You either have A> 'you are in a skill challenge what do you do' or B> 'you have this map with writing on it that you need to puzzle out'.
Which do you prefer? Now explain how that would play out in 4E.
B, and that is exactly what I do when I run a game (I already agreed that in this regard the DMG fails). In the 4E games I run this usually means that the players start discussing with each other what possible methods are open to them to examin the map, listing a couple of skills they think are involved in deciphering maps, and suggesting visiting knowledgable sages and sailors who probably could help. They then would discuss who should make the checks, asking me whether they can make a group check for the knowledges (to simulate them discussing with each other). I even allow them to state that they are going to ignore mister dumb barbarian in advance, in which case I would not count his check regardless of success or failure.
Once they made the knowledge checks, they would ask me if they their characters feel like they have deciphered the map or whether there are still some things unknown (in other words: enough successes = you are finished, enough failures for failing challenge = you are finished, and you think you deciphered the map correctly).
Depending on the circumstances, I might keep the challenge abstract and in which case am not going to detail what exactly each success or failure provides. When I have a pretty good idea of what each step and skill entails, I might give a detailed description of what each check results in. In that case the PCs might gain a partial success, something that is indeed not detailed in the DMG (but it is not prohibited either). For example, I am not going to detail the map in CORE1-3 at a level where I can describe what each check represents (if only because then I need to think about things like tides (rather difficult with a moon unlike RL), exact timing, distance, speed with which the PCs travel, weather and countless of other factors that I know nothing about and neither will most players).
Take for example searching a room.
Wrong example, since searching a room should never be a skill challenge in itself. It could be part of a bigger challenge, but in that case the circumstances should determine how that skill is used. If the PCs have all the time of the world, and actually take their time, then I would not count finding the item as part of the challenge (and in the new template it would be noted down as a 0 success; 1 maximum skill use). In that case taking 20 with the +8 modifier would be fine. If on the other hand time is of the essence, e.g. they have to keep a low profile and the longer they sneak around the spot, the bigger the chance of being spotted (e.g. Spoiler:Show
DALE1-3
), then I would ask for roll by the PC with the highest Perception with the rest assisting (assuming they are not doing something else at the moment), or I would allow them to take 20, but instead ask them to make a Bluff or Stealth check.
It's like the repairing the ship skill challenge brought up here. How does electing to haggle before buying supplies help you repair the ship? If you are successful at haggling then you need to do less work on the ship! Worse if your party takes the time to first scout out and then break into a place to steal the materials you are better off than by purchasing the materials and taking that time to work on the ship!
Have you read the skill challenge? You gain auto successes if you buy the supplies. Haggling itself does not get a failure or success towards the challenge as a whole, it reduces the amount of gold you need to spend to gain that auto-success. Not spending the gold does not provide you with a failure. Of course, failing on the Streetwise to find good supplies does generate a failure, and considering there is a time constraint that is not too weird (it is up to the DM to let describe the failure as the PCs using a lot of time to find the right stuff or to roleplay the scene as the PCs coming across inferior material with the PCs have a chance to learn about it or not).
In any event, assuming you get into a situation like you describe, one where one route is notably more complex then another, then you as a DM should increase the complexity of the challenge to simulate the fact that the PCs are taking bigger risks. It is just the other way round of the situation you used to describe a situation were the PCs come up with an easier sollution then the authors thought off. The fact that it is very hard in LFR to do this correctly due to page count limitations, xp limitations and player unpredictability, should not count against the game mechanic in itself. Again, DME.
For example, in CORE1-8 I tried very hard to come up with different paths to reach the goal of the challenge and then to come up with enough challenges and difficulties to make each paths equally complex. Perhaps somewhat unfair to the players, but the LFR restrictions make it hard to do otherwise and the majority of players want to be challenged.
Have you read the skill challenge? You gain auto successes if you buy the supplies. Haggling itself does not get a failure or success towards the challenge as a whole, it reduces the amount of gold you need to spend to gain that auto-success.
I must have misread it, because Diplomacy is listed as a primary skill and the text for it reads as follows: Diplomacy (DC 15 / 16): The PC attempts to convince a supplier that he should provide necessary resources to Captain Kloom by drawing on the PC’s own credibility or resources. This could result in getting a discount on the price of lumber or nails, or perhaps even getting something for free if the PC is sufficiently persuasive.
Which mentions nothing in regards to not counting a success or failure here.
The other skills that don't count as successes mention it expressly, am I not understanding how this works?
Madfox11 wrote:
In any event, assuming you get into a situation like you describe, one where one route is notably more complex then another, then you as a DM should increase the complexity of the challenge to simulate the fact that the PCs are taking bigger risks. It is just the other way round of the situation you used to describe a situation were the PCs come up with an easier sollution then the authors thought off. The fact that it is very hard in LFR to do this correctly due to page count limitations, xp limitations and player unpredictability, should not count against the game mechanic in itself. Again, DME.
Why not remove all the counting of successes/failures and just handle it without the skill challenge mechanic via DME?
As I read that part of the challenge you need 8 successes: 3 must come from securing materials, 2 must come from direct repairs, and the last 3 can come from either. But coming from either doesn't make sense in comparison.
Am I misunderstanding either you or this challenge? Or both?
-James PS: will comment more on the rest later when I have time.
But coming from either doesn't make sense in comparison.
Hmm... why? You're telling a story of how the group got it fixed not simulating the intricacies of ship building. You don't need to say at the outset 'The ship is 37% destroyed and will require precisely 2 checks of this and 3 checks of that to fix' - it could instead be 'Well, the ship is almost there but you're going to need to either juryrig a fix together out of the bits of wood you've got left or manage to locate a new ...' - or preferably just RP out both scenes with folks and stop when you're about right for whatever they happened to do. They did 4 to check on supplies and 4 for repairs, on their own? Great, moving on.
