target reflex, especially in the low-mid heroic (before they can shore up that defense adequately) you can easily kill the entire party.

Another good way is resist stacking. If you have a Warlock focusing on fire spells (Infernal Spec) and you throw tons of fire resist mobs at the party, you'll find all of a sudden they go pop.

The same with tons of minions. It's just too much to be AOE killed.


The DMG has lots of good guidelines to creating a balanced encounter, and those should be followed. The problem is that any total number you put on an encounter will be very vulnerable to party makeup.

Also you can have, say, 200 monsters in the monster manual for nearly limitless encounters. 200 encounters, at 6-10 encounters per level would mean most level 1-20 parties would face hitting exactly the same encounters, or forcing the DM to improvise which would destroy the entire system. Plus if you were in multiple campaigns you'd quickly learn 'the wolf encounter' as you would have seen it before.

 

[KidSnide]

There are two different issues here. One is how the DM can measure challenge so as to correctly predict how difficult an encounter will be. The other is to determine what rewards to give the PCs for completing the encounter.

For the first, the game needs to start with the assumption of a balanced party, where the quantity of magical items are used to adjust the party's effective level. The level (or CR) of the monster is based on the threat it poses to an ordinary party. Then, to the extent it's not obvious, the rules should give guidelines on how to adjust the threat level based on unusual circumstances. For example, monsters that have DR/magic pose more of a threat to martial-heavy parties that aren't equipped with magical weapons, regardless of level.

For the second, I suppose you could go through the arduous task of adjusting xp based on difficulty, but I think the issues you raise are yet another reason why PCs should be rewarded based on what they accomplish (whether treasure, story goals or whatever) rather than the difficulty they go through in order to accomplish it.

-KS

 

[Crazy Jerome]

As I said in the other topic, communicate the boundaries, not the average. You can include average, mean, median, etc. if they are useful, especially if not the exact mid point of the boundaries every time. But they aren't nearly as important as the boundaries.

It's much easier to explain a stock adult white dragon to a novice DM if you say something such as: Below level T, the party is most likely toast, if you play the dragon at all reasonable. Below level L, the party will probably lose a few members, unless they run immedidately or are lucky. Above level S, the dragon is a speed bump. Between levels L and S, increased numbers, luck, good tactics, nice DM, etc. will be less and less necessary as the party approaches S.

Naturally, you'd want some kind of one-line code block to reduce such boundary assumptions to a manageable bit.

 

[Boarstorm]

So... level ranges? That could work. Perhaps to shoehorn the concept into the "one-line code block" we could have a return to challenge codes.

Challenge Code A: Suitable for an equal number of level 1-3 PCs. Dangerous to higher-level PCs in groups of 1/2Level*5 per PC.
Challenge Code S: Suitable for a party of 4-6 PCs of level 20. Certain death to lower level PCs.

... as a "for instance."

 

[I'm A Banana]

Well this is why we have a DM to create encounters.

Well, how does a DM (especially a new one) learn to do that? What guidance (aside from general advice) should the game give? Or do you think general advice should be enough for anyone?

For the first, the game needs to start with the assumption of a balanced party, where the quantity of magical items are used to adjust the party's effective level.

A "balanced party" is already a pretty big assumption. Shouldn't my players all be able to play rogues if they want to without making my job as DM significantly harder? We can approach this by making broadly capable characters, but that cuts into D&D's class system (part of which is being not-as-capable in certain circumstances).

Magic items affecting level isn't a bad solution, but an item that deals, say, extra radiant damage, might be worth different amounts in campaigns that are about fighting undead than in campaigns that are about fighting angels.

So you need to assume a "balanced world," too. Which puts some limits on what a DM can easily change.

For the second, I suppose you could go through the arduous task of adjusting xp based on difficulty, but I think the issues you raise are yet another reason why PCs should be rewarded based on what they accomplish (whether treasure, story goals or whatever) rather than the difficulty they go through in order to accomplish it.

I'm fully on board with this idea. Though it does rule out using a 4e-style "XP Budget" to design anything. There's a lot of people who really like the elegance of that system, and with good reason. Though that seems largely about the first issue, not about the second.

As I said in the other topic, communicate the boundaries, not the average. You can include average, mean, median, etc. if they are useful, especially if not the exact mid point of the boundaries every time. But they aren't nearly as important as the boundaries.

Hmm...it's not a bad idea. Sort of like how old adventures had a "level range," we can now give monsters/encounters a level range. It helps get rid of the assumption that monster level is a precise judgement, since it's not a precise figure. There's still a slight issue with character powers, but it's mitigated -- even if the ranger with his favored enemy is about 2 levels more effective than usual, you just sprinkle a few threats in that are about EL+2, and call it a day.

Especially if combined with KidSnide's idea of XP as a simple reward system (rather than as a difficulty system), this ensures that you don't get extra rewards from fighting bigger things, necessarily.

Interesting ideas! Keep 'em coming!

 

[Janaxstrus]

A "balanced party" is already a pretty big assumption. Shouldn't my players all be able to play rogues if they want to without making my job as DM significantly harder? We can approach this by making broadly capable characters, but that cuts into D&D's class system (part of which is being not-as-capable in certain circumstances).

