The Art and Science of Skill Challenges

44 posts / 0 new
Last post
paraphrased wrote:
4e skill challenges don't allow roleplay

I think that's a big issue for a lot of 3.5 players and especially DMs in running skill challenges. Which is to say, they run them badly to horrible. And writers often make things that shouldn't be skill challenges. Examples:
18 Str/Con Character with no useful mental skills in a mental skill challenge tries to convince the DM that he could use Athletics. DM feels sorry for him because the DM hasn't read the errata and counts it as a Diplomacy check.

Character with a +1 in the appropriate skill announces that he's doing an action that likely involves that skill. DM has them roll and instant failure.

Writer makes what should be a straightforward diplomacy check into a skill challenge.

DM, in an effort to prevent rolls that kill the skill challenge, lists all the skills that can be used which turns it into a set of rolls.

-------

Plus people ignore the errata - there are no longer rounds, no character has to make a roll, and the DM should at most allow one to two assists.

Here's the way I think they should be done:
Highest skilled participating pc makes the roll
Second highest makes the assist
Just because the PCs are trying for a particular skill does not mean you can't end up in another one - best example is Diplomacy vs. Bluff or Intimidate

What does this mean?
Everyone can roleplay without the threat of failure by an individual. Just because the +1 perception character decides to look around the store doesn't mean he's going to roll the check.

DMs don't feel sorry for the 20 stat character with no useful skills. You know that 20 Str/16 Con/11 Wis Warforged with the high Athletics and Endurance and nothing else. This eliminates the bizarre skill rolls when those two skills aren't applicable and they're then forced to roll.

Players do need to meet the participation requirement and think about what they're saying - the guy who threatens the NPC who can't be intimidated as an example. Even if you're the guy with a Diplomacy 10 points higher than anyone else, the DM still needs to feel that you participated in the discussion.

---------

Thoughts? Other things that have worked for people?
I
Here's the way I think they should be done:
Highest skilled participating pc makes the roll
Second highest makes the assist
Just because the PCs are trying for a particular skill does not mean you can't end up in another one - best example is Diplomacy vs. Bluff or Intimidate
...
Thoughts? Other things that have worked for people?

I don't care for play where no matter who is actually involved in the encounter, the bests stats are used -- in fact, I generally find that as soon as someone says "who has the best X?" roleplaying takes a huge hit. I also dislike having people uninvolved. I'm not sure that the proposal addresses that issue.

What has worked for me is running skill challenges like this:

I want everyone to make a roll. In combat, everyone is a potential target, and making skill challenges 'special' in that you can avoid being a part of it makes it mechanically more advantageous to design characters as all-combat, no anything else. So I will ask everyone to make a roll either as a primary or as an assist. Frankly, if you have 5 people in a room, thinking that one person can talk and the rest have no effect at all is unlikely. They are part of the challenge, and must at least try something.

In practice this means that at least two prime movers are needed and then two - four people, even if they are incompetent, can assist. If your party cannot muster 2 people competent at any particular challenge, then I would argue that either you are really unlucky or, more likely, you are designing all-combat characters.

I will also allow people to use different skills to assist. For example, in diplomacy, one person could assist with bluff, insight, intimidate, history, religion very easily.

Overall I have found few problems with this. It seems balanced, everyone contributes and it allows people to roll with their +1 skills and know they won't screw over the party. It also, as far as I can see, does not require the rules of skill challenges to be changed.

- Graham
My perference as a DM, writer and player is to have a hybrid of the role playing backed up by the skill checks. Social situations it works very well IMHO. However, there are some situations that has not worked as well (typically very mechanical activities hard to role play). It may be hard to role play some knowledge checks (role play that you are thinking hard), but others possible (discussing points of history amongst the group).

I do warn players that if they role play their PC (for instance) as trying to bluff a NPC, then I will ask for a bluff check. If they are lousy at it, well, why is that different from being lousy at bluff but trying it anyway? Substitute whatever skill. I think it is reasonable to find out who is good at what, but I don't enjoy tables when players start calculating how to maximize the roll without even thinking about what it is they are doing.

The authors or DM need to provide enough insight (not the game mechanic kind) about the scene so that players might have ideas about what they might do. Sometimes though players are just stumped as to what to do, and then the DM has to provide some hints.

Some DMs are still not comfortable with allowing skills that are not explicitly listed in the skill challenge encounter although creativity should always be encouraged. Some DMs are not comfortable with a broad interpretation of what can be done with a skill. I often try challenging them (if they say "no") with a question, "If that skill does not do X, then which skill do you feel will accomplish X?" X being an activity, not an outcome.


Keith
Keith Hoffman LFR Writing Director for Waterdeep
Overall I have found few problems with this. It seems balanced, everyone contributes and it allows people to roll with their +1 skills and know they won't screw over the party. It also, as far as I can see, does not require the rules of skill challenges to be changed.

Actually, that's using the old rules of skill challenges - while it is certainly within the realm of DM power to decide that all characters must make rolls, that's half of the specifically deleted text, the 3rd paragraph of page 74.

And on the contrary, I think when combat monsters can't participate tends to make it more likely that they go away - forcing rolls makes it more likely you have a combat monster put in the position of barely contributing - which penalizes the party instead of the character. So DMs let them get away with a stupid skill attempt.

When you don't allow the stupid skill attempts, they turn into what they are - a roleless character in a skill challenge and the parties that they're with always seem to fail skill challenges because they don't cover all the vital skills.
18 Str/Con Character with no useful mental skills in a mental skill challenge tries to convince the DM that he could use Athletics. DM feels sorry for him because the DM hasn't read the errata and counts it as a Diplomacy check.

It doesn't help that the primary interaction skill for fighters is Intimidate, which, in LFR, is the Russian roulette of skills--maybe it'll work, maybe you'll cause an automatic failure just by trying the skill.

Just once, I'd like to see some other skill get that treatment. ("The mayor is deeply atheistic. Attempts to use the Religion skill cause the skill challenge to automatically end in a failure.")
Plus people ignore the errata - there are no longer rounds, no character has to make a roll, and the DM should at most allow one to two assists.

The original version was flawed but the errata deserves to be ignored, er, DME'ed away. Personally, I usually DME the skill challenge into something closer to the player-created Obsidian skill challenge system.

The entire point of the skill challenge system was to have a non-combat encounter in which everyone participates, as opposed to the old system where everyone sits back and watches as the rogue makes rolls to disarm the trap or the bard makes rolls to be diplomatic. The fact that the implementation of that idea was done so poorly doesn't make the goal any less worthy.

Basically, under the current core system, players are penalized for doing anything except determining quickly who has the highest bonus for a primary skill for the challenge, then sitting back and letting that person make all of the rolls. I consider that to be a bad thing.

-- Brian Gibbons.
It doesn't help that the primary interaction skill for fighters is Intimidate, which, in LFR, is the Russian roulette of skills--maybe it'll work, maybe you'll cause an automatic failure just by trying the skill.

Just once, I'd like to see some other skill get that treatment. ("The mayor is deeply atheistic. Attempts to use the Religion skill cause the skill challenge to automatically end in a failure.")

Thievery often has the same treatment.
Dave Kay LFR Writing Director Retiree dkay807 [at] yahoo [dot] com
I agree about intimidate. The writers seem to have something against it. The number of skill challenges I've seen in which intimidate involves a high DC, or just outright automatic failure (with no indication that it should be so) doesn't seem fair to those who have trained it.

