Doing Skill Challenges Right

General questions, debates, and rants about RPGs

Moderator: Moderators

RandomCasualty2
Prince
Posts: 3295
Joined: Sun May 25, 2008 4:22 pm

Post by RandomCasualty2 »

Voss wrote: There is no way they didn't know they were shoveling a pile of shit here.
Honestly, I'm not sure about. Lets remember how long it took the designers of 3E to figure out that polymorph was broken. It was well into 3.5 before they started to say "This shit doesn't work."
SphereOfFeetMan
Knight-Baron
Posts: 562
Joined: Fri Mar 07, 2008 7:54 pm

Post by SphereOfFeetMan »

I would guess that you are right RC.

Voss, did you know that not too long before 3.5 was released, Andy Collins stated that he hadn't known of the formula behind saving throws for the majority of 3e's run? (A reminder: Andy Collins was the lead designer of 3.5)
There is nothing worse than aggressive stupidity.
- Johann Wolfgang von Goethe
Voss
Prince
Posts: 3912
Joined: Fri Mar 07, 2008 7:54 pm

Post by Voss »

RC- I think it was more a matter of 'willing to admit' that shit didn't work.


Sphere- I missed that one. Thats.... pretty damn pathetic. Executions for stupidity all around.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Post by Username17 »

So giving things extra thought:

If you give players individual reservoirs of failure rather than party wide reservoirs of failure then having people perform low odds attempts to accomplish tasks is no longer baleful to the party but is instead potentially useful.

-Username17
SphereOfFeetMan
Knight-Baron
Posts: 562
Joined: Fri Mar 07, 2008 7:54 pm

Post by SphereOfFeetMan »

FrankTrollman wrote:If you give players individual reservoirs of failure rather than party wide reservoirs of failure then having people perform low odds attempts to accomplish tasks is no longer baleful to the party but is instead potentially useful.
That solves a lot of problems. It stops problem players from negating the efforts of the party. It stops possible intra-party resentments based upon differing roleplaying ability. It also creates a stronger sense of teamwork since your teammates can only help party efforts, and not frustrate them. Finally, it makes Pc's solely responsible for their actions, and therefore could generate more thoughtful contributions.
There is nothing worse than aggressive stupidity.
- Johann Wolfgang von Goethe
Shatner
Knight-Baron
Posts: 939
Joined: Fri Mar 07, 2008 7:54 pm

Post by Shatner »

FrankTrollman wrote:So giving things extra thought:

If you give players individual reservoirs of failure rather than party wide reservoirs of failure then having people perform low odds attempts to accomplish tasks is no longer baleful to the party but is instead potentially useful.

-Username17
I don't follow. What is a "reservoir of failure" and how is one managed on an individual and party-wide level?
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Post by Username17 »

Shatner wrote:
FrankTrollman wrote:So giving things extra thought:

If you give players individual reservoirs of failure rather than party wide reservoirs of failure then having people perform low odds attempts to accomplish tasks is no longer baleful to the party but is instead potentially useful.

-Username17
I don't follow. What is a "reservoir of failure" and how is one managed on an individual and party-wide level?
It's a direct reference to the 4e Skill Challenges, which I have been ranting about on a different board because people challenged me on my assumptions about them.

In a 4e D&D Skill challenge, your party fails at a complexity 3 challenge as soon as the total party generates 4 failed rolls. This means that anyone performing a low odds anything harms the party. Indeed, this means that any player character performing any action other than the most effective maneuver by the most effective team mate is actively harmful to the team's chances of success. That's bad, and it contributes to the 4e rubric where everyone just throws the Diplomancer forward and watches them roll Diplomacy nine times. Which is boring and retarded.

But I was thinking: if instead each character could generate a number of failures before that character was out of the challenge, then even players with low odds chances would still roll dice and role play. The Hobgoblin Malefactor would still want to look around because he might notice something important, rather than just have him shut his eyes and whistle because the Human Gadgeteer is much better at finding and identifying gizmos than he is.

