D&D 3E/3.5 Skill Challenges in D&D 3.5

I've played 4E, and I found I liked the general idea of Skill Challenges, a formal framework for non-combat challenges. The 4E mechanics had their problems, but the concept was appealing.

How would/should we go about implementing such a thing in 3.*?

One problem I saw with 4e was tied to the highly specialized/focused nature of skills in that edition: You either had the skill or you didn't, with no gradation or granularity. People selected Skills that they had the stats to back, so if they had it they were generally god-like in that area, and if they didn't have the stat/skill they were a pure liability. As a result, only PCs with the right skills ever participated.

Another issue I had was the meta-game aspect: For example, the party is pursuing someone across the desert. Tracking/Survival is an obvious element. So is Endurance. If horses are being used, Ride and Handle Animal come into play, making character Endurance less of an issue. Perhaps a Knowledge of Geography, to try and figure out where the enemy is heading.

When run in 4e, one character will likely make the tracking checks, which is fine. He'll likely make the Endurance, Survival and Ride checks, perhaps even the Handle Animal ones. And because he has good numbers, nobody else gets lost or tired, none of them get any saddle sores and their horses remain in good condition.

And if anyone who doesn't have the Skill and the Ability to back it up even reaches for a D20, they get slapped down.

I want to change that last part.

So overall I'd want to see some requirement that every PC has to contribute, with perhaps a minor penalty if they don't. Even an unskilled or low skilled person can contribute an Aid Another check on a lot of skills. And the Skilled person can Aid Another for the low skilled. "Here, you need to snug the saddle cinches evenly or your horse will end up with sores. And don't give him too much water all at once after a long hot ride. You can make him sick."

I'd also probably want a framework where the Challenge Rating of the Skill Challenge (for EXP purposes) would be based on the baseline DC of the Skill Checks involved.

Perhaps, instead of a "X number of successes before Y number of failures" goal, maybes a total required success count, with each failure subtracting from the total. More attempts, of course, takes more time, so a time limit element would seem appropriate.

Let's kick this around a bit, see if we can come up with something that works.

Not sure I can help much, but I’m very interested in the results!

Well, contribute what you can.

Lets talk about Skill DCs and CR for the challenge.

The standard for a common or casual success on a skill check is a 10. 15 for some that are supposed to be an actual challenge. So let's start there.

To raise the CR of the challenge, should we standardize on higher DCs, or more checks?

In our "Track the enemy across the desert" example, the Survival check DC for tracking starts with a base number derived from the terrain. "Desert" can be a sandy Sahara type place, or baked clay type soil, or anything in between, so we can vary that base DC from 5 to 20.

After that bonuses and penalties can apply based on the size and number of creatures, the amount of time since they passed, and weather conditions, giving us the ability to fine tune the DC by just about any amount we want.

Tiny or diminutive creature is +8 to the DC, every day that's passed adds another +1, any rain adds +1 per hour of rain, a covering of snow (or presumably blown sand) adds 10.

So what would we like the DC to be for a CR1 check. What should it be for a CR 5 check? CR 10?

Presume someone tries to substitute a skill, such as Knowledge Geography or even Knowledge Local, to deduce where the enemy might be heading? Creativity certainly should count for something.

Again, we can vary the DC up or down from our Survival base, but I don't think it should be more than five either way.

Now consider Endurance checks. Easier if the party moves after dark (Horses, it turns out, have low light vision). At the same time "Poor Visibility" adds to the Tracking DC, per the rules.

Broadening this from the one example, each Skill Challenge should have a Primary skill check , the main goal of the challenge. The DC for that sets the tone for the rest.

Secondary skill checks in the challenge should have lower DCs. I'm thinking that there should be at least one Secondary check involved per X points of the Primary check's DC.

So what is X? 3? 5?