I appear to be unable to communicate what I mean to you in this regard, or alternatively you are obstinantly refusing to accept what I am saying.
If the DM sits there and says "This is a skill challenge roll Arcana, History, Diplomacy, or Religion." That does all the things you are worried about.
If the DM sits there and says "You are confronted with an ancient tower, mysterious glyphs decorate it, and an Ghale Knight stands by the base of the tower and no door is apparent. How will you use your skills to surmount the challenge of gaining entry, and the chance to find what you seek?" That tends to avoid the things you are worried about, as the players are getting the narrative part, and the knowledge they are in a skill challenge.
It does so, but the invisible skill challenge carries a number of gameplay costs which vary from minor to dramatic.
The first problem with the invisible skill challenge can be summed up in an experience I had last week. We're playing a mod where we need to track down an old man who kidnapped a baby. The railroad dropped us off at a dirty bar inhabited by a bunch of hard-eyed, close-mouthed xenophobic hick miners. The DM tells us that as we enter they are giving us the hostile stare. Well, my character happened to be pretty good at streetwise. So, I say, "OK, I've seen tougher crowds at home in Westgate, I'll be fine. I roll streetwise and hit DC 30."
The DM says, "These are tough guys, they seem very unfriendly and don't tell you anything."
I am a little annoyed that using the appropriate skill with a good roll gets me absolutely nothing, not even a new bit of description, but I roll with the punches. "Fine, if they're tough, apparently, we need to show some spine and get some respect. Intimidate" (which is, of course, taking a chance since intimidate is an autofailure in 20-40% of skill challenges). I roll well again and land a 27.
The DM says, "These are tough guys. They still seem unfriendly and don't tell you anything."
At that point, I mentally checked out of the skill challenge entirely. Trying to play "guess the key skill" is a tedious and stupid game in an already tedious and bad mechanic, but (I am guessing) that the DM was refusing to hand out successes because the mod said that streetwise and intimidate were not key skills or that the only way to win their respect was to talk to the barkeep and drink his liquor that takes something higher than a DC 20 endurance check to choke down. (As a side note, where do they come up with this stuff. It's harder to gag the liquor down than it is to go without food for a week without suffering ill effects or to tread water for four hours? I think that kind of thing falls under the category of poison, not liquor).
To generalize from that experience, immediately you start running an invisible skill challenge, you have to toss the whole key skills mechanic out of the window. If you don't, it turns into "guess the key skill" as players wonder why the things that would ordinarily work don't.
Secondly, the invisible skill challenge can be unfair to players because their skills all start working differently and you never told them. Let's say your level 10 party has a pit that they have to get across. Outside of a skill challenge, they ask how far it is across, see that it is ten feet across and know that even the wizard can probably make that with ease. In a skill challenge, only characters who are trained in athletics and have strength as a primary stat should even make the attempt. If you describe "a narrow chasm" and it turns out to be DC 23 to jump across whether or not you have a running start, that is not very fair to the characters who would otherwise have acted rationally if they had not been mislead about the situation.
This works both ways. I played another mod last month and, in one of the skill challenges, the DM encouraged us to try to use arcana to disrupt the ritual the bad guys were doing. He seemed to think it was obvious. It wasn't because that is something that you can only do in a skill challenge. Outside of that context, you can't use arcana to disrupt a ritual. You can't use arcana to dispel an illusionary wall. So, unless you are told that you are in a skill challenge, you won't know that your character's abilities have entered skill challenge land.
This leads to a final problem with the invisible skill challenge: the DM/player relationship. The DM/player relationship depends upon trust. You have to trust that the DM is adjudicating the rules fairly and that he is not actively hiding information from you. To use an example from a particularly eggregious LG mod, "the hydras are hiding under the sand; PCs only get a spot check if they specifically say that they are looking for monsters hiding under the sand. Otherwise, the hydras get a surprise round." That is (at least I hope) widely recognized as an example of bad module writing and anyone who ran it as written would be an example of bad DMing. (Especially in LFR--in LG, there were some people who would have claimed that a good DM still didn't have flexibility to fix a bad module; in LFR, that excuse has vanished in a cloud of DME). However, whenever I have run invisible skill challenges, I have felt like I was doing exactly that. I was changing the rules on the players without telling them and without affording them the opportunity to start saying, "I look for an ambush under the sand, I look for an ambush from behind the hills, I look for monsters diving out of the sun, can I get my fracking spot check please?"
Finally, the invisible skill challenge only works better than a standard skill challenge because the skill challenge rules are a complete and utter failure. Consider this. The reason why you might want to run the skill challenge invisibly is because as soon as players start to look at the optimal tactics to deal with the challenge in the context of its rules, they start behaving in ways that are at odds with the narrative context. "You should search alone, I wouldn't want to contribute a failure." A good set of mechanics would encourage behavior that was appropriate to the narrative context. That you may obtain more in-character results from an invisible skill challenge only means that the rules are bad.
Contrast this with D&D combat. A paladin is supposed to seek out and challenge his foe and the gods strike that foe down if he refuses the challenge. (At least in theory, by the dominant interpretation of the fluff). And the rules give incentives for exactly that. The divine challenge requires that the paladin actively engage his foe and not run away from him. And they punish the foe if he attacks someone other than the paladin. A rageblood barbarian is supposed to charge and hit things with a big weapon and then dash off like a raging madman to hit something else as he drops the first one. And the rules encourage you to make your barbarian do exactly that. In combat, the rules generally encourage behavior that supports the narrative context. If combat rules encourage behavior that is at odds with the narrative context (as some thought that the playtest divine challenge rules did), those are recognized to be bad rules.
