D&D 5E The Problem with Skill Challenges and a Solution I Use

[Starfox]

To sum up the link: Insight is global (about the entire demeanor of an NPC), not local (rolled for each and any phrase uttered that might be a lie). Here's an example:

DM: Your NPC companion points you thru the tangle of buildings in the thieves' district to a dark alleyway with a single door lit by a blue lantern at the end.
Player: Is the NPC hiding something? *Rolls really high*
DM: The NPC keeps looking over his shoulder like he is nervous or expecting to be back stabbed at any moment. He has been very cooperative, but his words have been terse, like he is revealing only precisely what you ask.

Notice the DM did not say "yes, he's lying to you."

Now it's up to the players to converse more with the NPC, threaten him, or use what they already know to decide whether the NPC's jitteriness is because he is being coerced by the party to guide them, because the NPC has bad history with the thieves' guild, or because the NPC is setting them up for an ambush.

While this is fine and dandy, it also opens up a whole different can of worms - how much should the player be forced to read the DM, and how much are we measuring the player's social skills as opposed to the character's. If I gave the above response to a player with poor social skills, I might get "please answer the original question" back - and I think that is a fair comment. Just as a player cannot improvise every word of his character's Bluff, he cannot be expected to read every nuance of the DMs language.

In the situation above, I feel that a mediocre roll should yield the information given in the example, while a good roll should be more direct and focused. If the conversation is actually acted out, the information above is what anyone should be able to infer from how the DM role-plays the guide, while a character with a good skill roll should gain more. But I know this varies widely from table to table - how much information to give out and how is a big part of DMing style.

 

[Stormonu]

I think in most cases, multiple checks will help keep unlikely results from showing up too much. In other cases, I think beyond the stat bonus the DM should have the option to throw in a situational bonus. In the case above of the Ogre vs. Elf, the Ogre's large size could give him, say, a +4 size bonus on top of his +5 bonus from Strength. In the case of the lying individual, he might get a situational bonus because it's his "livlihood" - he may get, say, a +3 bonus to his Bluff check. Conversely, if the PC had some background in ferreting out the truth - say Spy or Law Enforcement, I might give the PC the same sort of bonus to Sense Motive check.

 

[N'raac]

I think in most cases, multiple checks will help keep unlikely results from showing up too much. In other cases, I think beyond the stat bonus the DM should have the option to throw in a situational bonus. In the case above of the Ogre vs. Elf, the Ogre's large size could give him, say, a +4 size bonus on top of his +5 bonus from Strength.

He gets a size bonus to carrying capacity (where a Small character gets a reduction), so this seems perfectly reasonable. The 10 STR Elf still wins on a 20 if we use the "take 10" approach, with about the same odds of rolling 10 or more better than the Ogre if both roll.

In the case of the lying individual, he might get a situational bonus because it's his "livlihood" - he may get, say, a +3 bonus to his Bluff check. Conversely, if the PC had some background in ferreting out the truth - say Spy or Law Enforcement, I might give the PC the same sort of bonus to Sense Motive check.

Seems the fact it's his livelihood or background would suggest it is a class skill and/or that he invest skill points in being good at the skill. Should every scholarly Wizard get a +3 to all knowledge checks because it's his background/livelihood? Surely the same logic would give all Clerics +3 in Religion, all Spellcasters +3 in Spellcraft and all arcane casters +3 in Knowledge Arcana.

Using multiple rolls would reduce unlikely results, but add to the rolling, so it's a tradeoff from the perspective of the OP comments. Using multiple dice (3d6 or 2d10) has a similar result in that it creates a bell curve, but avoids increasing the number of rolls.

I like the Take 10 approach. I wonder how much skill contest design was about always letting the player have a roll as many players seem to resent things "happening to them" without the opportunity to make a roll. If we always default to a player roll, could we remove skill contests? If the PC Bluffs, his DC is 10+opponent's Sense Motive bonus, but if the NPC Bluffs, the PC rolls Sense Motive against the NPC's 10+Bluff bonus.

We could take this further. PC attacks, he rolls vs the opponent's AC. But an AC 19 PC rolls d20+9 to resist an attack DC of 10+atacker's BAB. [Probably have to make low rolls on the defense roll potential criticals for this to work.] No more saves for NPC's - instead, the PC rolls d20 + spell level + stat modifier against 10+opponent's save bonus.