As for determining who has the best mod for a skill and letting them roll it, is that not what a party would do in real life? Surely they'd all realise that the charismatic paladin should probably do the talking, and the clever wizard ought to be the one researching the arcane symbol in the library, etc.

As for the 18 Strength/Con warforged with no skills, I have very little sympathy. If you're going to build a one-dimensional combat machine, why should the DM feel sorry for you and make a special effort to engage you in the skill challenge? What about the party members who took a more rounded stat and skill selection? Surely that's their reward for making such a character?

(Ok, fighters in particular do get the shaft, skills-wise. But hey, they're arguably one of, if not the best class in combat, so swings and roundabouts).
As for the 18 Strength/Con warforged with no skills, I have very little sympathy. If you're going to build a one-dimensional combat machine, why should the DM feel sorry for you and make a special effort to engage you in the skill challenge? What about the party members who took a more rounded stat and skill selection? Surely that's their reward for making such a character?

(Ok, fighters in particular do get the shaft, skills-wise. But hey, they're arguably one of, if not the best class in combat, so swings and roundabouts).

Hey, some of us str/con warforged fighters actually chose regions and bought magic items specifically so that we CAN participate in skill challenges...

I do agree that in most cases, parties should choose the best candidate for the job when engaged in skill challenges (from a tactical and RP perspective). I also feel that authors should make an effort to include a variety of skills in their challenges, or provide multiple challenges, each that caters to a different skillset.
Dave Kay LFR Writing Director Retiree dkay807 [at] yahoo [dot] com
Hey, some of us str/con warforged fighters actually chose regions and bought magic items specifically so that we CAN participate in skill challenges...

Fair enough, and good for you for doing so. I did the same with my Dwarf fighter. But she still only has three trained skills (will get a couple more via feats some day), so on some skill challenges has to sit out. That's just the way of it. I don't expect every skill challenge, or even every adventure, to require endurance and dungeoneering checks. Some mods certain characters will shine more than others. As long as the skill challenges vary between mods so every skill is used over the course of a few play sessions (there have been a very large number of mods with city-based 'Gather Information' skill challenges, with diplomacy primary, for example, and relatively few 'physical' skill challenges I've seen).
As for determining who has the best mod for a skill and letting them roll it, is that not what a party would do in real life? Surely they'd all realise that the charismatic paladin should probably do the talking, and the clever wizard ought to be the one researching the arcane symbol in the library, etc.

Sure, in the same way that, if players could choose, combats would be best structured as a one-on-one encounter between the combat monster PC and the enemy, with the other PCs just standing around buffing the main PC up.

The basic structure of D&D combat is that, generally, everyone participates. We don't average the combat skill of the party; having a weaker PC enter combat doesn't hurt the stronger PC's chances.

WOTC tried to create something similar for out-of-combat challenges, but so screwed it up that the end result is now basically just a more complicated version of the 3e single-PC single-skill everyone-else-aid-other system--essentially, post-errata, the 4e skill challenge is just the 3e system, with a bunch of sub-optimal options tacked on.

4e now gives you a choice between boring-but-optimal and flavorful-but-suboptimal. No thanks.

The basic flaw of the current skill challenge system is using failures as the end-point, which means that a PC can hurt the group by getting involved. Instead, flip things around so that staying out of the challenge is equivalent to failing (in the same way that the party doesn't benefit if someone stays out of combat).

Instead of saying that the party loses when they have X number of failures, say that the party loses if they have not achieved a certain number of successes in N rounds. In the former case, only the optimized need apply; in the latter, participation doesn't ever hurt the party.

While this largely works for right now (well, you'll have difficulty with encounters that are really single-PC encounters shoehorned into the skill challenge system), I expect that it will become more problematic in the future, as skill challenges and skill DCs become more based on the idea of a single PC taking care of the challenge.

-- Brian Gibbons.
One thing I don't like about skill challenges is how some DMs will ask for the Diplomacy check from the first character to open his mouth. That doesn't make sense to me.

Consider a situation where the party is looking for a tall man wearing a red cloak.

PC: "Mr. Merchant, have you seen a tall man wearing a red cloak?"
Merchant: "I see him visiting the fortune-teller all the time."

That's a situation where a skill roll is not needed. The merchant is happy to provide the information.

PC 1: "Mr. Merchant, when are you closing up for the day?"
Merchant: "Come sundown, I'll be heading home."
PC 1: "Have you seen a tall man wearing a red cloak?"
Merchant: "Can't say I have." (Someone with a high passive Insight check recognizes he's not being entirely honest.)
PC 2: "Please, it is very important that we find this man. Innocent lives may be at stake. Could you please think if you may have seen him? We won't mention this to anyone."

In this case, we are looking for some skill check to get the merchant to divulge this information. PC 2 is using Diplomacy. But PC 1 isn't using Diplomacy. He asks the first question, which the merchant happily answers, and when he asks the second question, the merchant doesn't answer. PC 1, by asking the second question, isn't trying to get the merchant to do anything he isn't normally inclined to do, so he's not using any skill.
Instead of saying that the party loses when they have X number of failures, say that the party loses if they have not achieved a certain number of successes in N rounds. In the former case, only the optimized need apply; in the latter, participation doesn't ever hurt the party.

Icy Spire does a nice job of this. Each "round," players can make their own checks, assist others, or sit out. Of course, each "round," each character must also make a
Show
n endurance
check, or lose a healing surge. There's a reward for "finishing quickly" and "working together as a team."

CORM1-1 has a similarly well-written skill challenge. It introduced the idea of "group challenges" - checks that all must make, and as long as at least half of the group makes the check, the group earns a success.

I'd like to see more of these challenges, especially one written with a changing "base skill." For example, "Chase Scenes" are common Skill Challenges. In addition to having characters make checks (or assist), the first round could require everyone to make an Athletics (easy) check to jump a fallen branch. Round 2 - again, in addition to the main challenge - Acrobatics (easy) to avoid low branches. Round 3 - how about Intimidate (easy) to shout commoners out of the way! Round 4, Endurance (easy) to keep up the long chase . . .

These "easy" skill checks are within the realm for everyone to make, and certainly give some reward for participating in the skill challenge.

I think that true role-playing can get a boost by the DM giving a bonus (+2/+4) for particularly well portrayed uses of a skill - rather than "Can I just roll?"

Dan Anderson @EpicUthrac
Total Confusion www.totalcon.com
LFR Calimshan Writing Director
LFR Epic Writing Director

LFR Myth Drannor Writing Director

What bothers me is that I've seen several cases where a skill challenge turned into a contest of "making up weak excuses for why one's best skill is applicable to the current situation, even if that doesn't make an ounce of sense".

For instance, using endurance in place of diplomacy (huh???). Players with enough ranks in the "fast talk DM" skill can apparently get away with that.
What bothers me is that I've seen several cases where a skill challenge turned into a contest of "making up weak excuses for why one's best skill is applicable to the current situation, even if that doesn't make an ounce of sense".

I think that's why the lack of a forced roll is a good thing, not a bad one. When the DM forces a roll to happen, then one of four things is going to happen:
The player has to roll the skill. Say 80% chance of failure.
The player makes something up and gets DM sympathy.
The player makes something up and the DM says no, you can't do that.
The player makes something up, DM lets him roll it anyway, then just gives him a failure.