They have to have a baseline for everything. As such, "balanced party of 4-5 PCs" is a good place to start. You can't write the DMG and have 12 paragraphs for each subject, 1 for each possible makeup of the party. It just won't work.

I would much rather work from the assumption of a balanced party, than change the game so the classes are more broadly capable.

 

[LostSoul]

So, given that it's hardly an exact science to determine individual monster threats, how do you give a DM guidelines? You need to give -- especially newbie DMs -- something to help them navigate the complexities of choosing challenges, and you need to give your designers some broad numbers that your monsters and skill DC's can sit in.

Tell me how you think it should be done. I'm listening.

An entry in the Monster Manual statblock that reads "Rumours/Information", gives a few DCs and the information related to that - weaknesses, special abilities, special attacks, some sort of comparison to normal men - "A troll is equal to 20 men-at-arms, or 5 if they are equipped with fire" (i.e. monster level).

Then let the players decide what they want to do with that information.

 

[Crazy Jerome]

Especially if combined with KidSnide's idea of XP as a simple reward system (rather than as a difficulty system), this ensures that you don't get extra rewards from fighting bigger things, necessarily.

Since what people want to reward varies so much, divorce the baseline for XP from the creatures entirely. CR or similar scales are for telling the DM how much or how little trouble he is putting the party up against. If that's all that scale has to do, it might do it well.

Then for the default XP rewards, I'd base XP on party level, with about a page or so of guidelines on how to modify this for particular styles. If a character is 8th level, goes on an adventure of moderate complexity and length (say, expected to finish in about three 4 hours sessions), then they get NNNN XP. Or even better, if they really make that "one hour adventure segment" thing work out, make it per hour. Do the XP advancement chart to match, you can even make this easy to remember--say, about 100 times character level per hour "adventure segment".

In the guidelines, explain that if your players are getting stuff done but running off on tangents all the time, it might be better to go with XP per hour played. Or if they like to cut up a lot and take longer, bump up the reward for completion of a segment, on the grounds that everyone is having fun at the slowr pace. Or don't, if you like slower advancement to go with your slower pace. Or if they like to cut up a lot and drive half the table crazy in the process, bump it down or be more strict about finishing stuff. Or if your fun is in meeting the challenges, only factor in adjustments for things that turned out too hard or too easy. The DM making those kinds of adjustments is mostly art instead of math, anyway. So get a base number with as minimal a math as possible, and advise how to tweak that number appropriately.

I'm convinced that most experienced DMs end up doing something like the above adjustments anyway. It's just harder to learn and harder to do if you are applying lots of fiddly formulas and evaluating every little bit separately. It's much easier for an up and coming DM to get to the end of the session, eyeball it, and say, "You know, you guys were laying into it, but that was tougher than normal. Have a +20% bonus."

 

[GM Dave]

1> With the campaign that I started in Jan (played 9 sessions) I dropped XP completely. I put the advancement in the hands of the players. If they want a level then I let them have a vote. IF one third vote No then they stay at their current level and if 2/3 vote to advance then they go up a level. I leave it up to the players to advance as fast or as slow as they want. If they find a sweet spot they are enjoying playing at then they might not advance at all.

2> If you do go in for some sort of XP calculation then the idea of a 'balanced' party is going to lead you into trouble. I know that I game with 9 people and some days I have all 9 and some days I have four or five. My encounters need to adjust on the fly since I never know till I get to the game night (and sometimes an hour into the night) how many players I have for a particular encounter (even then players are like cats and run off on other things meaning not the whole group is always present for a battle).

3> I gauge toughness of an encounter in two ways.

The first way is how much resources the players expended to win the encounter. This is often related to the amount of damage or long term effects that the monsters did to the players. (This is the danger value)

For example if an encounter lasted five rounds an all the monsters did was put a single player to sleep each round then the encounter cost the players very little in resources (no healing, no money, no limited supplies ~ dispels, remove curses, cure poisons, remove diseases).

For another example if an encounter ends with each player contracting Mummies Rot then this a tough encounter as the players will need to spend considerable resources getting each person cleared of the curse (which could take days or weeks).

The second way is how long it takes the players to overcome the obstacle. (This is the slog value).

A battle or obstacle that is cleared in a single round is an easier challenge because their are many opportunities to access resources outside of the encounter which are more difficult to do inside of an encounter (getting the remove curse scroll from the backpack which is likely buried towards the bottom or the rarely needed folding boat). A battle or obstacle that lasts five or ten rounds encourages players to dig through their supplies of easily replaceable things into their more limited resources (in PF a level 3 Wizard has only so many spells per day and if they cast a spell on every round of a combat then they will quickly go dry on a long slog. The same goes for the archer who is firing two or more arrows around unless the DM gives them an infinite quiver of arrows). Small amounts of damage if occuring every round will quickly add up in a slog fight (fighting inside a forest fire or house fire will wear down even paragon fighters).

4> XP considerations need to consider the comparable value and learning of actions without combat. That is unless you just want to play dungeon hack.