The Limit of Difficulty

[Choas]

Here is an extremely DM-centric question. What is the absolute limit of difficulty? If it could be a law of science, how could it be defined? Mathematically I would guess that the limit of difficulty would be a situation that takes 99.99% of the party's resources along with the players doing their absolute best and thinking their absolute best in order to overcome.

The point of this question is this: How do we reach the point that we make the players feel satisfied because the difficulty of an encounter claimed nearly everything the party had to solve it. This does not mean that the encounter has to destroy all the party resources, merely that it takes a lot of effort and all of each character's ability to resolve. Can we come up with some situations that make this situation more tangible ( with numbers).

The goal is to push yourself as a DM but still make the players feel like they had a load of fun while nearly being annihilated.

[Azireal]

Well the first problem that arises is that a situation that claims 99.99% of the party's possible strength would require that the dice not be a factor, simply assuming all 10's or 20's wouldn't suffice.

Beyond that, many players aren't even capable of accessing all 100% of a character's true strength, it could even be argued that it is impossible to attain 100%.

Beyond that even further, even if the players came across that situation it is debatable if they would even be able to beat it, as it would require absolutly no mistakes or holding back.

Even further than further, would you be happy if you were pushed to your absolute limit when you played an RPG? IMO it would just make the players think that they are somehow not supposed to be doing whatever it is that they are, but perhaps that's just me.

[Niko_Kaze]

My general opinion is that this question is one that has laid in the front of role playing games since the concept was thought up.

I generally do not think that any one encounter should be that tough. It simply doesn't work with randomness put into it. Instead a series of lead up encounters with one at the end that normally should be do able is the best way to go about it. Eating up about a 1/3 of the parties resources before a "big event" is generally enough to put that edge on it while remaining comfortable with the degree of success you as a DM can expect from the party.

Something else that can be done is not to build up the challenger directly but to instead hinder the party with terrain, moving ground, a storm that hinders their vision and ranged combat... or even a simple fight in the dark. For these types of fights an edge is put up because the party knows it should be able to handle it... but the element of the unknown makes it harder... they have to think more to deal with the additional problems.

All in all I believe that the solution to these sorts of problems is never a bigger or badder monster, instead it's the circumstances that makes the battle tough, challenging, and ultimately memoriable.

Now if you do this with every battle though, your players will quickly tire of it. They need those easier fights to build confidence, and ability in their character. Those easier battles are when they'll start trying new things, figure how to go about different tasks while in combat and feel in control (which is just as important for a player). If every fight is a knock down drag out then the players quickly bore from the fact that it doesn't matter how good they get... they just get lucky enough to win because the next monster is just going to be that much bigger and badder.

[Azireal]

I wholeheartedly agree with everything that Niko said.


I think this is the first time this has happened, great now we have to deal with the hellfire and brimstone. this sucks.

[Choas]

Both of you post very valid points as to the problems one has when trying to make an end all be all of encounters. I agree with the scope of Niko:

the Encounter is sum of it's parts and not the other way around: aka the big bad monster does not make the encounter, although he may be the star of the show.

Another great point Niko brings up is that this absolute limit of difficulty most probably should be looked at in the length of a session, not the length of an encounter.

100% resources would mean that the party is dead. So 99.99% is the most difficult available that still leaves the players victorious.

I feel that the 99.99% is possible. But I like to push the limits.

My only real suggestion is an escalating/de-escalating model for a session long encounter, basically the encounter has certain goals that if not met make the encounter harder. If an adventure's excitement is mapped out like a wave, it has it's ups and downs. Imagine taking an encounter to this level and mapping out the large hits to resources ( hp loss, power usage, consumable usage) and then mixing it in with smaller components ( such as complicated terrain, objectives that if not met, make the scenario harder, ect). You have an up motion of conflict and player response and a down motion of monkey wrenches and player made solutions to the problems presented in the encounter.

So...with an escalating model, we can semi-project an extended duration of difficulty to stretch out player resources and push the feeling of fun, player-made input, and encounter difficulty to it's maximum.

I hope i am not too abstract.

[Niko_Kaze]

It is possible... but not something to shoot for. Consider statistics for the answer: Your truthful statistician will tell you that the minimum margin of error is 3%... so any difference of less than your margin of error is too close to offer meaningful analysis. They'll also tell you that any sample data that is worth studying statistically will fall under the bell curve with a rate of 3% outliers... that means 97% of your data collected is going to be under the bell curve... with the majority of the data within the first standard deviation (67% if I remember correctly) with most of the rest coming in the second deviation and just about 10% in the third standard deviation. Everything else is in that 3% outlier area...