Let's consider this in more detail. Fred the Fighter has AC 19, and his opponent has a +6 attack bonus. He would need a 13 to hit, so 40% chance. Fred rolling d20+9 vs DC 16 needs a 7, so a 70% chance to be missed. The rounding is an issue, as is the "success on an equal roll" aspect. The DC of avoiding an attack would have to be 12 + Attack Bonus. The Save would need to be similarly adjusted to avoid changing the odds. Otherwise a DC 17 Spell vs a +5 save bonus goes from needing a 12+ (45% chance of a save) to a d20+7 vs DC 15, so an 8 required (35% the target is not affected).

It's a bit easier for the PC to win that opposed skill challenge, since the average roll against him would be 10.5, not 10. The advantage always falls to the person making the roll, which would always be the PC, so this change would be to the PC's advantage over the long haul. If two PCs oppose one another, both get to roll, eliminating that advantage and going back to the opposed roll system. Maybe some NPC's (important ones) similarly get to roll, but mooks and cannon fodder always Take 10.

 

[Rhenny]

I like all of the ideas that have to do with taking 10 and adding bonuses, yet, with D&DNext I feel that there is an attempt to get away from the numbers. One of the joys I'm finding DMing Next is that I don't need to worry so much about finding and adding bonuses or penalties to rolls. It has freed me up to really narrate and make the game flow more smoothly. (advantage/disadvantage as an example)

For me, if the player wants to push against a door that an Orc is holding closed, I can do it by just deciding straight up, that will be a moderate check (DC 15 because the Orc has a really good position and has prepared himself for the PC attempt....or I could say DC 10 because the Orc is just leaning on the door, not really digging in). Then the player rolls the d20 and adds strength modifier, and it is done. The key is, I don't have to look at the Orc stat block to determine its strength or decide what bonuses to apply. It becomes a quicker more intuitive decision that keeps the game moving very quickly.

In terms of the sense motive checks, I still find it tough to decide how much to "act" and how much to "tell", but at least if I set the DC based on what I determine the NPC/monster to be like, it just makes more sense to me. Not all NPC or monsters will have a DC 25....just ones that are used to being deceptive. Oh...here's another problem with sense motive....when a PC who isn't used to dealing with Lizardmen, interacts with a Lizardman, how will the PC know what body movements, facial expressions, tells and ticks indicate different dispositions? I find it much easier as a DM to set a DC of 20 or 25 in situations like that rather than applying a penalty to the PC player's roll. And, since I can set the DC for that situation, I feel it is just more consistent to set DCs for nearly all interaction, exploration, action resolution attempts. Does that make sense?

 

[Starfox]

If what you really want are more centered probabilities, best of three works really well. For repeated rolls, a linear distribution like 1d20 works out well.

Should every scholarly Wizard get a +3 to all knowledge checks because it's his background/livelihood? Surely the same logic would give all Clerics +3 in Religion, all Spellcasters +3 in Spellcraft and all arcane casters +3 in Knowledge Arcana.

This sounds very much like the class skill bonus classes get on certain skills in Pathfinder. Problem is only that there is no skill for strength feats.

when a PC who isn't used to dealing with Lizardmen, interacts with a Lizardman, how will the PC know what body movements, facial expressions, tells and ticks indicate different dispositions? I find it much easier as a DM to set a DC of 20 or 25 in situations like that rather than applying a penalty to the PC player's roll. And, since I can set the DC for that situation, I feel it is just more consistent to set DCs for nearly all interaction, exploration, action resolution attempts. Does that make sense?

Yes, I agree this is often a wise way to go. Sense Motive can't always be opposed by Bluff.

 

[N'raac]

Oh...here's another problem with sense motive....when a PC who isn't used to dealing with Lizardmen, interacts with a Lizardman, how will the PC know what body movements, facial expressions, tells and ticks indicate different dispositions? I find it much easier as a DM to set a DC of 20 or 25 in situations like that rather than applying a penalty to the PC player's roll. And, since I can set the DC for that situation, I feel it is just more consistent to set DCs for nearly all interaction, exploration, action resolution attempts. Does that make sense?

Sorry, no, it does not make sense to me. How is setting, say, a DC 5 points higher for dealing with an unfamiliar species with very different body movements, facial expressions, tells and ticks any more intuitive than a 5 point penalty to the roll? And, again, does the Lizardman get the same penalty to his ability to interact with the human? Imposing a set DC increase (decrease) for various levels of cultural unfamiliarity makes sense, but my preference would be some consistency. If the player envisions his character having an intuitive understanding of social situations in general, and the GM decides huge penalties are in order when dealing with other races, that's a recipe for conflict at the gaming table. I'd rather see the ability to point to specific examples (eg. "similar humanoids such as dwarves or elves", "very different humanoid species such as trolls or myconids", or "completely different creatures such as otherplanar beings").