A different, but similar method of accumulating success and failure inverts the paradigm and rewards everyone for participating. By having Team Success and Individual Failure, cooperation and contribution is always rewarded or at worst irrelevant and hopefully funny.

-Username17
RandomCasualty2
Prince
Posts: 3295
Joined: Sun May 25, 2008 4:22 pm

Post by RandomCasualty2 »

Actually pretty much the best skill use for a nonskilled character is do aid another. In the DMG they talk about doing a climb skill challenge by just having everyone just do aid another on the best climber. Especially given that aid another Dcs are always 10, you're going to always be able to aid another at higher levels, even if untrained.

Basically you just have your best character make the rolls with like a +8 bonus from Aid another.
Last edited by RandomCasualty2 on Wed Jun 18, 2008 6:18 pm, edited 1 time in total.
Voss
Prince
Posts: 3912
Joined: Fri Mar 07, 2008 7:54 pm

Post by Voss »

Ugh. That makes for a fun game. Everybody give Bob a bonus so he succeeds every time. Uh-oh. Now we have to Intimidate someone. Now everybody aid Fred so he auto-wins.

Eh... with how that works out, all enemies should automatically attack the party, rather than risk automatic diplomatizing.
Manxome
Knight-Baron
Posts: 977
Joined: Fri Mar 07, 2008 7:54 pm

Post by Manxome »

The tricky bit with team success and individual failure is that your chances of success always go up as your party gets larger. That is realistic in many cases, but it means that you need to know the party's size as well as level in order to set a reasonable target number, and it leads to illogical results in some other cases (e.g. a horde of 1000 barbarians can diplomatize someone better than a bard because they get at least 1000 chances to roll natural 20s).

However, if you assume that party size is fairly consistent and that we only care about rolls that directly involve the PCs, it probably works pretty well.
User avatar
virgil
King
Posts: 6339
Joined: Fri Mar 07, 2008 7:54 pm

Post by virgil »

The horde of barbarians technique shouldn't work if you don't allow natural 20s to auto-succeed on skill checks, which means that getting help only helps when the assistant has competence. Also, diminishing returns could be applied.
Come see Sprockets & Serials
How do you confuse a barbarian?
Put a greatsword a maul and a greataxe in a room and ask them to take their pick
EXPLOSIVE RUNES!
Manxome
Knight-Baron
Posts: 977
Joined: Fri Mar 07, 2008 7:54 pm

Post by Manxome »

The horde of barbarians technique shouldn't work if you don't allow natural 20s to auto-succeed on skill checks, which means that getting help only helps when the assistant has competence.
The point of that example wasn't to show how you could cheese the game, but to point out that there are some tasks where it is really weird (from a "realism" perspective) to think that lots of unskilled people will do better than one (or a few) skilled people. Even with your rule, that's still possible in some cases. It's not necessarily a big enough problem to make one want to use a different system, but it's worth considering.

Your suggested rule also leads to the following conclusions, which may or may not be desirable:

1) None of the barbarians, individually or collectively, can ever possibly succeed at the task in question.
2) If one of the barbarians were traveling with the bard, he couldn't possibly aid the bard's diplomacy attempt, so he's back to sitting on his hands while the bard rolls diplomacy more times.
3) You've pushed someone entirely off the RNG
Also, diminishing returns could be applied.
I don't see an obvious, easy-to-implement way to do this, especially if you want to maintain the principle that letting someone bad at the task attempt to help will never hurt the team. If you do, please share.
User avatar
Talisman
Duke
Posts: 1109
Joined: Fri Mar 07, 2008 7:54 pm
Location: The Cliffs of Insanity!

Post by Talisman »

Why should someone who is bad at the task never hurt the team? I can understandwhat y'all are talking about here, but it seems to be taking it too far to go from "individual failures always hurt the team" to "individual failures never hurt the team unless everyone fails horrifically."