Maybe that's something LFR skill challenges could work better on for their introduction block text... moving the DMs toward the more immersive approach rather than the crunchy one.
But this still isn't a mechanical flaw with skill challenges, its a DMing issue.
Didn't stop us recently when a "yahoo" did exactly that, and we were not just rolling on 3+
Maybe and maybe not. I've had at least one table where a TPK was averted because the paladin rolled a crit with a x3 crit weapon while smiting evil at the last possible moment before I killed the entire party. (Don't ask me why he waited that long to use the smite. Weak parties playing up should use good tactics). But that does not for one moment change the fact that the party was in a very bad situation in that combat and that they were unlikely to come out of it alive.
Now, when your yahoo rolls his skill at 30% he changes your odds dramatically. Now, maybe you all roll skills where you succeed on a 5+ and you end up at something like a 50% success rate (that is not calculated--it turns out I didn't save the spreadsheet where I plugged the math in last time and I've got to go to work in the morning so I'm not recreating it right now). It's not unlikely that you will succeed anyway--especially if one or two players have not this time cards. (They make a very very dramatic difference in the math of a skill challenge--heck, even the +1 secondary effects make a big difference since they can turn failures into success a good proportion of the time). In any particular instance, it's quite possible that you will still succeed, but overall, the odds are not good.
Also, the math complex enough that most people do not understand the probabilities (the designers certainly didn't as was evidenced by the initial DC chart) so that when you do fail in that situation, people are more likely to blame it on luck than on tactics. "We rolled three dice under five in a row, what could we have done?" "Two ones in a row. You can't win if that happens." But the truth is that you can. Two ones won't fail a skill challenge (assuming that the ones fail--which they don't need to do all the time if you are using only skills that you are good at). Two ones and the yahoo who failed on a 9 because he tried something he wasn't good at will fail the skill challenge.
This is much like the D&D miniatures game in that the dice can and will frequently go against you, but there are a lot of things that you can do to stay in the game. If you roll all ones and your foe rolls all twenties, you will lose no matter what, but if you are careful to always flank whenever possible, to avoid giving him any good area attack opportunities, to always stop behind cover, never make a basic attack when you could charge (into a flank if possible), and move your minis so that the maximum number of them come into attack range at the same time so that you are able to concentrate your force more quickly than your opponent, you will often win against people who do not do that--even if your dice are rolling around the bottom 33% of the bell curve area and his are rolling in the 66% area of the curve. A lot of people never realize that and blame the dice every time they lose a minis match--because it is easy to see a one or a two and hard to see the three things you could have done differently that would have made the four sixes into hits instead of a misses.
The math is forgiving at the complexity that LFR skill challenges should be set at. Many LFR skill challenges are way more complex than they should be, most should be complexity 1 or 2 challenges, way to many are 3 or 4... gods help us that we see complexity 6...heck the DMG stops at complexity 5 (12 successes)...
Again poor use of the rules by authors and DMs doesn't make the rules themselves bad.
Apparently I have been misusing the terminology. I though that a complexity six skill challenge had six successes before four failures and complexity 12 would be 12 successes before 12 failures. (I'll admit that I've paid much more attention to the math and tactics than to the terminology).
That said, it is telling if LFR mods should not use more than complexity 1 or 2 skill challenges--because the effects that I am discussing become more and more pronounced the higher the complexity of the skill challenge. If we must have skill challenges in mods (and unless there is a WotC directive saying that we have to use them in order for LFR to continue, we should be able to just stop writing them and stop inserting them during the editing process), it would be much better to put a hard stop at complexity 2 skill challenges and tell writers to just use multiple complexity 1 or 2 skill challenges rather than writing any higher complexity ones.
Now, don't get me wrong, complexity 1 and 2 skill challenges are still bad and we would still be better off without them because, even if the incentives they create are not nearly as strong as the higher complexity challenges, they are still perverse incentives that encourage player behavior that is generally at odds with rather than in harmony with the narrative context.
And in your example, did the players whose characters were not good in those checks make the rolls? And if they did, had the failures a consequence or were they just ignored? Personally, my ability to immerse myself in the game is hurt when all the players start rolling the dice with no rhyme or reason (then again, in 3E nobody knew anything unless trained in that knowledge skill). At least in a skill challenge in 4E, players use the skills their characters are good at instead of randomly rolling the dice until one of them rolled the right number to go on to the next scene.
In my mods, it depended on the situation.
In Long Cold Watches of the Night, I was anticipating that nearly everyone would fail the skill checks. They were there in case someone was playing a character who was so good at X, Y, or Z, that he should have a chance to spot the skullduggery before it kicked him in the stomach. Other ones were there because I figured that anyone who did put ranks into profession: cook or something similar deserved a chance to guess at the nature of the mystery meat. In short, failure was the expectation. If you succeeded on the skill checks, the story changed.
In Signs and Portents, most of the skill checks were easy. You don't need to be a honey-tongued bard to convince the nightsoil collector to tell you what he saw; mostly you just need to think to ask around--"who comes by here, first thing in the morning?" etc. It was also aided by the fact that I deliberately placed a number of trails of clues that could lead to the bad guys independently. So, if you were only asking the kind of direct questions that got the "Didn't hear nothing; didn't see nothing" response and didn't have the sense motive to figure out that they were following the don't snitch code and didn't have the (marginal) diplomacy needed to talk them into talking to you, there were plenty of other avenues to pursue. In a lot of cases, I expected you to be able to bribe your way to success with relatively modest cost too. And there was always magic if you were really having no luck. Some good questions on a speak with dead, a well-timed charm or suggestion, or simply brute forcing the situation by scrying (assuming you know who you want to scry on) and teleporting (assuming you land on target, etc) could also give you a shortcut along any of the investigative routes or jump start one that was stalling. But in any case, the trick was supposed to be much more in asking the right questions than in rolling the right numbers.