None of those options for the player making something up are all that great for the DM. #2 is the DM caving, #3 means the player just does #1 anyway, and #4 is the DM being a jerk. So making something up is worthwhile, because if you don't, you get the 80% chance of failure for certain.

On the other hand, the DM isn't forcing a roll, then it doesn't work the same way.
For my part, I don't think the skill challenge mechanic is salvageable. I've played through quite a few of them and have not really enjoyed any of them so my attitude now is just to roll with it. Find the person with the highest bonus, everyone assist and roll dice until you win, then move on to something that might actually be fun.

Even when the skill challenge mechanic does--sort of--work, it removes whatever the PCs were trying to an element of abstraction so far removed from what the PCs are, in game, trying to do, that you are playing a mini-game rather than participating in a story. (For instance, the party is trying to slip over the wall into the forbidden zone. If there is no skill challenge, they might go into a shop with a view of the forbidden zone, spend an excessive amount of time with one or two characters window shopping and/or haggling while the other characters observe the guard patrol and learn their schedule, then the PCs slip to the zone in a break between patrols and one character delays the guards (by reporting a crime and generally being an ass) while the other characters slip over the wall in that window of opportunity. The character who reported the crime then fey steps (or ethereal strides) to the top of the wall when the guards turn away and hops over. The distinct differences in this approach from the skill challenge approach are as follows:

1. The characters make a coherent plan that works together to effect an actual goal rather than every character who is good at a semi-appropriate skill describing an uncoordinated attempt to use that skill and other characters aiding that skill in the effort to get an individual success three to twelve times.

2. The characters deal with actual obstacles rather than arbitrarily scaling DCs. The PCs need to get over a 15 foot wall without being stopped. If three people need to make DC 12 athletics checks, then you can't do anything to make it easier (though some DMs might grant an autosuccess for use of fey step or ethereal stride). On the other hand, if you are treating the wall as the obstacle rather than the need to get a set number of successes, then the PCs may be able to find a way to smuggle a ladder (or a ladder-like object) to the wall and make it accomplishable by the less athletic. In a skill challenge, I have yet to see a DM reduce the DC to account for player precautions or preparations. At most, bringing the ladder usually gives one character a +2 item bonus or something.

3. Once the task is accomplished or failed, the planned approach to dealing with concrete obstacles is done. On the other hand, I have seen several occasions where the players have come up with a way to accomplish the "task" set by a skill challenge with fewer checks than the skill challenge required. In those cases, I have on several occasions been told, "OK, you're over the wall now, but the skill challenge still requires two more successes, so someone roll another stealth check or another athletics check."

I do not think these are simply problems with skill challenges being run poorly. I think, instead, that they are inherent to a fundamentally flawed system.
For my part, I don't think the skill challenge mechanic is salvageable. I've played through quite a few of them and have not really enjoyed any of them so my attitude now is just to roll with it. Find the person with the highest bonus, everyone assist and roll dice until you win, then move on to something that might actually be fun.

Even when the skill challenge mechanic does--sort of--work, it removes whatever the PCs were trying to an element of abstraction so far removed from what the PCs are, in game, trying to do, that you are playing a mini-game rather than participating in a story. (For instance, the party is trying to slip over the wall into the forbidden zone. If there is no skill challenge, they might go into a shop with a view of the forbidden zone, spend an excessive amount of time with one or two characters window shopping and/or haggling while the other characters observe the guard patrol and learn their schedule, then the PCs slip to the zone in a break between patrols and one character delays the guards (by reporting a crime and generally being an ass) while the other characters slip over the wall in that window of opportunity. The character who reported the crime then fey steps (or ethereal strides) to the top of the wall when the guards turn away and hops over.

From my side of the DM screen you just described a perfectly valid skill challenge to me. Let's see:

Bluff or Diplomacy to keep the shopkeeper interested in them as customers and not to draw any unwanted attention or get shoved out of the shop.

Insight and Perception to get a feel for the watch patrols and to notice that towerguard who keeps a look over this stretch of wall.

Athletics or Acrobatics to scale the wall maybe aided by a ladder that is "smuggled in" with a stealth check. Auto success if the ladder is present, but I agree some DMs tend to be less flexible with the rules.

And finally another Bluff or Diplomacy to distract the guards, maybe substituted for a streetwise to send a pack of children chasing behind a ball at them and maybe a Stealth check to get onto the wall and hide immediately.

I think the biggest problem with SC is divided in two parts. First the author of a model should be very careful when designing a SC. Mike Mearls his column in Dragon should be obligatory for every LFR writer in that regard. The right balance of skills and the right amount of scenes gives the DM the tools needed to bring the SC to life.

The second part is the synergy between the DM and his players. My group has been playing 4th edition for over a year already and we quickly learned that a SC stands or falls with both the DM and the players. Key is to not ask a roll for nearly everything and don't be afraid to improvise or do something entirely different. In a recent playtest we came up with an entire other line of thought than the adventure was written for. An impromptu SC was set up by our DM, tying in with the SC written in the adventure and we had a blast. That specific path now is part of the SC in the actual adventure and when I ran it last week my players had boatloads of fun, they even did far more checks than required for the SC but as long as the scenes progressed and the players had the feeling they were accomplishing something I just kept them going.

Last in 4th edition it is important to equip your character for a SC. You have feats a plenty and using 1 or 2 to get some extra skills or multiclass feat can really round out your skill selection. I always try to take at least one physical, one social and one knowledge skill making sure I can always try to engage in an SC.
I guess where I deviate from you Elder Basilisk is that I view all of D&D as flawed systems. The questions come up with any of the mechanics, from levels to hit points to classes and there aren't any perfect answers. I think the skill challenges are far from perfect, but they do a neat mechanical trick, which is to make the non-combat portion of the game matter to the players as part of the system. Does it do it perfectly as a pure role play mode or a perfect mechanistic system? Nope. It is one of the many things you have to get past to enjoy. Can't get past it? Cool, make suggestions, try other ideas, experiment. Tell others about your ideas, you may yet hit the perfect mechanic.

Living Forgotten Realms may not be the ideal place to do these experiments however, it is a pretty legalistic place, but if you really can't get past the whole skill challenge thing they are a part of the game experience. Me, I can get past my issues and still have a blast playing with fun people, even if part of my brain twitches when I think about some of the mechanics. I experiment in my home games and play it by the book at LFR events and I have fun at both. I think how you approach skill challenges matters, if you go in negative it will almost always be negative, if you go in positive it will almost always be positive. When I play I try to role play during skill challenges, when I DM I try to lead the players to role play them as well. Everybody seems to enjoy themselves and it sure beats working.
It is no surprise that what I described can be framed as a skill challenge. The basic scenario is ripped out of core 1-1. However, I have yet to see a skill challenge go down that way, and I don't expect that I ever will. Here is why:

Planning is about overcoming an obstacle as a team. If your team has weaknesses (for instance, Sir Clanksalot the paladin in a stealth scenario), you have to come up with plans that do not depend upon your weaknesses.