Example: When I run, people often try to Aid Another on Diplomacy. My houserule is that, if the aider fails horribly (natural 1) they impose a -4 penalty on the main Diplo check, instead of the usual +2. It makes sense because more participants = more chances to screw up, but the main diplomat is (presumably) skilled enough to cover all but the worst gaffes.
MartinHarper wrote:Babies are difficult to acquire in comparison to other sources of nutrition.
User avatar
virgil
King
Posts: 6339
Joined: Fri Mar 07, 2008 7:54 pm

Post by virgil »

I was just assuming that the horde was lower level than the bard, where it is acceptable for them to not be on the RNG in comparison.

As for diminishing returns, the idea would be that when you have 10+ getting collective on a task, you begin combining their efforts. You start requiring a group of 2+ act/roll as one person, so that there are only five rollers.

Maybe certain skill challenges allow a logarithmic increase to the level of each roller when it's combined in that manner, such that 1000 Level 2 barbarians act as 4 Level 18 barbarians. This would be for large-scale construction challenges, likely.

Diminishing returns does marginalize the individual in this manner. But that's to be expected when you try to have an army help with a single skill challenge.
Come see Sprockets & Serials
How do you confuse a barbarian?
Put a greatsword a maul and a greataxe in a room and ask them to take their pick
EXPLOSIVE RUNES!
Draco_Argentum
Duke
Posts: 2434
Joined: Fri Mar 07, 2008 7:54 pm

Post by Draco_Argentum »

Just cap the maximum number of people that can help at the party's size. All the PCs can help, which is good. Enemies won't be able to horde up and auto beat the PCs at the challenge.

Or, for level based skills just ratchet up the DCs. A horde of guys with a small success rate will be equal to PCs with a much larger success rate.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Post by Username17 »

Certainly a possibility is to have tasks be given a "size" which dictates how many people can work on them. Obviously it pays to give out tasks of a size that is roughly that of the PCs most of the time.

-Username17
RandomCasualty2
Prince
Posts: 3295
Joined: Sun May 25, 2008 4:22 pm

Post by RandomCasualty2 »

Draco_Argentum wrote:Just cap the maximum number of people that can help at the party's size. All the PCs can help, which is good. Enemies won't be able to horde up and auto beat the PCs at the challenge.

Or, for level based skills just ratchet up the DCs. A horde of guys with a small success rate will be equal to PCs with a much larger success rate.
I just think the Aid another scenario is really boring, so I don't think I'd want to allow it at all. It discourages Pcs from trying to be creative and instead encourages them to just spam aid another on the guy with the largest bonus. That sucks. You shouldn't be able to contribute no successes yourself and just help other people.

One variant skill challenge idea that would be interesting is setting a maximum limit to the number of rounds in the challenge that ends the challenge instead of the number of failures. Instead of a generic pool of successes for the party, Each character can succeed or fail at the challenge individually based on how many successes they have.

To determine if the overall challenge is a win or failure, a certain percentage of the party must succeed. This means that basically you can use aid another to help the people slacking behind if you finish early and have rounds to burn. Alternately, you could just admit defeat from the beginning and invest in helping people. But if you just help people, you can't succeed individually. There may be a time when you want to just aid another, but you certainly wouldn't want to do that all the time.
Draco_Argentum
Duke
Posts: 2434
Joined: Fri Mar 07, 2008 7:54 pm

Post by Draco_Argentum »

RandomCasualty2 wrote:I just think the Aid another scenario is really boring, so I don't think I'd want to allow it at all. It discourages Pcs from trying to be creative and instead encourages them to just spam aid another on the guy with the largest bonus. That sucks. You shouldn't be able to contribute no successes yourself and just help other people.
I meant that as an add to Frank's idea. Everyone still tries to get successes of their own. But there is a limit to the number of people, including the number of people who just spam aid another.

You idea for a max duration sounds decent. How would that work with a challenge where the PCs brought successes along at the start. The prior military victory coming to a diplomacy example or similar.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Post by Username17 »

Probably the best way to do priorly gained successes is to allow people to cash them in during a round. So if you already defeated the Siege Golem you can have one of your characters spend their challenge round dropping a success on the pile. You could even reward player skills by allowing unusual skills to be used to bust that up to two successes (with no consequence of failure).