And, unlike a high complexity skill challenge such as Bald 1-1 or Waterdeep 1-1, if you picked a good line of inquiry right away and were able to see the picture accurately with only a few pieces, you didn't have to sit through three more scenes in order to get the requisite number of successes. (This happened in both of those mods when I played them).
What is more, the DC is not entirely arbitrary as it was in 3E so the players actually know in advance the risks they take by getting involved in the challenge at hand.
Well, players know what rough level the DCs are going to be, provided they know whether or not they are in a skill challenge. In a skill challenge, everything scales so you should never try anything you're not good at unless there is no other choice and you are the best your group has. Outside of a skill challenge, you may well find things that are impractical even in skills that you are good at (leaping off a 40 foot mast will probably hurt you significantly in the heroic tier even if dex is a good stat and you're trained in acrobatics) or that are doable even though you are terrible (at 10th level, even an 8 strength wizard can easily jump over a 10 foot pit with a running start).
In 3e, most of the arbitrariness of skill checks came from exactly the kind of scaling that is employed in the 4e skill challenge mechanic. If you stuck to the DCs in the Player's Handbook, you could actually predict whether or not you could succeed fairly well.
In the end though, your example IS a skill challenge where different skills provide different clues and at some point the players have enough clues to get at the next encounter. All you have to do is assign a complexity based on the number of clues you think are needed to get at the next encounter and some thoughts on the consequences of failures* and you are set. I find it odd to blame the system for something the players do and the DM accepts. Granted, the DMG tells the DM to inform the players what the primary skills are and that is a bad idea, since it stiffles the creativity of the players. If a DM on the otherhand makes sure the players are aware of the goals of the skill challenge and then let them work towards it, then creativity is going to be rewarded. Remember, the DMG does tell the DM to make sure the players tell HOW they are going to use a particular skill.
* In regards to the consequences of failure, you are correct that at times it can lead to weird results if regardless of actions taken the consequences of failure are exact the same. Again, that is not the result of the skill challenge mechanic (nowhere does the rules say this),
They certainly imply that the relevant question is whether you succeed or fail at the skill challenge as a whole and that individual successes and failures only matter in that context.
My argument is that the success or failure of individual skill checks should often have more significance than simply a success or failure on the road to completing the skill challenge. To take an extreme example, let's imagine that a group decided to go off the rails, and flirt with the deep end of the alignment pool in the Cor 1-1 skill challenge: to sneak into the forbidden zone, they are going to start a fire in a populated city block and then sneak into the forbidden zone while everyone is distracted putting out the fire and saving the citizens from being burned to death. So, we'll imagine that there are a number of skill checks involved in setting a fire and it comes out to about the same complexity. Still, in this case, failure at whatever skill checks were necessary to avoid getting caught for starting the fire should have consequences far more dire than simply accruing one failure towards the consequences of failing the skill challenge. Outside of the skill challenge context, the rest of the successes would be irrelevant if it turned out that the roll to rig the fuse failed so that it burned out before catching the inn on fire. And, in a historically informed game, being observed starting the fire would turn the situation into a question of whether you are hanged for arson rather than wether or not you get into the forbidden zone. (In a light and fluffy version of the realms, maybe they just let you go and say "don't do it again" but in most ancient, medieval, and early modern societies, arson was more severely punished than murder. The way FR is usually depicted in its serious moments, it should be there too).
but partially because of inexperience and partially because of the limitations of writing a LFR adventure. The LFR rules are that a skill challenge cannot end the adventure since there is no time and space to devise countless alternatives nor can a DM drop in the next adventure on a fly to fill the remaining hours of the slot. Space considerations and not knowing who is going to play the adventure is also going to severely limit the ability of an author to detail more than one or two potential outcomes. It is one of the reasons why the RPGA came with DME: to give DMs more freedom to go along with the plans of the PCs.
I submit that if a mechanic does not work without extensive DME, it is a bad mechanic.
And that is bad how? For me it hurts my ability to immerse myself in the story if the dumb barbarian tries to outsmart the count in a discussion without a chance to make a fool out of himself by mixing up or not remembering specific facts at all (in other words failing at a History check*). For me it is important that there is a consequence to a dice roll, why make that roll otherwise?
It is bad that the math of skill challenges tends toward becoming autosuccess or autofailure because: A. Rolling dice and hoping you don't roll a 1 or 2 is boring. For players, you are not succeeding, you are avoiding contributing a failure. You are dodging bullets, not shooting them. For DMs, especially those who do not grasp this kind of probability, it is also tedious. I have had several DMs, say things like, "this is stupid! OK, everyone roll your DC 10 athletics or acrobatics check. I know you'll all succeed." They do not feel like there is a point to it unless there is a significant risk of failure on the PCs' part. On the other hand, if there were a significant risk of failure on the PCs' part, we would be doomed to fail the skill challenge as a whole two times in three. B. It discourages participation. In the name of role-playing the skill challenge, a lot of local DMs have a habit of asking you to roll skill check X if you were talking. So, unless you think you're likely to be good at the skill you expect to be called out, you'd better keep your mouth shut. C. It encourages the "let the bard with his +30 diplomacy do it all" mentality (that was how it went down the second time I played Drag 1-2; and don't ask me how he got his diplomacy that high) D. It forces players to try to shoehorn the skills that they are good at into the challenge rather than figure out alternate ways to get around a concrete obstacle. (For example, I gather there was a recent mod where everyone was supposed to roll an athletics check or something as a part of a skill challenge to climb a cliff. Our DM decided that it was a stupid skill challenge and just used the rules for climbing and the given height of the cliff. After the fighter and my warlord reached the top through a combination of climbing and fey step, we lowered ropes and hauled up the wizard and the warlock who would have been utterly useless at skill challenge scaled DC type climbing. Since we knew they couldn't make the check, we found a way to overcome the obstacle without them having to make the check. In a skill challenge, they would have tried to use Arcana or dungeoneering to climb the cliff in whatever BS way they thought of at the time and immersion would have been out of the window. (Given the particular table I was at, the wizard player would have tried to BS his way out of it and the warlock's player would probably have stared helplessly, but the net effect is the same).