Skill challenges, on the other hand, are about achieving X successes before Y failures. If your team has weaknesses, players can either come up win increasingly farfetched scenarios to use skills that cater to their strengths instead, or, more commonly, keep their mouths shut and just let the players who are good at the skills roll. Thus in the described scenario, the party does not worry about a Stealth check for Sir Clanksalot. They know that everyone won't have to roll a stealth check so Sir Clanksalot uses Aid Other while Nimblefingers Tinytoes sneaks across the alley.

Planning requires coming up with a plan that will work for everyone in your party and interacting with the world in a way that facilitates the plan. Characters planning to get over the wall will scout it out--walk the circumference of the wall and see if there are any areas where the alley is narrower than others. They will see if there are any areas that are poorly lit. They will see if there are any houses that are taller than the wall on the opposite side of the street.

Skill challenges, on the other hand require coming up with a rational excuse to use a skill that someone in your party is good at six or twelve times. You may try to use perception to notice if there is an area where the alley is more poorly lit or where there is ivy growing on the wall or something, but 3/4 DMs will not even give you a real success for that--just a +2 on the next person's roll. Consequently, that level of thought is only useful when you need to coax an Aid Other type effect out of a DM with better odds of success than using the same skill as the other guy.

In a scenario that calls for planning, also, some failures are more consequential than others. Failure on the Bluff or Diplomacy roll to stay in the shop simply results in being asked to leave and having to go to the shop across the road, get a disguise, or actually buy something. Failure on the Perception checks to check out the guard rotation of the wall, on the other hand, is likely to result in the guard showing up at the wrong time and ruining the whole plan. Likewise, in Sembia, failure on the bluff/diplomacy check when reporting the crime to distract the patrol, could result in the character being given a thorough beating (though that could actually be better than success for distraction).

In a skill challenge, on the other hand, it would usually be bad form for a judge to inflict penlaties beyond a simple failure on a player. For instance, when I ran the core special, I had a player who insisted on rolling intimidate checks with his fighter even though he wasn't especially good at it and I hinted that the people were tough to intimidate. So, we impromptu narrated that he found an urchin and threatened to beat him up if he didn't spill some info. First check failed and the urchin ran away. Second check succeeded and another guy gave him some info. Third attempt failed again and the urchin ran away and said "my brother's an enforcer with the XYZ gang; you can't treat me like that." Now, given the scenario, I think that, as a part of the narrative, it would be perfectly justified for that kind of a tactic and a failure to result in additional combats or a lost healing surge or two for the fighter as he spends his time brawling in the streets. On the other hand, it is strongly discouraged by the LFR rules to do that kind of thing (DME could conceivably cover it but that would be stretching quite a bit). It is also not fair to the player or the character. What I ended up doing was simply describing a scene or two later when he met with the rest of the party--he was running away from a mob made up of the relatives and fellow gang members of the people he had been intimidating. Bottom line, in a skill challenge, a failure is almost always just a failure.

The failure of the skill challenge model is that the player behavior that it encourages is almost 180 degrees opposite from the kind of behavior that would be required to overcome conceptually similar challenges outside the skill challenge format.

It is also very difficult to go for half-measures. For instance, one time I played Wat 1-1, our party made all the right choices and got the name of the correct pawn shop guy on our third skill check. Our DM, on the other hand, had to drag the darn thing out for another seven or eight skill checks with every named NPC in the mod so that we could make our twelve required successes. That is a serious risk you run when trying to make the skill challenge concrete. It is simply not possible to ask the right questions or go to the right people, or simply jump to the correct conclusion, and get your final info early. You have to slog through all of the vignettes in order to succeed.

That is why, despite your being able to model my proposed scenario as a skill challenge, you will never have actual skill challenges work out that way. The skill challenge framework requires several more levels of abstraction that make concrete planning impossible and additionally, the set success/failure ratio and consequences mean that plans will not have their expected results for success ("but this is complexity five--you haven't won yet; you've got seven more skill checks to go") or failure ("well, you failed your third athletics check climbing the cliff of ultimate doom--you plummet to the ground and... sustain a failure. Move on. No, you cannot eliminate the failure by using your feather fall utility power. Stop trying to cheat.")

From my side of the DM screen you just described a perfectly valid skill challenge to me. Let's see:

Bluff or Diplomacy to keep the shopkeeper interested in them as customers and not to draw any unwanted attention or get shoved out of the shop.

Insight and Perception to get a feel for the watch patrols and to notice that towerguard who keeps a look over this stretch of wall.

Athletics or Acrobatics to scale the wall maybe aided by a ladder that is "smuggled in" with a stealth check. Auto success if the ladder is present, but I agree some DMs tend to be less flexible with the rules.

And finally another Bluff or Diplomacy to distract the guards, maybe substituted for a streetwise to send a pack of children chasing behind a ball at them and maybe a Stealth check to get onto the wall and hide immediately.

I think the biggest problem with SC is divided in two parts. First the author of a model should be very careful when designing a SC. Mike Mearls his column in Dragon should be obligatory for every LFR writer in that regard. The right balance of skills and the right amount of scenes gives the DM the tools needed to bring the SC to life.

The second part is the synergy between the DM and his players. My group has been playing 4th edition for over a year already and we quickly learned that a SC stands or falls with both the DM and the players. Key is to not ask a roll for nearly everything and don't be afraid to improvise or do something entirely different. In a recent playtest we came up with an entire other line of thought than the adventure was written for. An impromptu SC was set up by our DM, tying in with the SC written in the adventure and we had a blast. That specific path now is part of the SC in the actual adventure and when I ran it last week my players had boatloads of fun, they even did far more checks than required for the SC but as long as the scenes progressed and the players had the feeling they were accomplishing something I just kept them going.

Last in 4th edition it is important to equip your character for a SC. You have feats a plenty and using 1 or 2 to get some extra skills or multiclass feat can really round out your skill selection. I always try to take at least one physical, one social and one knowledge skill making sure I can always try to engage in an SC.

I guess where I deviate from you Elder Basilisk is that I view all of D&D as flawed systems. The questions come up with any of the mechanics, from levels to hit points to classes and there aren't any perfect answers. I think the skill challenges are far from perfect, but they do a neat mechanical trick, which is to make the non-combat portion of the game matter to the players as part of the system. Does it do it perfectly as a pure role play mode or a perfect mechanistic system? Nope. It is one of the many things you have to get past to enjoy. Can't get past it? Cool, make suggestions, try other ideas, experiment. Tell others about your ideas, you may yet hit the perfect mechanic.

At present, my idea is to roll six to fourteen dice as quickly as possible and get on to the interesting and fun parts of the module. Or to ignore skill challenges entirely and let players roll individual skill checks for individual tasks as they come up and just give them all the xp and milestones for free--along with the logical consequences of their skill checks.

It is always possible to treat skill challenges in mods exactly the way most of us treated the weapon vs armor table from first edition: ignore them entirely and hope that they go away next edition. In most cases, I suspect that it is the best strategy for having fun.
At present, my idea is to roll six to fourteen dice as quickly as possible and get on to the interesting and fun parts of the module. Or to ignore skill challenges entirely and let players roll individual skill checks for individual tasks as they come up and just give them all the xp and milestones for free--along with the logical consequences of their skill checks.

It is always possible to treat skill challenges in mods exactly the way most of us treated the weapon vs armor table from first edition: ignore them entirely and hope that they go away next edition. In most cases, I suspect that it is the best strategy for having fun.