So looking at the Negotiation: normally you are rolling out Diplomacy, Bluff, and Insight. But if you recently killed Korlax the Dragon you can cash that in for a success and make a free bonus Intimidate or Monster Lord test to potentially double it. This means that a party barbarian or sage can throw down special like on a test or two even though they don't have any of the basic skills. On the flip side, you can have someone spend challenge actions carrying explosives that have already been produced or purchased around while doing your whole sabotage thing. The key is that finding things for people to do is your biggest challenge in the challenge.

--

Probably what you want for a challenge is:
  • Size: The number of actions that can be taken each challenge round. My suggestion is to have most challenge sizes be slightly larger than the team size so that people can find it useful to own dogs and shit ("And then Rex will try to follow that game trail with his nose...")

    Difficulty: This is the number of successes the test requires.

    Length: This is the number of challenge rounds the test allows. In most cases, this should be a hidden number.
Now, it's entirely possible to have other rules. Like have skills that have already been used "turn off" or simply become more difficult. That would encourage people to move skills around. Possibly doing certain things (like using other skills for an entire round or more) should reset skills and allow you to use them again normally. Possibly not, mostly depends on how many skills people have.

A very important option in such challenges is the delegation action. That is, you can spend your action hiring or procuring a specialist to use their skill on your behalf.

-Username17
MartinHarper
Knight-Baron
Posts: 703
Joined: Fri Mar 07, 2008 7:54 pm

Post by MartinHarper »

So, something like:
Complexity 1 (2 successes in 1 round)
Complexity 2 (4 successes in 2 rounds)
Complexity 3 (6 successes in 3 rounds)
Complexity 4 (8 successes in 4 rounds)

With 4 PCs, 55% success rate, we get nice results:
Complexity 1: 76%
Complexity 2: 74%
Complexity 3: 74%
Complexity 4: 74%

If the PCs all get a +2 circumstance bonus:
Complexity 1: 87%
Complexity 2: 89%
Complexity 3: 92%
Complexity 4: 93%

If they get a -2 circumstance bonus:
Complexity 1: 61%
Complexity 2: 52%
Complexity 3: 47%
Complexity 4: 44%

Looks tough for three PCs. With 55% success rate:
Complexity 1: 57%
Complexity 2: 44%
Complexity 3: 36%
Complexity 4: 30%

5 PCs makes it real easy. With 55% success rate:
Complexity 1: 87%
Complexity 2: 90%
Complexity 3: 92%
Complexity 4: 94%

I'm getting the feeling that rolling lots of dice and counting successes is inherently divergent, so the system doesn't handle additional PCs or small changes in success rate very gracefully.
Username17
Serious Badass
Posts: 29894
Joined: Fri Mar 07, 2008 7:54 pm

Post by Username17 »

Rolling more dice and getting more successes makes it more likely that you won't succeed if you were unlikely to succeed and more likely to win if you were probably going to win. 50% stays the same if the ratios stay the same, but a 49% goes down and a 51% goes up.

What should probably stay about the same is challenge Size vs. Required Hits ratios. This means that if the challenge is size 3 your expected result (success or failure overall) should be the same with 3 people working on it as they would be if it was size 6 with 6 people working on it.

Complex problems which allow more rounds should probably require a different number of expected Hits per round than ones which allowed less rounds. My initial thought was that the ratio of demanded hits should go up as turns did (making longer challenges more difficult inherently). But after thinking about it more, they should probably actually demand less hits per round because players are going to burn through super powers and automatic Hits and their top skills quickly, meaning that later challenge turns are going to end up with players throwing their lot in with weird tertiary skills that might not even work. The kitchen sink effect of challenge turns will cause later challenge turns to be inherently less fruitful than early challenge turns, meaning that longer challenges can demand a lower standard of success than shorter ones.

-Username17
Post Reply