Another skill challenge solution

[Fredrik Svanberg]

I read Stalker0's new Skill Challenge system (found here: http://www.enworld.org/showthread.php?t=229796) and while the math probably is sound and it looks like it would work just fine, I still think it's a little too complicated. Perhaps I have misunderstood the original skill challenge system in the book, and after I've posted this I will go and read it again just to make sure, but it seems to me that the core concept of gaining a set number of successes before too many failures is flawed.

I would suggest that instead of counting failed rolls, the system should be changed as follows:

Complexity shows how many successes a party must get in a certain amount of time, expressed as "rounds". E.g complexity 1 would mean 4 successes in 2 rounds, etc. During each round each character would be allowed to use a skill, according to the rules in the DMG, including helping another character.

Note that for non-standard parties (i.e. parties with more or less than 5 members) the number of successes would have to be adjusted. Smaller parties should face challenges with fewer required successes in the same amount of rounds, and vice versa. Just grabbing numbers from thin air here; 25% less for a 4-person party, 33% less for a 3-person party, 50% less for a 2-person party and 66% less for a party of one.

A round doesn't have to be a combat round in a skill challenge. It can be any amount of time that the DM wants. Some skill challenges are part of combat encounters and in those cases combat rounds would be appropriate. Other challenges could use rounds that are a minute, an hour or even a day long. In some cases the actual length of a round isn't important but it can be used for dramatic effect when the characters are working against a deadline.

I haven't checked if this simple suggestion is enough to "fix" skill challenges or if the DCs and complexity values still needs an overhaul. I bet mathematically talented people like Stalker0 could make a quick estimate and find out if this would work or if it would just make things worse. I will do my own calculations later when I have more time, I just wanted to throw this idea into the proverbial circle to see it get torn apart by razor-sharp minds.

 

[Stalker0]

If we assume that each party member can make 1 roll per round, 4 successes in 2 rounds means that a party of 5 has 10 rolls they can make, and must succeed 4 times. This is effectively saying you need 4 successes before you get 7 failures, because once you get 7 failures, you cannot get the number of successes you need in time.

If we assume the standard 50% success rate on each individual role, the final answer is:
82.81%, which isn't bad at all.

Let's check the variance, lets assume 55% success, then 60%, then 45%.
45%: 73.40%
55%: 89.90%
60%: 94.57%

That variance is a bit high, but still a bit better than WOTC's standard system. However, your problem comes in when you try and change the complexity. For example, let's change the number to 5 successes in 2 rounds. That drops the win rate from 82.81% to 62.30%. The reason is with this system I can't change the success and failure number independently. The second I change the success number, I'm also changing the failure number.

Still, for a quick fix using that 4 successes in 2 rounds thing should work pretty well, I just wouldn't try to scale it beyond that.

 

[Fredrik Svanberg]

Thanks for looking at this.

With this system I wouldn't try to change complexity like that. I would stick with 4/2, 6/3, 8/4, etc, as the book says, and if I wanted to make a challenge harder or easier I would probably mess around with the DCs instead.

I can also award bonus rounds for creative skill use. Finding ways to buy the group some more time to finish the challenge would be very helpful, and I can think of a few dramatic ideas for it. E.g. in a combat the number of rounds wouldn't even have to be fixed to begin with but could depend on the fighter holding back enemies in a doorway while the rest of the party tries to solve the challenge. Have the whole party making strength checks to hold back the crushing walls while the rogue works on the mechanism to stop them. And so on. By throwing raw ability checks in the mix I can let characters who don't have a suitable skill but an impressive ability shine too - at an adjusted DC of course.

I think I'm going to try this out on my group whenever I get a chance to DM again. Your post about how the original system is broken will hopefully help convince them that something has to be done and I think my version will work better for me. If not, I'll try yours.

 

[Starfox]

I was going to post something much like this, but the boards swallowed my post.

Anyway, I like your proposed system. The standard system suffered because it made "aid other" almost a required activity if you didn't have a 75% chance to succeed. Which ran directly counter to the "activate all the players" idea of skill challenges.

Making this scale for more complex challenges needs more work, tough.

TORG had a Dramatic Skill Resolution System that was somewhat like Skill Challenges yet very different. I'm writing it out here in abbreviated form to give inspiration.

* You always had 10 rounds to complete the task. 10 rounds was a good static time in TORG, where rounds went by pretty quick; it might be so for D&D too,

* Dramatic Skill resolution had 4 required steps, and the situation changed as you completed each step. Say you wanted to escape a room filling with water in four easy steps. (Using 3.0 skill names because they come easier to me)

A: Reach up to the sluice controls (Climb, Strength (for lifting))
B: Open the drainage hole (Strength to lift the plug).
C: Close the sluice gates (Disable device). Guards start shooting through the now dry sluice gates.
D: Open the door once the water has drained out sufficiently (Strength with a decreasing difficulty for each round)

In this case, the timer would stop once you completed step C, replacing it with a threat of getting shot from above. For added fun, add a constrictor snake you have to fight in the pit.