* I know you are convinced the rules should be same, so that a failed check means knowing nothing as opposed to misremembering things, but in 4E game philosophy there are many parts were the specific trumps the generic especially to make the game easier to run and more fun. Outside of a skill challenge misremembering facts is complicated and it is by definition going to be arbitrary in consequence. During a skill challenge you hopefully know what happens with a failed challenge and let the failed check work towards that end.
You've got me mixed up with James on this point. In general, I do dislike the arbitrarily scaling DCs that skill challenges require. They essentially amount to saying that your characters don't ever get better at anything because the DCs go up just as quickly as your skills. (Not quite because of attribute increases, but close enough). They also preclude the possibility of character development. You will never come back to a skill challenge similar to one that was hard before and ace it because now you're higher level and more awesome. This is partially because the mechanics of skill challenges mitigate so heavily against any concrete imagination that you generally don't know exactly what you did, only that you got X successes or Y failures and moved on with the mod. But more importantly, in this context, it is because any similar skill challenges you encounter at higher level are likely to be just as hard for you as the first one was. The DCs scale and DMs and authors will never decide that "now you're 16th level, this investigation stuff is beyond you; no skill challenge is required."
However, in the kind of case you are discussing (which is, I think, different than the one James was discussing where you look at something and roll to see if you remember something), I think it's perfectly fine for a failure at history to be detrimental. That is because it is pushing the normal boundaries of the history skill. Ordinarily, history doesn't persuade people. However, in this context, you are rolling history rather than diplomacy because you want to persuade people on the basis of what you know. In other words, you are taking what you may or may not know and using it. That can certainly fail.
I never said the math is forgiving, although with the current DCs it actually is (I have never seen a skill challenge fail and I did see PCs untrained make checks very often).
I have seen skill challenges fail on several occasions. The last time I ran East 1-1, they failed the challenge (or would have if I hadn't stopped running skill challenges--as it happened, they failed enough skill checks that if I had been running a skill challenge, they would have failed it). Ditto for the second skill challenge in Dragon Coast 1-2. (My party failed that one too--we got to one more success required and our DM told us we couldn't win without at least one bluff success and no amount of creativity could change that. So, we found the guy with the 10 charisma and all tried to aid other... and promply failed three times in a row). This monday, my party failed the sneak up the hill skill challenge in Impiltur 1-4. Two ones and a three and it was over. (I should have tried using dungeoneering rather than figuring that +9 athletics was enough to forgoe the lame attempt at BSing the DM--or I should have packed a that'll do instead of figuring that a 16 strength 4th level melee ranger with a +1 weapon with +2 proficiency needs two daring exploits).
I don't know how you manage to never see a skill challenge fail though because the probabilities are clear and even with the revised DCs, the odds dictate that people rolling untrained skills will have a noticable failure rate on skill challenges unless: A. there is a lot of Aid Other going on, B. There is a fair amount of Not this Time and other reward card bonuses in use or C. The particular skill challenge doesn't have failure at three failures (like the Spec 1-1 maze or the spec 1-2 climb the chain skill challenges), or D. the DM just lets the PCs' off the hook. E. Maybe playing at low tier, skill challenge DCs are just that much easier, but I doubt it.
I am arguing that there should be a chance of failure, especially if those unskilled at something are getting involved through that untrained skill. If you make a skill challenge forgiving, being good at a skill becomes unimportant and you invalidate a particular aspect of the characters.
Skill challenges can maintain a challenge without being as punitive as they are at the moment. The essential problem is that they are set up to encourage everyone to take part and then punish players when they do. (While they did explicitly state that players don't have to roll if they don't want to in the errata to the DMG, a lot of DMs have trouble with that--heck, the designers even appear to have trouble with that or with players aiding other if the podcasts are any indication). Regardless of the erratta, the skill challenge changes the game from one where players employ definite abilities to reach concrete goals and turns it into a collaberative storytelling exercise. Even if there hadn't been the original "everyone must participate" rule in the DMG, collaberative story time is a format where everyone is supposed to participate. But you can't participate if you can't roll. And you shouldn't participate if you can't find a way to roll something that you're good at. So, people end up either spinning complete BS ("I pray to Amaunator to be merciful to us in this desert--I roll a religion check to help us endure the heat") or being the yahoo who screws things up for everyone else.
If you insist on keeping the collaberative storytelling element (which I actually think we would be better off jettisoning into the scrapbin of weapon vs. armor type and wandering harlot tables), then I think it would be much better if successes were tracked against time or against some other finite and definable resource rather than measured against failures. That way, you are trying to succeed rather than trying to avoid failure. If it is still possible to fail the skill challenge (as, for instance in Stalker0 of ENWorld's Obsidian version of the skill challenge system where you need to garner the requisite successes before 3 rounds go by rather than before three failures), then the failure to contribute a success is still a definite consequence. But the probability of success is much easier to understand that way. And it doesn't punish players for participation.