That may be your answer, but I suspect it might not be the answer all your players would prefer. That said, if you are so filled with bitterness over skill challenges, then your answer may be the best your players are going to get. If you can't run a fun skill challenge, then that's just the way it is. Like I say, I struggle with a number of mechanics, doesn't mean that mean I let the players get worry about my hang ups.
It is no surprise that what I described can be framed as a skill challenge. The basic scenario is ripped out of core 1-1. However, I have yet to see a skill challenge go down that way, and I don't expect that I ever will. Here is why:

Planning is about overcoming an obstacle as a team. ... .

Skill challenges, on the other hand, are about achieving X successes before Y failures. ...

Planning requires coming up with a plan that will work for everyone in your party and interacting with the world in a way that facilitates the plan. ....

Skill challenges, on the other hand require coming up with a rational excuse to use a skill that someone in your party is good at six or twelve times. ....

I agree with many of your points and this, to me, expresses the key failure of most skill challenges.

The level of abstraction is to complete and too removed from the adventure (and often, as written, the players goal is non-obvious).

I am starting to feel as if the real point of the skill challenge is not to "'have a non-combat encounter in which everyone participates" - you can already do that and it's called Role Playing. If it is logical for a skill to influence the outcome, the skill can influence the outcome whether it is wrapped in a skill challenge or not.

Rather, I suspect that the real purpose of a skill challenge is to codify and provide a framework for turning roleplaying into an encounter (and thus to allow it to be formally rewarded with experience).

To put this another way, it seems as if there are two different types of skill challenges. In the 'natural' skill challenge, it is relatively clear how the players actions lead to their eventual success - each step in the path moves them closer to their success. In the ideal case, the final (skill challenge winning) success coincides with players reaching their goal. In the 'abstract' skill challenge, the players look for various ways to use their skills but the connection between each of these success (or failures) is not obviously connected to the final success or failure of the skill challenge.

The former of these (the 'natural' skill challenge) do not need the skill challenge mechanism to work: They work just fine as roleplaying encounters without needing the success/failure formula overlay - the formalization into a skill challenge only serves to determine the experience earned. To relate this to the earlier posts - the players in this case have made and executed a plan and their success or failure is dependant upon the plan and how well they rolled and less on the skill challenge formula. These are also the ones that, in my experience, have players walking away saying "That was a great skill challenge, to bad more of them aren't like that".

The latter of these (the 'abstract' skill challenges) are the problem. Rather than working together to solve a specific problem, the players are trying to find ways to justify the use of their particular skills to address parts of the environment. The connection between the different players actions and the eventual success/failure of the skill challenge is non-obvious at best (and sometimes non-existant) and if you take away the skill challenge formula, you end up with a series of disconnected skill checks. And yes, this could be argued as disproving my statement that the skill challenge formula serves no purpose but to create an encounter/reward system but I don't see it that way - rather I see it as a litmus test. If the skill challenge doesn't hold together as a roleplaying encounter without the skill challenge overlay (i.e. if it is disconnected rolls rather than a plan coming together) it is flawed and perhaps shouldn't be used as a skill challenge in the first place.

In brief: 1) Role playing encounters should not be abstracted to the point where there is no connection between the results of the character's actions and their eventual success (and encounters that are that abstracted feel forced, flat and unsatisfying to most players). 2) If there is an apparent connection between the results of the character's actions and their eventual success, the Skill Challenge Overlay is not necessary for the determination of the eventual (in-game) outcome of the character's actions. 3) If the Skill Challenge Overlay is not necessary to determine the eventual outcome of the character's actions, it's only remaning role is to determine the rewards for that success (i.e. to provide a way to reward players for successfully solving the initial challenge).

and finally:


An alternate approach: Design the encounter as you would if it were a Skill Challenge, but present it as an obstacle/goal in the game world to the players and resolve their attempts to overcome the problem, basing their eventual success on whether or not they have overcome the challenge. And yes, in some cases, this may be a sequence of obstacles (as in a chase) or a single task (as in a diplomatic negotiation). Then, behind the scenes or in plain sight, track the number of successes and failures on the way to that success - all the while making it clear to the players that no number of failures or successes will lead to an endrun short circuit to a final success for the characters - only the character's actions can do that. (In other words, the skill challenge becomes a metagame concept used to determine the players' reward but the players' relative success at the skill challenge has no effect on the characters - their success or failure depends on the results of their actions not the abstract Skill Challenge).

And then take those successes and failures and use that to retroactively determine what complexity of skill challenge they succeeded at. If they only had 4 successes before hitting that third failure, they get experience as if it were a complexity one skill challenge while if they reached 10 successes before getting that third failure they might be rewarded as if it were a complexity 4 skill challenge (with some preset maximum complexity for those cases where they overcame the challenge without ever reaching those three failures).

The reduces the Skill Challenge Overlay into nothing more than a bookkeeping mechanism to be used to determine the reward for the Roleplaying encounter while doing away with the disconnect between the results of the players actiuons and their eventual success (or failure).

Carl
Personally I say a good rule of thumb is that you go around the table asking how people will help the situation. No one is allowed to pass twice in a row and no one is allowed to assist without explaining how they want to contribute. I think the majority of bonuses should come from good roleplaying and creativity.

I don't think dice rolling for skill checks should ever take the place of creativity and speaking in character.
Personally I say a good rule of thumb is that you go around the table asking how people will help the situation. No one is allowed to pass twice in a row and no one is allowed to assist without explaining how they want to contribute. I think the majority of bonuses should come from good roleplaying and creativity.

I don't think dice rolling for skill checks should ever take the place of creativity and speaking in character.

The problem with any of the "no-one is allowed to pass X times" and "Aid other is highly discouraged" ideas is that they are highly unfair to the PCs given the current mathematical structure of skill challenges. If you have to get 12 successes before 3 failures, then even with an 80% chance of success on every roll you make (which is pretty darn good for skill checks that are not using Aid Other), you will only succeed a little more than 50% of the time. Now, let's imagine that, due to the DM's perverse desire to make sure that everyone participates--even those who know perfectly well that their participation in any non-Aid Other fashion is only like to result in failure for the group--you add a few rolls with a 40% success rate. Now your odds of success are significantly below 50%.

The current rules explicitly allow Aid Other and explicitly allow passing because the probabilities of the skill challenge mechanism do not work out without those kludges. In the RPGA, the various bonuses from reward cards and especially the "That'll Do" card disguise the mathematical weakness of the approach you advocate, but try it on a high complexity skill challenge with people who don't have cards and it'll fall apart faster than wonderbread in water.
That may be your answer, but I suspect it might not be the answer all your players would prefer. That said, if you are so filled with bitterness over skill challenges, then your answer may be the best your players are going to get. If you can't run a fun skill challenge, then that's just the way it is. Like I say, I struggle with a number of mechanics, doesn't mean that mean I let the players get worry about my hang ups.

That's the answer that I prefer as a player. In my local area, a lot of judges (fortunately) don't bother with skill challenges--just give people the xp and roleplay through the situation. Unfortunately, some DMs still insist on running through skill challenges so I roll with it as best I can and try to avoid rolling my eyes too much.

