D&D General Skill challenges: action resolution that centres the fiction

[pemerton]

If you need to roll ten on your skill checks (so pretty average difficulty)

That's not particularly typical.

At 1st level, a good skill bonus is +8 (+3 stat, +5 training). And medium difficulty at 1st level is DC 12 (Rules Compendium, p 26). Even with a poor stat, training will be +5 which makes 7 or 8 the number of success.

At 30th level, a good skill bonus is +34 (+8 stat, +5 training, +6 sundries from character build and items, +15 level). And medium difficulty is DC 32. Even with a poor stat (+0 or +1 at that level), that's 12 less sundries for a medium DC, and if the player wants to succeed on those skill checks they will acquire those sundries.

That's before we get to resource use, which is relevant even at 1st level but of course far more relevant at 30th (given the extreme depth of 4e character builds as levels are gained).

 

[clearstream]

Oh wow! The math seems pretty broken to me. Do I read this correctly? If you need to roll ten on your skill checks (so pretty average difficulty) the party has only 22% chance of beating a complexity two skill challenge and only 2% chance of beating a complexity five challenge?

It was very broken, but remember that's in 2008. The 2010 RC works to patch up the maths by prescribing a mix of DCs and translating what had been vague before into concrete rules for Advantages on higher complexity SCs.

 

[clearstream]

That's not particularly typical.

At 1st level, a good skill bonus is +8 (+3 stat, +5 training). And medium difficulty at 1st level is DC 12 (Rules Compendium, p 26). Even with a poor stat, training will be +5 which makes 7 or 8 the number of success.

At 30th level, a good skill bonus is +34 (+8 stat, +5 training, +6 sundries from character build and items, +15 level). And medium difficulty is DC 32. Even with a poor stat (+0 or +1 at that level), that's 12 less sundries for a medium DC, and if the player wants to succeed on those skill checks they will acquire those sundries.

That's before we get to resource use, which is relevant even at 1st level but of course far more relevant at 30th (given the extreme depth of 4e character builds as levels are gained).

By 2008, SCs were widely acknowledge as broken - in terms of the underlying probabilities. The 2010 rules are what I would think folk should use.

With 5e they made an extremely similar mistake on Group checks, which they later patched over with modified advice on how to run them. I would have thought they'd have learned by then!

EDIT The problems created included - as one poster put it - "I guess I have to sit-out this encounter while McSkilly hogs all the fun."

 

[Manbearcat]

Oh wow! The math seems pretty broken to me. Do I read this correctly? If you need to roll ten on your skill checks (so pretty average difficulty) the party has only 22% chance of beating a complexity two skill challenge and only 2% chance of beating a complexity five challenge?

It was very broken, but remember that's in 2008. The 2010 RC works to patch up the maths by prescribing a mix of DCs and translating what had been vague before into concrete rules for Advantages on higher complexity SCs.

You guys need to scroll down to post 6 which links to the instantaneous errata by WotC (mid-July 2008).

Skill Challenges were fine after that errata. They improved after 2010.

Again, I’ve been running them since inception. In my running of them it’s probably something like 7.5-10 % failure rate across the distribution of all of the SCs I’ve run.

And here is the thing. The base maths become MUCH more adversarial as you climb into Epic Tier…so, on paper, you should be failing more, right? Wrong. Failure rate remains roughly the same in Epic because resource suite expansion giving the PCs the ability to call upon means to (a) outright earn successes (Rituals, Dailies, more willingness to make Coin fungible toward that end) and (b) the ability to call upon Encounter Powers to reroll checks (the original Advantage) or take Failures “off the board” or buff your allies checks.

 

[pemerton]

In the context specifically of a game text that includes what I will refer to imprecisely as "task resolution", would you say that SCs can bring in what I will equally imprecisely refer to as "intention resolution"? That is to say, would it make sense for a group to view the razor between a simple series of tasks, and an SC, as being pursuit of an overarching intention (that N-tasks will resolve)?

I don't know.

But anyone who wants to adjudicate action consequences (be they successes or failures) simply by extrapolation of immediate, granular causal consequences of the task performed by the acting character will struggle with skill challenge adjudication. For instance, they won't be able to introduce the canyon, as per @Manbearcat's example:

You are riding full throttle on horseback, with precious relic in tow (freshly stolen from an evil god's temple in order to bring it...wherever), through treacherous terrain where a single misstep by your horse could mean disaster. You are in the middle of the skill challenge when you fail a check that should indicate that you look like an inept fool and fall off your horse (even though you're an accomplished rider). Orrrrrrrrrrrrrrrrr...the check could signify a critical failure within the fiction external to your riding. Perhaps you reach a dead end with an enormous jump (difficult DC) over a rushing river canyon facing you. Outside of the obvious jump attempt, there are a number of interesting decision points that could spawn from this failure at your ride check (and corresponding skill use possibilities).

But without that sort of consequence narration, it's not clear to me how the GM is going to establish overall success or failure at the goal in a way that coheres with the emerging fiction.

In the tradition of wargaming, he may have been thinking of the die-roll as covering those myriad possible factors - wood chips hitting your eye and so on.

In which case, it's no longer true that the GM is establishing all the parameters of the fiction prior to the action being declared and the dice rolled.