Apparently I have been misusing the terminology. I though that a complexity six skill challenge had six successes before four failures and complexity 12 would be 12 successes before 12 failures. (I'll admit that I've paid much more attention to the math and tactics than to the terminology).
Before we proceed with the discusion, please rereade the skill challenge rules as well as the official errata. It is odd that you are so venomously opposed to something, when you don't even know the rules that cover them. Not only is this statement completely incorrect, you also seem to have some odd ideas about the DCs.
Mind you, your DM in your example should also reread the rules, because even the DMG clearly mentions something about using unmentioned skills. I am sure, that DM would have similarly screwed up in your 3E example. Your examples about WATE1-1 (and probably BALD1-1) are also more about DMs screwing up (either that or players screwing up because I have had players being so stuck in weird dice roll mode that no matter what I tried to did not get into the problem solving mode).
As for the pit example, why is it that you keep up dragging examples that are simply wrong? Getting over the pit should NOT be a skill challenge in itself, unless of course, there are external factors raising the DC to the expected levels. In your case the sides of the pit should be muddy and slippery, or crumbling or whatever else the author can think of to explain the raise in DC.
As for why not let a DM decide on success without the skill challenge: in itself because it would become rather arbitrary. As of yet, I have not had players deviate more then 1 or 2 checks from the chosen complexity. Then again, during preperations I spend some time contemplating the goals of the skill challenge, and what checks and actions might be necessary to achieve them. I then add or remove obstacles to get at the correct complexity, knowing enough of the scene to do so on the fly while running the game. In itself this is no different from a fight where I look at the aimed challenge and decide upon tactics and terrain to make sure that aimed challenge is reached.
O, and to quote CORE1-3 skill challenge text: "Buying a commodity outright counts as one automatic success, but the cost must be paid for out of the PCs’ own pockets (it cannot be “drawn in advance” against the treasure parcels from this adventure)."
Hence, buying stuff is indeed considerably less risky in CORE1-3 then stealing items. The use of the ritual Make Whole also provides 2 automatic successes (or removes failures by removing the need for certain supplies). So PCs who spend money only have to succeed at 3 checks before failing at 3 (remember: this challenge needs to be changed because of the errata). Checks that could easily be represented by finding and recognizing superior materials.
It does so, but the invisible skill challenge carries a number of gameplay costs which vary from minor to dramatic.
I didn't say make it invisible... though I tend to run them invisibly in what I'm saying but place a 10 sided dice on the table in clear view of the players moving it up as they generate successes and add another in view when they get a failure.
So the fact they are in a skill challenge remains apparent, but its secondary to the RP that is going on.
Elder_basilisk wrote:
That said, it is telling if LFR mods should not use more than complexity 1 or 2 skill challenges--because the effects that I am discussing become more and more pronounced the higher the complexity of the skill challenge.
That's not why they shouldn't use more than complexity 1 or 2 skill challenges, its because most uses of skill challenges in the format of a typical LFR module are not suitable for larger skill challenges.
I posted (here) an example of the opening sequence of Raider's of the Lost Ark comparing it to a skill challenge - Complexity 2 - 6 successes before 3 failures. That's a lot of stuff getting done for a maximum of 9 skill checks. I'm unable to remember an LFR skill challenge that needed more than those 6 successes where it also made narative sense to need more than 6 successes. But that also doesn't mean that such a skill challenge cannot happen. It just needs to make sense narratively.
If you think of a module as a movie and each "encounter" is a key scene of the movie - its much easier to understand where skill challenges fit. They are the parts of the action movie where the heroes are doing something exciting that isn't a fight. Just talking to 5 folks, not a skill challenge... racing against time to reach a destination, a skill challenge.
Sadly there are a number of LFR mods where they are "just talking to 5 folks" or where their time frames are wrong dramatically (LURU1-1 I'm looking at you), these poor uses of skill challenges (author) combined with DM's not understanding the rules for one reason or another lead to lots of bad play experiences with them.
(Oh and all I think the "Update" of the DMG rules tells us is that WotC really didn't playtest those rules before releasing them into the wild.)
Actually, one of the more fun skill challenges in my home campaign was a long overland march through an enemy country (wilderness part) where the challenge as a whole covered several days. The track took 3 days. Each day of travel I asked for 1 Stealth and 1 Nature check (I had thought of a couple of replacement skills and assists). I would also have asked for a Stealth checks for each night, but the PCs had that covered with a ritual (effectively granting them 2 auto successes). I then added two specific encounters as seperate scenes, one a social interaction and one an arcane problem with each of the scenes being "mini"-challenges that added 1 success or failure towards the overall challenge. In short: a complexity 4 skill challenge.
Sure, I could have handwaved it, but the skill challenge mechanics provided me with a good structure and a non-arbitrary way to decide on xp and DCs. It also forced me to think about the structure of the encounter and the concequences of failure which certainly helped me run that encounter for my players. It also helped the players give the idea that they were sneaking through enemy territory, how risky that was and it gave them the feeling they had control over whether or not they would meet the enemy instead of it just being the result of a random encounter dice roll.
Side-note: Arcana actually allows PCs to disable magical traps and magic circles all over the official books, so even though it might not be officially mentioned in the PHB, most players I met have no problem remembering that the skill can be used as such.
Before we proceed with the discusion, please rereade the skill challenge rules as well as the official errata. It is odd that you are so venomously opposed to something, when you don't even know the rules that cover them. Not only is this statement completely incorrect, you also seem to have some odd ideas about the DCs.