In a simple Tracking challenge the Primary DC should be 15. Endurance or Ride *AND* Handle Animal would be a secondary. That makes me think the secondaries would be one per Primary DC - 10 / 5. So our DC 15 simple challenge would have one secondary. A DC 20 primary should have two secondaries.

A basic trapped door in a dungeon could then be expressed as a Skill Challenge: Search is the Primary, to find it. Disable Device would be a Secondary, as would Sleight of Hand/Open Locks. Magical traps might call for a Knowledge Arcana check, so the party knows what kind of cover or precautions they need to take.

Note that the Search DC could be relatively easy, say a 15 or 20, while the Disable and Open DCs could be higher. (Any lock that can be opened on a 20 or less isn't really locked, since a minimally skilled Rogue can Take 20 and get it every time.)

Thoughts? Am I off on a wrong track?

One of my issues with skill challenges is the 2X successes before X failures doesn't fit the possible scenarios. Sometimes, it's an endurance contest and a net X successes means success. Other times, it is a race between groups and the first to reach X successes achieves their goal.

For the tracking scenario I'd suggest two thresholds: Net X successes and the quarry is found. Net Y failures and the quarry is lost for good. If the speed of the party is slower than the quarry, that's an automatic failure per roll period. If the speed of the party is faster than the quarry that offers a automatic success in a roll period (or adds one if the party manages a tracking check or doubles the number of successes developed in the period... depending on terrain, and how you envision the chase).

So overall I'd want to see some requirement that every PC has to contribute, with perhaps a minor penalty if they don't. Even an unskilled or low skilled person can contribute an Aid Another check on a lot of skills. And the Skilled person can Aid Another for the low skilled. "Here, you need to snug the saddle cinches evenly or your horse will end up with sores. And don't give him too much water all at once after a long hot ride. You can make him sick."

Click to expand...

Skill challenges in 4e had that requirement which was another thing I chafed at. Sometimes everyone contributing makes sense. Other times, it gets in the way. One of the things I'd suggest is more like the system floated during the 5e playtest: everyone picks a role appropriate to the task at hand and becomes responsible for the success there.

For the track the quarry across the desert, you need a tracking station, a survival station, a mount caregiving station, maybe a navigation station, and maybe a scanning for danger station. Each PC can fill a single role. Any role unfilled automatically fails -- though that failure may not directly impact the challenge. "We tracked him to his lair! Great! Where are we again? Which way is home? I thought you were navigating!"

Challenge rating / Encounter Level determination is somewhat harder. I'd suggest CR should be based on the highest required DC. Something like a CR = (highest DC - 15) or level of spell that would bypass the challenge.

Movement rates are always an issue.

According to RAW, a tracker moves at half speed when tracking, unless they take a serious negative to the skill check or have the Fast Tracker ability (Rangers get that at some point).

But, as the tale of the Tortoise and the Hare points out, how fast you *can* move isn't always how fast you *do* move. That's where Endurance checks come it. And tactics.

You know that your quarry needs water just like everyone else. If your Knowledge Geography or Knowledge Local check is good enough, you may know where they're heading, so you can try to cut them off. Anything they might do to lead you astray or to cover their tracks now plays against them, since it slows them down.

In the chase scene I'd probably skip the X success v Y failures test, and go straight to a time challenge. Failures waste time. if you don't catch them before they get to that water hole, or their reinforcements, you can't catch them at all. Or perhaps you'll fall too far behind, or night will fall. Something that says "Too late".

The victory conditions should probably be one of several standard solutions, rather than always Success v Failure counts.

So let's build a generic Skill Challenge.

The party must achieve X before the adversary/situation reaches Y.

My party is 7th level, so I want this to be CR 5.

That suggests a Primary skill check DC, and a number of secondary Skill Check DCs.

Award a 10% EXP bonus for full party participation, and a 10% Exp penalty if less than half contribute.