This serves two functions. First, it sets expectations - the player and GM can be placed on common ground that a 10 point penalty/DC increase applies to Sense Motive of a Fire Elemental, so the penalty for a lizard man cannot be that high, but a 2 point penalty applies to even very similar creatures, and a lizard man seems no more different than a myconid, so the appropriate penalty must be somewhere between 2 and 5. It's not "just as easy as reading a human", but it's not a nigh impossible task which should see a DC of 25 or 30 either. Second, it provides consistency, so we don't see a DC 20 set for Sensing the Motive of a Water Elemental today, and a few weeks later the Sense Motive on an Orc is set at DC 30.

This sounds very much like the class skill bonus classes get on certain skills in Pathfinder. Problem is only that there is no skill for strength feats.

It does, indeed. SHOULD there be a skill for strength feats? Assuming there is, should an Ogre have this skill, or is he simply a dumb brute who relies on brute strength, rather than any actual skill?

I'm inclined to extrapolate. Since much of the skill of tug of war is getting your opponent off their feet, Acrobatics (which subsumes Balance) may be a good choice, especially as this has largely become the Athletic Abilities skill. Climb is another possibility - noting and using footholds seems likely to help in such a struggle. Perhaps we can adapt the Aid Another skill (since Pathfinder lacks synergies) - use your own Climb and Balance skills in similar fashion to Aid Another to gain a bonus on your STR check, or perhaps we beef this up with a bonus based on how well each of those complementary skills were rolled.

If the tug of war is really game-important, then let's spend some time on it, and let the characters spell out how they will leverage their skills towards victory. If not, let's get on with it, perhaps simply with a STR roll modified by Size.

 

[Starfox]

4E had an Athletics skill that was Strength-based that I used for these, even if the rules recommended using straight Strength rolls.

 

[Klaus]

DM Advice: Setting DCs for Contests rather than Having Contests
<SNIP>

I'm sensing that your problem is with the swinginess of the d20. The answer seems to trade it for 2d10, or even 3d6. That creates a bell curve, and ability score modifier become much more important. That makes it virtually impossible for the Str 10 elf to win a contest against the Str 21 Ogre.

 

[Quickleaf]

While this is fine and dandy, it also opens up a whole different can of worms - how much should the player be forced to read the DM, and how much are we measuring the player's social skills as opposed to the character's. If I gave the above response to a player with poor social skills, I might get "please answer the original question" back - and I think that is a fair comment. Just as a player cannot improvise every word of his character's Bluff, he cannot be expected to read every nuance of the DMs language.

In the situation above, I feel that a mediocre roll should yield the information given in the example, while a good roll should be more direct and focused. If the conversation is actually acted out, the information above is what anyone should be able to infer from how the DM role-plays the guide, while a character with a good skill roll should gain more. But I know this varies widely from table to table - how much information to give out and how is a big part of DMing style.

Yeah, it is up to each DM. My point was more along the lines of the old argument that "sense motive isn't detect lie, and diplomacy isn't charm person."

Of course, this argument can be applied by extension to the elf vs. ogre tug-of-war. Like with insight, where I was pointing out the importance of details and just what a "success" means, you could argue that the elf succeeding looks different than the ogre succeeding.

For example, the elf wins by feeding the ogre a bunch of rope, then when the lumbering ogre is off balance, the elf throws a coil of rope around a boulder for leverage and yanks with all his might. The ogre topples forward, and the elf collapses backward.

The ogre wins by just yanking with sheer strength and no strategy. Depending on just how much the ogre wins by, the elf might end up face down in the dirt, with rope burns on his hands, or still clutching the rope as he is catapulted thru the air toward the ogre.

Even in a straight Strength vs. Strength test, there is room for creatively interpreting the results of die rolls to explain the fiction.

Of course of [MENTION=18333]Neechen[/MENTION]'s problem really is the swinginess of a d20, then the solution [MENTION=607]Klaus[/MENTION] and others have been saying is self-evident (though it does create some problems of its own).

 

[Warbringer]

I'm sensing that your problem is with the swinginess of the d20. The answer seems to trade it for 2d10, or even 3d6. That creates a bell curve, and ability score modifier become much more important. That makes it virtually impossible for the Str 10 elf to win a contest against the Str 21 Ogre.

I find it quicker just set a target of either 10+ or 11+ modifier, depending on whether you're a pro palyer DM or not