I have re-read the rules and other than the error in the terminology, everything I said about skill challenges is still correct. (I did not grow to realize quite how unutterably bad the skill challenge mechanics were until I graphed the probabilities).
Now, as for the DCs, they are pretty much irrelevant to the arguments I posed. (And I'm not sure what "odd" ideas I have about them--I was as surprised as anyone else when a 20 failed the Incident at the Gorge of Gauros endurance check skill challenge). The ease of failing skill challenges (if someone in your party does something foolish) is only one problem. The other side of the problem is autosuccess. Those are systemic issues that are inherent to the probability distribution of the X successes before Y failures rules. Wherever the DCs are set, that problem will still be there.
But since you seem enamored of something about the DCs as though it would make skill challenges into a good mechanic, I'll address the errattaed DCs.
PCs have three basic skill levels. They are good at a skill. A PC is good at a skill that keys off of an attack or a secondary stat and that s/he is trained in. For example, fighters are typically good at athletics. Good skills range from +8 to +12 at first level and increase a bit faster than 1/2 levels from there. +8 or +9 is more typical than +12 which usually indicates a character with a 20 stat and a racial bonus or a racial and a background bonus. For example, an 18 Con dwarf fighter wearing scale armor comes up with +11 endurance. For this discussion, when I mention a character who is good at a skill, I will be using the equivalent of a +9 at first level. That's trained with an 18 in the relevant stat with no armor check or other penalties; while you can do better, it will only be in one or two skills which may or may not be applicable to the challenge at hand.
They are decent at a skill. This PC is trained in the skill but it does not key off of an attack stat or a secondary stat. A good example of this is a typical warlord with endurance trained. These skills will usually range from +4 to +7 at first level and increase at about 1/2 levels. For this discussion, I will be using a +6; the equivalent of a trained character with a 12 or 13 in the relevant stat and no penalties.
They are weak at a skill. This PC is not trained in the skill and it keys off a dump stat. These skills will usually range from -3 to +2 at first level. Your typical fighter is in this range with diplomacy and your typical wizard is weak in athletics. The weakest numbers would be for characters with armor check penalties. For example, an 10 strength half-elf charisma paladin wearing plate armor with a shield. For this discussion, I will be rating weak as +0. Characters are frequently a bit worse or a bit better, but there's not too much difference one way or another.
So, how does that relate to the DCs for skill challenges.
If we start with easy difficulty, characters who are good at or decent at a skill usually cannot fail and will succeed even on a 1. If every character in the party were weak in the required skills, they would still have a 68% chance or so to succeed on a complexity 2 (six successes) skill challenge.
Moving up to moderate difficulties, if everyone were weak in a skill, they would end up with around a 20% chance of succeeding on the skill challenge against those DCs. If everyone were good at the skills, they would run between a 68% and 80% chance to succeed on the skill challenge because of the variations in moderate skill. If everyone were good, they would run between 99% and 100% chance of success on the skill challenge. Now, on the face of it, that sounds pretty reasonable, but what happens if you have two party members who try skills they are weak in (+0), two who are moderate (+6) and thus figure it's worth trying, and two who are good (+9). Now the party is down to a 70% success rate. But, unless everyone succeeds, it's not over after those six tries. If someone who is weak in the skill tries again, they're down to around a 60% chance of succeeding at the skill challenge as a whole. If it's one of the people who automatically succeeds, they go up to 80%.
How about the hard difficulty? Now, the party with all weak skills has a 1% chance of success. The party with all moderate skills, has a 32% chance of success. And the party who never rolls any checks that don't fall into the good category has a 68% chance of success on the overall skill challenge. But how about that party that tries two of each? They have about a 14% chance of succeeding. Let's keep them a bit more under control though: Now we'll assume that they one yahoo who will give it a try even though he's weak in a skill, and two people will roll with skills that they are decent at; everyone else will be rolling the good skills. The party's odds of succeeding at the skill challenge? 39%. Clearly "sit down, shut up, and don't contribute a failure" is the most basic and essential tactic for hard DC skill challenges--because as soon as someone does that, you cannot recover your odds unless you can reach 100% success on individual rolls and just do that over and over.
So, what's the result of this analysis.
First, standard skill challenges using the easy DCs are nearly worthless. Unless the party is deliberately attempting to fail, they will succeed almost every time. In a modified format, they can avoid pointlessness, but are not much fun as skill challenges. For instance, in Aglarond 1-2 (as it was run for me both times I played the mod--I haven't read the module, so some of the details may be wrong; the example is more useful for the principles involved than as a critique of the specific skill challenge in Aglarond 1-2), every party member needs to make a check or lose a healing surge when the ship beaches in the storm. It uses the easy DC table. This is still an obstacle to the party, however, because: A. every player has to roll and players who are weak in all of the required skills cannot dodge it. B. each individual failure has a significant and lasting consequence--thus, even if the party never rolls three failures, the one or two characters who do fail suffer consequences which means it shouldn't be handwaved.
That said, it has always felt more like a surge tax on physically uninclined characters than a harrowing experience of escaping a shipwreck to me so despite the non-pointlessness of the skill challenge, I did not think much of it as an encounter. But if you were going to use an easy DC skill challenge, you should eliminate the ability of weak-skilled characters to dodge the bullet and make the consequences of each individual failure more significant than the collective consequences of success or failure in the skill challenge as a whole. Otherwise, you might as well just write, "X happens, you win" and skip the die rolling.
Moderate DC skill challenges are a different beast entirely. Complexity 1 and 2 skill challenges with moderate DCs are the only places that the math comes close to working. (And thus if you insist on keeping writing skill challenges, those are the ones that you should use). All parties need to do to obtain a good chance of success is to make sure they don't use weak skills or that, if they do, they have a few people who are good at the checks to counterbalance them. And in this particular case, the good people with their 100% success rate on individual checks will sometimes counteract the weak characters with the 55% success rate on individual checks.