The goal is something like surface to the table for player leverage those impactful factors they could reasonably observe and have power to interact with, and let the die roll model those innumerable factors - secret badgerhorses etc.

These distinctions between what might be interacted with, and what not, are in many cases arbitrary (or, perhaps, conventional).

D&D emphasises distances but not surface conditions - but both are impactful for a jump. It would be no less arbitrary for the GM to track all the surface conditions, or the winds, and to establish the distance of the jump as something to be read from the dice rolls!

The problems created included - as one poster put it - "I guess I have to sit-out this encounter while McSkilly hogs all the fun."

What problem? I've never seen this to be an issue in skill challenge resolution.

 

[clearstream]

You guys need to scroll down to post 6 which links to the instantaneous errata by WotC (mid-July 2008).

Skill Challenges were fine after that errata. They improved after 2010.

That's what I'm saying here: it's advisable to use the 2010 rules.

Again, I’ve been running then since inception. In my running of them it’s probably something like 7.5-10 % failure rate across the distribution of all of the SCs I’ve run.

Absolutely - having run them from inception, you've grokked the system. That's what I'm pointing out. I have read accounts here on Enworld of groups failing to grok the far more straightforward (mathematically) Group checks.

 

[pemerton]

You guys need to scroll down to post 6 which links to the instantaneous errata by WotC (mid-July 2008).

Skill Challenges were fine after that errata.

But what about all the players trapped in that month of 2008? Who's thinking of them!

 

[clearstream]

But what about all the players trapped in that month of 2008? Who's thinking of them!

Or those quoting 2008 rules hard at me

To be fair, from this vantage point looking back, one of the first tasks is to be sure what version of SCs folk are referring to. Is it right to say our baseline is 2010 then?

 

[Manbearcat]

That's what I'm saying here: it's advisable to use the 2010 rules.


Absolutely - having run them from inception, you've grokked the system. That's what I'm pointing out. I have read accounts here on Enworld of groups failing to grok the far more straightforward (mathematically) Group checks.

Sure.

User error is common in TTRPGing. Being quite deft at both the GMing role and the player role of any given game is a difficult thing. Its a skill to be earned via thoughtfulness, comprehension of principles/techniques/engine interactions and best practices, humility, and disciplined practice.

That is a good thing!

What isn't a good thing is what I see happen in a lot of Skill Challenge threads (historically). Refrains whereby people are basically committed to a decade + long effort of casting this sort of resolution and this particular game, 4e D&D, in as negative a light as possible and doing ZERO inventory of their own role in the problems their having):

• Fail Forward doesn't work!
• The fiction doesn't change after resolution!
• The fiction is meaningless!
• The fiction is too meaningful and there is no actual "game" there!
• The math is broken so the system doesn't work!
• This is an exercise in useless dice rolling and free form roleplay!
• You can't use Powers or Rituals in Skill Challenges! They're totally prescriptive railroads!
• The fiction of 4e is nonsensical!
• Intent/Goal-based consequences of action resolution equals meaningless, nonsensical, agencyless play!
• Fire doesn't cause ignition of flammable materials (and dozens of other wrong ideas about keywords and "Target - Creature")!

I could go on and on and on with these sorts of things.

The reality is, there was...and remains...a very large contingent of anti-4e sentiment by a particular (and particularly raucous and in your face...here and in real life) group of D&D players who have died on every hill...at every opportunity...for 14 years...overwhelmingly wrongly...who will never course-correct their wrongness and put it on the record because inevitably they're going to trot out the same nonsense 2 years later at first opportunity.

 

[clearstream]

I don't know.

But anyone who wants to adjudicate action consequences (be they successes or failures) simply by extrapolation of immediate, granular causal consequences of the task performed by the acting character will struggle with skill challenge adjudication. For instance, they won't be able to introduce the canyon, as per @Manbearcat's example:

But without that sort of consequence narration, it's not clear to me how the GM is going to establish overall success or failure at the goal in a way that coheres with the emerging fiction.

That's a great example, thanks! I think folk using task resolution picture that they do narrate the consequences... of the task. It feels like the broader narration @Manbearcat brings in orients toward the intention. I think it becomes necessary, really, to work within the ambit of the intention, as that is where things must if successful land. It must come to legitimately follow.

In which case, it's no longer true that the GM is establishing all the parameters of the fiction prior to the action being declared and the dice rolled.

Yes, and we often see folk at the table narrate that extra detail from the roll. A previously unnoted detail is referenced - "You trip over a corner of the carpet..." or whatever, when we didn't mention a carpet until now.

These distinctions between what might be interacted with, and what not, are in many cases arbitrary (or, perhaps, conventional).

D&D emphasises distances but not surface conditions - but both are impactful for a jump. It would be no less arbitrary for the GM to track all the surface conditions, or the winds, and to establish the distance of the jump as something to be read from the dice rolls!

Motives for not tracking all those things include the questionable accuracy in doing so, their possible irrelevance, and simple human limitations. It seems totally uncontentious to me to say that wargamers don't suppose there is no mud, potholds, blowing leaves, etc on the field where they aren't expressly listed and modelled, but leave it up to dice to say if those too-many-to-count factors had an impact.