I haven't DMed a whole lot but the last time I did, I ran the skill challenge in stealth mode--having the PCs react to the situation, recording successes and failures, and telling them after the fact. I am a bit concerned about adopting this approach in the future, however, because while skill challenges in general are hard to fail if you adopt the appropriate strategies (never roll anything that you are not trained in--and if you can help it, never roll anything that does not also use one of your strong stats as well; otherwise Aid Other; and make sure to get some Aid Others regardless), they are very easy to fail if you just have one or two players rolling checks with skills that they are not good at. If I run the skill challenge in stealth mode, it is likely that the players will not adopt appropriate strategies and statistically, they will fail far more often than they ought to. So, if I run in stealth mode, the players are likely to have fun, but I can't justly give them a failure unless they completely botch it. On the other hand, I have yet to meet a player who enjoys playing skill challenges out of the book. At best, people gamely give it a try.
So, if I run in stealth mode, the players are likely to have fun, but I can't justly give them a failure unless they completely botch it. On the other hand, I have yet to meet a player who enjoys playing skill challenges out of the book. At best, people gamely give it a try.

I think what you call stealth mode is what the designers typically think of how you're supposed to run skill challenges. When they succeed, you give positive feedback, when they start to fail, you give negative.

Perhaps if you had the players write on a notecard what their 4 highest skills are or skills which are greater than (level/2)+6, then when people are roleplaying, you ask the people with the skills to make the rolls.
Perhaps if you had the players write on a notecard what their 4 highest skills are or skills which are greater than (level/2)+6, then when people are roleplaying, you ask the people with the skills to make the rolls.

To what end?

And I mean by that, not 'how can you make the best of being forced into dealing with skill challenges' but rather 'do skill challenges help the game'?

If the goal is to get by the guards and climb over the wall for the party, how should this run in a purely roleplayed sense? Let's go through this and see if it would fit in with the skill challenge framework (not, mind you, if we could make the framework fit or vice versa).

Also a question regarding how you run these challenges: if the role-played goals for the challenge were met already but the number of successes had yet to be reached, what do you do? Likewise if the result of a failed challenge is, say dramatic, how do you handle the final failure not really supporting that?

And lastly, what about skill challenges is helpful beyond a bare sense of things? How does tracking numbers of 'successes' vs 'failures' help here?

-James
To what end?

And I mean by that, not 'how can you make the best of being forced into dealing with skill challenges' but rather 'do skill challenges help the game'?

I'm not sure what the point of the question is - skill challenges are part of the game and will continue to be part of LFR. There are lots of different ways to run them - some are fun, some are not.

Elder Basilisk has a way he likes to run them that the players enjoy, but he feels as if he has to dump the rules to do it. But he doesn't. You can roleplay your actions and have the best players roll their skill checks - no matter how well you might roleplaying ideas about how to overcome the skill challenge, maybe you still fail anyway. Same thing happens all the time in combat - you do the right tactics, but the monsters get the combo of luck and good tactics and overwhelm you.
The problem with any of the "no-one is allowed to pass X times" and "Aid other is highly discouraged" ideas is that they are highly unfair to the PCs given the current mathematical structure of skill challenges. If you have to get 12 successes before 3 failures

well actualy, you'll almost always see 12 sucesses before 6 failures, which means the lucky number players need to roll is 8s. So if the skill checks are about DC 15 (for first level), and the player is trained with an average stat of 14, then they will get a +7 and need an 8 or better.

If the party needs to roll 8s for a skill challenge, then they will have about a 50/50 of succeeding the challenge. The party can "even their odds" by either being creative, or by using skills above +7 (which every party has).

By the "rules as written" you can have everyone pass while the one paladin with +11 diplomacy makes all the checks and the rest of the party roll to assist.
well actualy, you'll almost always see 12 sucesses before 6 failures, which means the lucky number players need to roll is 8s. So if the skill checks are about DC 15 (for first level), and the player is trained with an average stat of 14, then they will get a +7 and need an 8 or better.

If the party needs to roll 8s for a skill challenge, then they will have about a 50/50 of succeeding the challenge. The party can "even their odds" by either being creative, or by using skills above +7 (which every party has).

By the "rules as written" you can have everyone pass while the one paladin with +11 diplomacy makes all the checks and the rest of the party roll to assist.

Errata makes this 3.
Well, the point of this question is whether or not skill challenges are a success. If they are, then we can run, dance for joy and use them prodigiously. If they are not, then, as writers or as judges, we should stop using them. They can keep skill challenges in the DMG, but they aren't making us put them in modules. They can put skill challenges in modules, but they can't make us run them. (And in fact, DM Empowerment could be seen as permission to throw them under the bus and back up a few times--I generally have not done this mostly because I am not sold on DM Empowerment to the extent that some others are).

My general feeling (if it wasn't clear already) is that, in their original form and as currently errated, they are a complete and utter failure (that cannot and should not be redeemed). I have a few ways of dealing with them and I am not happy with any of them. The mechanic rewards and supports tactics that are 180 degrees from the role-playing situations that they are meant to model and actively punishes the tactics that the situations would normally call for in the absence of the skill challenge mechanic. I only know of one location in my neck of the woods where DMs will typically run them as written. (And since I DM there on occasion, it's not exactly unanimous). At the other two major LFR play hubs in the area, it is typical to see them ignored entirely, run in stealth mode, or run with major house rules that dramatically change the math of the challenge.

Now, they are a part of 4th edition, yes. On the other hand, that does not mean that they will necessarily continue to be a major part of LFR. I hope that as more people get over the honeymoon period of 4e and have more experience with skill challenges, their utter failure will be recognized and module writers will quietly stop putting them in modules--or will start modifying the mechanic and come up with encounters that are labeled skill challenges but which no longer reflect the basic 4e skill challenge mechanics. (The maze encounter in the recent LFR core special is an example of this. It is treated as a skill challenge for the purpose of XP, but it deviates significantly from the standard skill challenge mechanic by allowing an unlimited number of failures with set consequences for each failure). And if word of this filters up to the top, we may mercifully see skill challenges removed from 5th edition or heavily modified and relegated to a sidebar like Weapon vs. Armor type was in 2nd edition. Bad ideas do not have to be eternal.

I'm not sure what the point of the question is - skill challenges are part of the game and will continue to be part of LFR. There are lots of different ways to run them - some are fun, some are not.

Elder Basilisk has a way he likes to run them that the players enjoy, but he feels as if he has to dump the rules to do it. But he doesn't. You can roleplay your actions and have the best players roll their skill checks - no matter how well you might roleplaying ideas about how to overcome the skill challenge, maybe you still fail anyway. Same thing happens all the time in combat - you do the right tactics, but the monsters get the combo of luck and good tactics and overwhelm you.

well on the upside the latest podcast said they would have advanced skill challenge rules in the DMG2 *crosses fingers that they make sense*
I think the biggest reason players and DMs don't like skill challenges is because the DM is treating it as a scripted, unalive, no options but what it is written, challenge. Also some players treat it as something they are more likely to fail at, and missing the rolls are more drastic to them.