So what should the DC of the primary skill check be, for a given Skill Challenge CR?
What would be the DC of the Secondary skill checks?
How many different secondary skills checks should there be?
How many successful checks should be required to meet the Skill Challenge?
What constitutes a failure?

We need formulas or methods to answer these questions. Suggestions welcome.

Greenfield said:

Movement rates are always an issue.

According to RAW, a tracker moves at half speed when tracking, unless they take a serious negative to the skill check or have the Fast Tracker ability (Rangers get that at some point).

But, as the tale of the Tortoise and the Hare points out, how fast you *can* move isn't always how fast you *do* move. That's where Endurance checks come it. And tactics.

You know that your quarry needs water just like everyone else. If your Knowledge Geography or Knowledge Local check is good enough, you may know where they're heading, so you can try to cut them off. Anything they might do to lead you astray or to cover their tracks now plays against them, since it slows them down.

In the chase scene I'd probably skip the X success v Y failures test, and go straight to a time challenge. Failures waste time. if you don't catch them before they get to that water hole, or their reinforcements, you can't catch them at all. Or perhaps you'll fall too far behind, or night will fall. Something that says "Too late".

The victory conditions should probably be one of several standard solutions, rather than always Success v Failure counts.

So let's build a generic Skill Challenge.

The party must achieve X before the adversary/situation reaches Y.

My party is 7th level, so I want this to be CR 5.

That suggests a Primary skill check DC, and a number of secondary Skill Check DCs.

Award a 10% EXP bonus for full party participation, and a 10% Exp penalty if less than half contribute.

So what should the DC of the primary skill check be, for a given Skill Challenge CR?
What would be the DC of the Secondary skill checks?
How many different secondary skills checks should there be?
How many successful checks should be required to meet the Skill Challenge?
What constitutes a failure?

We need formulas or methods to answer these questions. Suggestions welcome.

Click to expand...

The question becomes how likely do you want a single check to succeed? I'd suggest something between 65 - 75% success rate for an appropriately adept character. For CR 5, that would mean an appropriately talented character would have a bonus of +5 (skill ranks) + 4.5 (attribute 18-20) = about +10 total. A 65% success rate means rolling a 8 or better and that pegs the DC at DC 18 for the primary skill. At CR 15, the base bonus is +15 + 7-8 or +22-23 total so we'd be looking at DC around 30-31 . The formula at that point would be DC = 10 + 1.5 * CR (to account to steadily increasing primary attributes and the growing possibilities of skill bonuses).

As for the number of primary skill checks, we're back to it depends. Overall, what is the preferred probability of success for an appropriate group? If it is a simple endurance contest (requires net X successes) then the probable number or rolls will be 3X (every 3 rolls will result in 2 successes and 1 failure giving a net increase of 1). A race challenge where only success counts for either side is straightforward to calculate if the opposition probability of success is known. A classic skill challenge of 2X successes before X failures is more opaque.

The number and value of specific secondary checks should be customised by challenge. here should always be some to allow different character types obvious ways to participate. There are a couple of types of checks I like to include: those that offer situational bonuses and those that prevent circumstantial problems.

Secondary skills are probably a little easier to a little harder depending on the failure ramifications. If the skill check will grant a bonus to another primary check on success but offers little downside beyond wasted time, raise it a couple of points to DC 20. If the secondary check is meant to prevent a negative effect and doesn't otherwise move the challenge to completion, drop it a couple of points to DC 15-16.

You could use the Influence or Pursuit types of multiple checks with various scaling DC's that are in the Ultimate Intrigue book for Pathfinder. The complete chapter is linked here:
http://www.d20pfsrd.com/gamemastering/other-rules/intrigue/

JoeElf said:

You could use the Influence or Pursuit types of multiple checks with various scaling DC's that are in the Ultimate Intrigue book for Pathfinder. The complete chapter is linked here:
http://www.d20pfsrd.com/gamemastering/other-rules/intrigue/

Click to expand...

Thanks. I'll have to look at that when I have time.