 

[Fredrik Svanberg]

The scaling works fine if you just stick to the complexities in the DMG. It actually becomes slightly easier, on average, with higher complexity.

For complexity 1 using standard DC, a party of 5 using good skills (+9, 45% chance of success at DC 20) will get 4.25 successes in 2 rounds, on average. This is a level 1 party, I haven't checked at higher levels...

Complexity 2: 6.75 successes in 3 rounds.

Complexity 3: 9 successes in 4 rounds.

Complexity 4: 11.25 successes in 5 rounds.

Complexity 5: 13.5 successes in 6 rounds.

I feel that this margin in favour of the characters is nice to give parties that aren't perfectly optimized a break.

 

[Starfox]

The scaling works fine if you just stick to the complexities in the DMG. It actually becomes slightly easier, on average, with higher complexity...

It IS a little odd that it becomes easier to succeed at more complex tasks. I think this might work well in actual gameplay, tough. It is quite anticlimatic to fail at a long challenge, while failing a shrt one is not as drastic. We just have to adjust the level and xp rewards of skill challenges.

 

[Aoirorentsu]

Another reason why this system is good

You know, it occurs to me that with a time limit rather than a failure limit, the tactical choice is pretty dang cool.

Eg do I take fewer skill rolls, each of which is more likely to "hit"? (because of Aid Another)

OR do I take more skill rolls, each of which is a bit less likely to "hit"?


Sorry for not diving too hard into the math (probability was never my strong suit- curse you, Monty Haul!!!!!), just wanted to say I think it's pretty cool.

 

[Fredrik Svanberg]

Thanks, I think it's pretty cool too.

I won't have time to play any DnD for another three weeks now so if anyone tries out this system for real in a game I would love it if you posted about the experience here.

 

[Starfox]

I'm about to use this, but I'd still love a system where increasing the complexity actually increases the challenge. I'll think more on it.

 

[Starfox]

Proposed New Skill Challege System

In this system, the challenge needs an initiative to act upon. Generally, the challenge always acts last in the round. In some cases, where the challenge is based on a trap or character, it might have a regular initiative score, but this makes things harder by giving the players less time. It also introduces something called friction points; a mechanic used to determine when the time allowed for the challenge runs out.

Each round on the start of the challenge's turn, check to see if there is enough friction points to cause a failure. The number of friction points needed to fail is twice the number of successes the task requires. At the end of the challenge's turn, it gain a number of friction points equal to the number of players.

This gives the players in a 5-person party one extra round to complete a challenge. For larger and smaller parties, the time allotted varies because of rounding, but they get at least one extra round as well. This makes simpler challenges easier - it is more valuable to go from 2 to 3 rounds than it is to go from 6 to 7 rounds.

It is important that friction points be based on the number of players, not on the number of characters present. If the players split up, or if some characters opt not to participate, things get a lot harder. Everyone is encouraged to participate. Since your efforts never cause harm, any form of participation is helpful.

I did some math:

I have not calculated the exact probabilities (I am too lazy, but feel free to if you like), just the ratio of expected successes to required successes. If this ratio is over 1, your odds are good. The formula for this ratio is: [2C/(P(4C+n))], where P is the probability of success, n is the number of players and C is the complexity of the task. The break-even point (ratio 1) should give roughly a 50% chance of success. This break-even point requires that [P = 2C/(4C+n)]. This work out to the following minimum success probabilities, based on complexity:

5 players
C P
1 .31
2 .35
3 .38
4 .40
5 .41

3 players
1 .36
2 .4
3 .42
4 .43
5 .44

1 player (clearly a difficult one)
1 .44
2 .46
3 .47
4 .48
5 .48


Reservations on the math; I haven't done serious math in fifteen years.

This work out to these required skill levels to have a 50%chance of success at a moderate difficulty challenge in a party of five.

Complexity Minimum skill bonus
1 +1
2 +2 (just barely)
3 +2
4 +2
5 +3

Remember, a 50% chance to succeed is very, very bad. This, this gives us an absolute worst-case scenario: a skill challenge should never get harder than this. Players expect to succeed much more than that.

The chance to succeed goes down as the complexity increases, but only very slightly so. Then again, players are likely to be running out of ideas and resources, which should make a long challenge harder.

In a mixed encounter, NPCs can work at counter purposes to the players, adding friction points as they score successes. Make sure the total xp value of the NPCs and the skill challenge does not become excessive.