I've DMed WATE 1-1 several times and the first half to 2/3rds of the session is one massive skill challenge. Every time though it has been extremely fun, lots of laughter, and not seeming contrived or limited. All of the skill rolls fall in the normal roll playing of the players asking questions and participating in various activity's self evident in the area, as a DM I rarely say no to what they want to do or try, so if their skill isn't on the list, but they role play it out I'll still have them roll the appropriate skill for the consequence and I use the DM golden rule and give them a circumstantial bonus on a future roll or more depending on what they do (+/-2). I only tell them at the very start they are going to be in a skill challenge, and if their character try to do something the success or failure will count so make sure your character does it well.. or maybe things wont turn out all roses. Often,a player will want to role play their character in a way they aren't good at, but that is where the fun comes into play... the failures spur other role playing opportunities and laughter around the table.. and the option to do something they aren't good at alows them to develop the personality of their character.. so what, some really uncouth people in real life think they can charm anyone, people in Faerun are no different, not even heros and boy is it fun to see the players trying to deal with the ugly dragonborn who wants to be everyone's friend. Some of the strict strategy players get mad if the other players try to do something they are bad at, but the skill challenges aren't for the straight combat nonroleplayers.. the skill challenges are meant to give the role players a turn to shine. Also.. don't be afraid of modify the skill challenge if the players are taking it a direction that isn't written down..work with them, feed them information based on the context of what they are trying, and just keep track of the successes and failures with key skill related to what they are doing. To resolve the issue of players being weary of rolling and failing, I like for some diplomacy and bluff skill rolls to ask the player their bonus and roll for them behind the screen. Then they don't know exactly what kind of response they are getting, it is less stressful then knowing it is a definite failure, and it is much more tense (and fun for most) to have some of it unknown if the hatchet is about to drop.

If you find your players aren't into role playing, make sure you have an equivalent EXP combat encounter on the side and prepare a hook to start that to move the adventure along and skip the skill challenge... just make sure whatever you are doing is fun for the players! It is OK to switch gears a little if your table doesn't enjoy a skill challenge or combat.. Sometimes I find that everyone likes role playing and I'll have the combats turn into a skill challenge with the banter of the monsters vs. pc's in the fight..or in a skill challenge if most people seem bored I'll throw in an equiv exp level fight that is triggered by an ambush, or some particularly awful failure or idea by the restless adventurers and work in the info to move them on in some fashion.

I know this was long winded, but hopefully my experience running successful and fun skill challenges and what I've learned will help others learn how to play captivating, natural, and fun skill challenges. Just remember the mechanics behind the skill challenge are only for a way to measure success and the consequences of the actions of the PCs in the world. The mechanics here should never be part of the foreground, the players and their creativity should drive the skill challenge with gentle shaping by the DM along with interesting narration and responses from the world.
If you are a DM running WATE1-1 and have a table with several replayers, I strongly recommend switching around some names and who provides what information. While easier to do thoroughly if you know this situation is coming, you can do it ad hoc with a marker. The "correct" pawnshop is now a different shop name and shopkeeper name. Use it as a framework or template and not a rigid script.

Please do welcome creative proposals for alternative skills. Perhaps use endurance for a drinking contest or acrobatics for a dancing contest. Introduce a theme tavern or shop that might make arcana, religion, history or nature more reasonable choices, or just some scholarly type in the bar, but able to answer questions for that smart person.

If you have a table of players who strongly prefer combat, simplify the skill challenge to "finding a gang who knows something" and going into an alley, and fighting the gang, preferably unarmed to make the contest different. If the PCs win, they get the information. It is not as realistic but as this is a game, doing this is an option. Draw on stat blocks from elsewhere in the adventure and just make sure it is a similar challenge level to award the xp. (Doing this well is a little tricky so not recommended for a beginning DM.)

Keith
Keith Hoffman LFR Writing Director for Waterdeep
Keith, THANK YOU for the validation that you are allowed to mix things up in a module. That has been the biggest argument from other DMs(that you cant change the module) against my LFR suggestions to make things flexible and more alive for the players in the past.

Also, thank you so much for a wonderful role playing LFR module, it has been my favorite so far and I hope more of this sort come out! Please give my thanks to Claire as well. It is very difficult to fully role play out this skill challenge and have time for the combats at the end...so I usually have to be extremely creative at that point, but the encounter setting and creature abilities, especially the last ones, provide excellent flexibility. My favorite combat moment was:
My favorite combat moment in Wate 1-1:
In the inn where a barbarian warforge named "War" took two OA's from the drakes and did a long jump out of the second story inn window to follow the gnomes. The gnomes failed a save check from laughter and failed their stealth because they were laughing too hard... and then the poor warforge was almost dead from the jump and then failed a couple of attempts to climb the rope.

I don't want to high jack the thread completely for wate 1-1, so feel free to respond off thread or we could start another, but..
The biggest contrived moment, suggestion welcome:

When a group fell for the gnomes trick upstairs and ran out to try to catch them. The module reads that the gnome escapes to move the encounter to the street with the fallen statue, but the players had a hard time believing that they could of escaped, with the wagon out of site before they were able to run back outside through the back. I made the back door locked so they burnned time going around front or getting through the door, but at a full run it still seemed implausible the gnomes were that much more quick in getting the wagon away. I made it so that the gnomes were basically hauling butt and left an easy trail of distraught people and knocked over items from their reckless run away.

If you have any suggestions on ideal ways to cut the last two encounters in the modules short that fit well with the intention please let me know! Again, thank you.
I've played and run skill challenges that were fun and engaging, and ones that were boring and borderline painful.

The main difference seems to be the DM's attitude and flexibility. Does the DM groan and sigh when he announces that you are about to have a skill challenge? Does he allow only skills listed? Disallow any bonuses to rolls for role playing or creative (but sensible) ideas? Then it's probably going to be a bad skill challenge.

Other people have given some good suggestions, like giving a +2 for role-playing or successfully using a skill that doesn't really apply.

Adjudicating non-combat encounters has always been difficult in D&D. Skill challenges are not perfect, but in my experience they can be a fun addition to the game with the right approach and attitude.

Allen.
If the players are more interested in story or RP than the combat, or the combat has turned into senseless reduction of hps, the DM should feel empowered to call an end to the combat.

Show
In the case of adventures in Waterdeep, unless the villains are assassins sent to kill the PCs and/or other murder, most thieves, drunks, or random crooks will not fight to the death. They know the penalties for murder are a lot worse than for thievery in Waterdeep. In the case of the thieves in WATE1-1, once the wererat is down and the battle has turned, feel free to have the thieves surrender. In the case of the gnomes, they really want to get away rather than fight, but if they have tried a couple of tricks or so, i.e., you have milked the encounter for that aspect, don't feel you have to play out the entire chase scene across the City. Neither group are willing to die to avoid going to prison.

Note: that may not be true in more lawless areas of Waterdeep, such as Downshadow. Or with undead, who don't necessarily pay attention to the law anyway.


Keith
Keith Hoffman LFR Writing Director for Waterdeep
I've played through a lot of LFR skill challenges, and overall I've liked about half of them. In general, I'd say that a good DM (and group) can rescue a poorly-written SC, while poor ones can doom even the best-written ones to mediocrity - but when you get a good DM running a well-structured SC, with players who are prepared to treat it as a role-playing encounter rather than a set of dice rolls, the result is just really great fun.

My best advice for DMs and writers alike would be to structure the skill challenge purely as a roleplaying challenge, and let the players take the lead primarily by working out what they want to do, rather than what skill they want to use. They know their characters, so they'll take paths that work to their strengths in any case. When the encounter calls for a skill check, ask for one.