Greenfield said:

One problem I saw with 4e was tied to the highly specialized/focused nature of skills in that edition: You either had the skill or you didn't, with no gradation or granularity. People selected Skills that they had the stats to back, so if they had it they were generally god-like in that area, and if they didn't have the stat/skill they were a pure liability. As a result, only PCs with the right skills ever participated.

Click to expand...

Isn't this only exacerbated in 3.5? If you don't put ranks in a skill, you're not getting any better. In 4e, you at least get half your level to the check, even if you lack a good ability modifier. In 3.5, if you lack the ranks, you're incompetent compared to someone with the ranks (especially with skills that can only be used trained), and this issue becomes bigger and bigger at higher levels.

Another problem is setting the DCs. In 4e, the worst in the party at climbing or bluffing will have a slim but a fair chance at attempting a moderate DC skill check, and that's because the DCs and the skill bonuses advance with some degree of respect to one another. In 3.5, a level 8 PC with full ranks in a skill, a fair or good ability modifier, and a skill-boosting magic item (plentiful and stronger in 3.5) will pretty much auto-succeed on checks that an untrained PC will have a 5%-10% chance of passing -- sometimes 0% because a natural 20 isn't a guaranteed success in skill checks. Thus, what could possibly be a "moderate" or "difficult" DC in 3.5, when the gap between the untrained and the specialist is so huge?

I've thought about implementing 4e-style skill challenges in my 3.5 game, but this issue with the DCs and the way the good keep getting better and the bad keep getting (comparatively) worse cause it to be too much of a hassle. A very real problem with the first iterations of 4e skill challenges was that the chance of success was too low in longer challenges, all thanks to poorly thought out skill DCs. In 3.5, it's next to impossible to set DCs that stay fair and robust from level 1 to 10 and to 20 -- the same issue the early 4e version had would only be magnified.

I wasn't trying to make this a "My favorite system is better than yours" thing.

In 4e, you're either trained in the skill or you aren't. No granularity to speak of. In 3.5 characters get skill points per level, and can use them to add to existing skills or invest in new ones. Some are paid for at a 2-for-1 price because they're "Cross Class", which discourages some people investing in them, and limits how many skill ranks they can actually have.

Not touting one over the other, but since 4 seems to an "All or Nothing" kind of thing with skills, my observation was that people only invested in skills they could do well in, that they had the Ability bonus to back up. What I saw in play was that, if you had a skill, you were functionally a god in that skill. If you didn't have training in the skill, you also didn't have the ability score to back it, and thus you sucked at it. With no elective skill points to assign after creation (short of spending a feat), there weren't any areas where someone might become moderately competent over time.

Now it's true that everyone gets a bit more competent over time (+1 every two levels), the general rule was that DCs also increased at the same rate, so your numbers might look better as you advance, but it's an illusion, a point inflation that makes the apparent gains meaningless. Your actual success rate didn't improve over time. You spend a career as good at what you do as you were at 1st level.

In your example of Climbing, in 4e if the party had one good climber, he could effectively carry an army on his back, and they never needed to even try. In fact, if they did try it could cause them all to fail. (Okay, not really an army, but an adventuring party of just about any size.)

That's a feature I'm trying to move away from in this effort.

As for setting the DCs: That's kind of what the framework I was proposing was working towards.

You want a Skill Challenge with an effective Challenge Rating of X. Dcs for skill checks should be based on X, say X + 14 just to pick a number. so a CR 1 Skill Challenge has Skill checks with a DC of 15. If your group is going to be a bit weak in some areas, call for more skill checks in that area, but lower the individual skill check DC by one or two per additional required dice roll.

Now just fill in the blanks: What is the goal or theme of the Skill Challenge? What skill checks are appropriate? What Challenge Rating do you want the whole thing to be?

We might find that CR +14 isn't the right target. I don't know. It's kind of why I was asking people here.