So the best you can shoot for as a DM and expect your players to come out ahead with is 96% of their resources spent, and that is if your margin of error is only 3% as a DM... most people aren't even close to that so should leave themselves planning to eat 85% of the parties resources (note in statistics if your margin of error is over 5% then your study is trash and you need to start over!) Part of this is padding against the Dice. Since we rely so heavily on probability in D&D (and from what I understand even more directly in 4e than in past editions) we have to factor in the chances of automatic failures and successes into the encounter. Even if you were to run the same encounter 50 times (32 is enough to be statistically significant) with the same party at the same level of preparedness you'll get different results each time just because of the dice -- we all know this -- however depending where the difficulty of the challenge is you'll have a higher or lower failure rate for the party (to an extent... the bell curve gets a bit funky here since the general idea is to not have an even curve! However statistics can account even for this... learn how on your own time!).

All and all what I'm saying in this rambling message is:

1. Know thy party and plan accordingly
2. Leave yourself error room since you make errors and the dice force errors to happen.
3. You won't know if you hit the 99.9% until after the encounter... and you won't be able to produce the same results twice since the margin of error is bigger than the difference of resources spent in this case.

And Finally:

Big encounters should be designed to eat up 95% of the parties resources for the "BIG BAD END OF THE WORLD" Fight. Planning at 95% means you'll still have about a 20% chance of party failure and another 20% chance of TPK.

EDIT: Funny bit of information -- Gary Gygax included a basic lesson in statistics and probability in the front of the Advanced Dungeons and Dragons DMG. It explained standard deviation, what the bell curve represents and means, and how the dice affect the outcomes of the game, and what changing the different dice would do to your game. It is a shame that basic information like that isn't included in the books anymore since it really helps explain how things shape in the game.

[Choas]

Wow...never knew that was part of the game ( Gygax's info), thanks Niko.

Mathematically and statistically, my gut tells me that what you said sounds pretty right on from what I have experienced from DMing some tough shiz.

But...what about player cognitive skills and complex decision making that does not involve a dice roll( much the same as your difficult terrain or the fact that monster reinforces would come if condition X is not met). These make the encounter vastly interesting but also directly affect party resources in a different way ...time and the decision of sacrificing focus in a combat encounter.

A good scenario would be that the players are fighting a group of equal skill to them mathematically but with no consumables. The party knows that in 4 rounds that unless they kill the caster of the enemy, it will summon a group of reinforcements. However, each time they strike the caster, the cleric of the group gets to roll a d4, on a 4 he can heal any struck party member for a healing surge. Better yet, the fight is in a room where the party is filled with a particular poison built to kill them in 10 rounds and deals gradually greater damage each round. They can stop the poison by hitting a ventilation shaft switch which is behind the caster that is being guarded by a number of enemies equal to the party -1.

What does the party choose to do? Is this too hard? The enemies are equal but the circumstances and pressures of time ensure that this battle is very critical in decision making.

What say ya'll?

[MindForge]

First thing first.. you have to know your players and PC's to really answer the question... a group of strangers at a game table is probably going to be randomish for you as a DM... I usually know how my players are going to react... they are quick to flee if the situation gets dangerous because they know that I create things they can't beat... I might have a beholder wandering hallways that just should be avoided... better be avoided.

I think the best way to test players without killing them is through chain encounters... back to back waves of 3-4 encounters. I have a few boss style fights like this. Three fights back to back is rough and tests the players tactical strategy and teamwork because the combat lasts so long. I usually give them a brief rest to recharge encounter powers but that's it. The entire idea is that they don't really have time to heal themselves all the way... they literally have a few minutes (sometimes)... other times I might not let the first two fights have any time... no encounter power charging... it really depends on how my group is doing - if they are doing good... no rest... doing poorly, give them a rest... they will never know that you bent the rule for them because the rule was to offer them a break if they had a really rough fight...

Fourth Edition is unforgivable at high levels... players work together or die. It is easy for an at level encounter to stress a party's resources... 11 of them to go... Combat should get more deadly but there should be fewer but longer fights... that is how we do it and I have had fights with two down, two with just a couple hit points and one going to collapse from temp HP loss in two rounds... Boss fights are like this in my games because I know the group. I know what resources they can fight and I would not create something that they can't beat but at the same time... to take inspiration from Batman comics... the group might need skill challenges in the middle of a fight or lose a healing surge... I love doing this to my group. I used to think that players had too many healing surges and thought about ways to take them away and the skill challenges are a great way to do this. Let's say in the middle of combat the floor breaks out from the players and they all need to make skill challenges to prevent slamming into the ground below which knocks the wind out... taking a healing surge. Create a 3 phase fight with three short skill challenges and you have a very difficult encounter considering that healing is on the line.