I ran SPEC1-1 this weekend, and the first skill challenge there was one of the best we've had - one of the best role-playing scenes we've had, even. The way it was written up allowed me to present it in a way that had them clamouring to take part, and the skill checks flowed naturally from there - greatly helped by the number and variety of well-described checks and results in the scenario.

One pitfall I've seen is that, when a player suggests a course of action, the DM will tend to assume their character is trying to carry it out. Don't be shy to encourage players to throw out suggestions for other players' PCs to run with.
I think one of the ongoing problems is writers turning basic interaction into skill challenges. This creates two distinct problems. First, many of these are straightforward problems with relatively obvious solutions. So, after the players have done the three checks that should solve the stated problem, the DM left with an encounter that says the players need 6 or 8 to move on.

Second, anything that is an official skill challenge gets an experience budget. So something that should have been 10 minutes of roleplay becomes an exercise in the DM thinking up excuses to why the players aren't done yet and then makes the rest of the adventure weaker by syphoning off the XP budget.

So, in short the secret of skill challenges tends to be making sure they should be skill challenges in the first place. I tend to run my skill challenges transparently, that is to say that the players don't know if it is a 'skill challenge' or if that are simply doing a bunch for things that need skill checks. If the players can 'solve' the problem in a few checks, I tend to ignore the official success number I should be waiting for.
This little signature is my official and insignificant protest to the (not so new now) community redesign. The layout is lousy. The colour scheme burns the eyes. The wiki is a crippled monstrosity. So many posters have abandoned this site that some major forums are going days without posts. The 4e General Discussion board regularly has posts on the front page from two or even three days ago. This is pathetic. Since I have to assume Wizards has a vested interest in an active community I wish someone in charge would fix this mess.
I have found that the most important things in a skill challenge are:

1. Conveying to the Players WHAT The goal of the skill challenge is. If you say to the PCs:

"Ok you enter a labyrinth, this is a skill challenge, the primary skills are:"
That utterly fails. I have seen several DMs do this.

Instead say:

"Travelling up ahead, you enter a long tunnel, at the end of it are three possible exits. After checking each one one out, they also have three possible exits each. It appears that the way ahead is not clear. However, it's obvious that the bandits came through this area, what do you do?"

You have conveyed the same information here, but in the second case, PCs will give their responses in character, and will find something within their PC's abilities to do, and that gives you, the DM, the opportunity to decide if it works or not.

That brings me to my second point:

2. Ignore the bit in the mod that says what skills are primary and what is not, also ignore the banning of skills. That is a guideline.

Many skills like intimidate (the main fighter interpersonal skill) etc, are perfectly viable if used wisely and intelligently. Just because threatening a statue guardian with destruction might not work, doesn't mean you can't appeal to it's sense of preservation regarding the TOMB, etc that could be threatened if a fight was initiated.

3. Mod writers, remember that skill challenges should NOT replace simple checks.

As Telvin3d said: this is NOT on. Need to convince a lone peasant to help? part of a greater skill challenge definitely, one all by itself, hell NO. There are a few out there where a simple diplomacy check/thievery will do. If as a DM you are confronted with this, try and use the old art of a slot zero or good prepping to change it into a much larger and more realistic challenge - such as convincing a GROUP of peasants/refugees/merchants, with several different 'scenes' and characters involved, such as the dodgy fence, the greedy old codger, and the loyal but slow farmer.

4. Remember to encourage participation, Don't Let the ones with the highest mods do everything.

This is the MOST important point. Don't let it turn into a discussion of modifiers and the mechanic, by assigning a goal, and leaving it to the players to work out how to achieve said goal, the fighter will say stuff like "I throw down boxes and stuff to slow down the mob chasing us" (rolls athletics) whilst the Barbarian Intimidates by "growling and sneering back".

What is important to remember is that it is a team role-playing game, as soon as it becomes a battle for who has the right skill or is the most awesome at it, the game stops being fun and the skill challenge stops being role-play. By all means encourage people to play to their strengths, but if one PC has a +11 diplo and one has a +13, doesn't mean PC 1 has to be quiet/just assist/not come up with thier own ideas.

p.s: important thread Andy.
4. Remember to encourage participation, Don't Let the ones with the highest mods do everything.

This is the MOST important point. Don't let it turn into a discussion of modifiers and the mechanic, by assigning a goal, and leaving it to the players to work out how to achieve said goal, the fighter will say stuff like "I throw down boxes and stuff to slow down the mob chasing us" (rolls athletics) whilst the Barbarian Intimidates by "growling and sneering back".

What is important to remember is that it is a team role-playing game, as soon as it becomes a battle for who has the right skill or is the most awesome at it, the game stops being fun and the skill challenge stops being role-play. By all means encourage people to play to their strengths, but if one PC has a +11 diplo and one has a +13, doesn't mean PC 1 has to be quiet/just assist/not come up with thier own ideas.

Sure, but especially with high complexity skill challenges, you suffer a non-trivial chance of complete failure if you do anything other than "find the highest mod, get as many assists, do it until you are done." The math of skill challenges is extremely unforgiving and you get extremely dramatic swings in the success failure ratio by apparently minor changes in the success chance for the individual skill checks. As a player, it is generally my goal to get the guy with the highest mod to roll the check and I'd prefer not to be pushed into 30-50% failure range by DMs who think that everyone ought to participate. (On a complexity 12 skill challenge, it looks like even if every skill check your party rolls has a 75% chance of success, it looks like you have more than a 50% chance of failing the skill challenge--add a couple of 65%s or a 50% in there and you're probably toast).

I can understand where you are coming from, but from the player's side of the table, it is a strategy for losing.
With the errata to skill challenges, the prevalence of the 'That'll Do' card, and so on, it's not hard to pass one. What's harder is ensuring a skill challenge isn't a staid bit of dice-rolling that prohibits the role-play it supposedly encourages. What is important is to be mindful of both extremes, don't force the fighter to roll the diplomacy, but at the same time, don't just sit there and let the Waterduvian half-elf with Astral Speech roll Diplomacy 8 times.

Each PC is likely to be good at SOMETHING, so encourage their participation. Again, you are assuming a presented situation. Consider the following:

"A couple of Bandit Guards stand by the doorway where you have to enter. If you fight them, it will create a ruckus that will surely alert your quarry, giving them time to escape."

Bad:
DM: "Ok There are some Guards minding the door, you have to talk/get your way pass."
Players:"Bruce activates astral speech, has a +19 diplomacy"
*Dice clatter 4 times*

Good:
Warlock: "I use bluff to convince the Guards that the militia are coming, known bandits should scarper or they'll be nicked!"
DM: The Guards look concerned and and one of them goes to look round the corner
Wizard: "I use Arcana and Ghost Sound to make it sound like a group of soldiers are coming this way:
DM: "The Guards look really sheepish now
Warlord: "I use diplomacy to convince them that their pay aint worth getting arrested"
DM: "The guards look about ready to bolt for it, with the sounds of the militia becoming louder:
Fighter: "Using intimidate, I point out that they aren't really tough enough to fight off the militia"
DM: "On that note, the guards decide that their measly wage aint worth the hassle, and they scarper, leaving the doorway unguarded for your entry"

I have seen both types of skill challenges, the first is based on the premise of always going with the highest mod. It's painful to both play and run, when you consider the point of a role-playing game.