But appearances can be deceiving here because moderate DC skill challenges can be turned into automatic successes with a modicum of planning on the players' parts. Just two Aid Others per skill check boosts the moderately skilled characters into 100% success rate territory as well.
Hard DC skill challenges are a dramatic change from either of the two previous ones. One yahoo really does sink the party's odds and the characters with good skills are not enough to make up for it. Aid Other can still boost the good skilled characters to autosuccess, but even using Aid Other, moderately skilled characters will struggle to get much better than even odds and weak skill characters should still just go home.
Now there are a few factors that complicate this.
1. Skill challenges do not come with labels that designate them as easy, moderate, or hard. This creates a number of complications for players: A. What is good strategy in an easy skill challenge and won't hurt you if you're rolling against moderate DCs is suicide in a hard skill challenge. Rolling that skill that you are just OK at is fine as long as the DC is moderate, but if it's a hard DC, you are hurting the party by attempting it. Since you don't generally know ahead of time how challenging the attempt is, this turns hard skill challenges into "Gotcha" encounters and forces players who want to approach skill challenges strategically to treat every skill challenge as though it were against hard DCs.
2. At least in LFR, skill challenges frequently have more than one difficulty setting. For instance, in Sense of Wonder, Insight used the easy DC table, bluff, intimidate, and thievery used the hard DC table, and streetwise, heal, athletics, etc used the moderate DC tables. (All pre-errata, but presumably this would be adjusted by DMs). This is probably a good thing as far as making skill challenges tell a story (though I still don't think you can do anything with a skill challenge that you couldn't do better by discarding some or all of the mechanics). However, like the opaqueness of the DCs discussed above, this encourages players to treat everything as though it were a hard DC challenge. You never know if you'll be rolling on the easy table or the hard one until you try.
But the fundamental mathematical problem that skill challenges pose is that, when you are dealing with large numbers of rolls, the likelihood of failures goes up dramatically. Consequently, there is a very narrow band of probabilities where it can have the kind of probable outcomes that we would judge desirable. And +2 or +3 here and there is usually enough to turn any DCs that ordinarily would make success attainable but not automatic into automatic success while. Any DCs that avoid the autosuccess challenge tend to produce automatic failures (such a the weak skilled characters who have only a 1% chance of winning a hard DC skill challenge). In any d20 system when you need to have odds hovering in the 75-90% range for the system to succeed, you are only a short step away from ending up at 100%.
The revised DCs in the errata don't change how odds of an individual success relate to odds of success on the skill challenge. They just take the odds from the sucky end of the graph to the rosy end... sometimes. But it's still a distribution that is bad for the game. Just because it used to be worse doesn't mean that it's good now.
Mind you, your DM in your example should also reread the rules, because even the DMG clearly mentions something about using unmentioned skills. I am sure, that DM would have similarly screwed up in your 3E example.
Maybe and maybe not. The 3.x assumption regarding skills was that they were for concrete purposes and generally functioned against fixed DCs. This led to some silliness with diplomacy, but there was much more of a sense that there are rules for what your character can do and if you try something that is within those rules, it should work. That differs greatly from the 4e model that functions in skill challenges where the rules are specific to each situation and are not expected to interact with each other. As a consequence, the 3e model encouraged DMs to ask, "can you do that?" The current model encourages DMs to ask, "is that a primary skill?" To get more specific, the 3e style answer to, "I dominate him and have him tell us where his friends are" is "that's against his nature, he gets a save... failed, ok, he tells you." The 4e equivalent--"I use rending fear of Khirad to make him tell us where his friends are" is much more likely to get, "Hmm, is that one automatic success or two? Roll four more successes and you'll get somewhere." In the context of skill challenges, 4th edition tries to box creativity into the context of the minigame rather than allowing creativity to bypass the minigame entirely. Without the minigame, you don't have that tendency.
Your examples about WATE1-1 (and probably BALD1-1) are also more about DMs screwing up (either that or players screwing up because I have had players being so stuck in weird dice roll mode that no matter what I tried to did not get into the problem solving mode).
As for the pit example, why is it that you keep up dragging examples that are simply wrong?
Because it is a very clear example of something that a character would be able to do outside a skill challenge but cannot do as soon as he enters skill challenge-land. The example is only wrong if you insist on looking at the pit as the skill challenge in and of itself rather than as something that might frequently show up in various skill challenges.
Getting over the pit should NOT be a skill challenge in itself, unless of course, there are external factors raising the DC to the expected levels. In your case the sides of the pit should be muddy and slippery, or crumbling or whatever else the author can think of to explain the raise in DC.
No one is saying that getting over a pit is a skill challenge. What I am saying is that, while the character would consider jumping over things outside of the skill challenge environment, as soon as he enters the skill challenge environment, jumping over things has to immediately depart from his list of live options. If the DM is, for instance, running an urban chase skill challenge, the wizard had better not try to jump from rooftop to rooftop, even if he was doing so in the fight immediately prior to the skill challenge.
As for why not let a DM decide on success without the skill challenge: in itself because it would become rather arbitrary. As of yet, I have not had players deviate more then 1 or 2 checks from the chosen complexity. Then again, during preperations I spend some time contemplating the goals of the skill challenge, and what checks and actions might be necessary to achieve them. I then add or remove obstacles to get at the correct complexity, knowing enough of the scene to do so on the fly while running the game. In itself this is no different from a fight where I look at the aimed challenge and decide upon tactics and terrain to make sure that aimed challenge is reached.
