My Alt.Skill Challenge System

[Chris_Nightwing]

I have observed the disaster that is the current skill challenge system and I have observed the two systems proposed by Stalker0 (whose input I would appreciate should he see this). Then a thought struck me that one of the most excellent systems in 4th edition could be adapted for use in skill challenges: namely, the disease track.

The disease track is essentially a state-machine, where the probability of changing from one state to another (or staying the same in this case) can be calculated from a skill roll. This means the maths is very simple, as it can be analysed as a Markov Chain, hence summarised in a matrix and raised to an infinite power to measure the likelyhood of success or failure. I haven't finished playing around with the numbers (it's late here) but here's an outline:

The players start the challenge looking not for a particular number of successes or failures, but progression either up or down the skill track. If a skill check is failed, they progress down towards failure, if succeeded, they progress up towards success. A challenge shouldn't end on one bad roll, so there are always two stages below the starting position. Difficulty is determined by the DC of the tasks at each stage, not (much) by the number of consecutive successes required (see below for maths). Here is a very simple example, complexity 2:

Success! Win a gold piece, an xp and everyone likes you!
¦
You spot a nearby great big tree, which you think might be sturdy enough to get everyone across. Medium (DC 15) Athletics (to climb atop and use your weight to fell it), Endurance (to chop it down, hard work)
¦
You are confronted by a treacherous chasm (too wide to jump) which you must cross to continue your adventure. Easy (DC 10) Nature (chasms, eh?), Perception (look over there it's a-)
¦
You flounce around, throwing stones into the chasm out of boredom, maybe that wasn't such a good idea. Easy (DC 10) Endurance (to stay perfectly still), Acrobatics (to catch the bored PC's rock), Perception (to spot rocks tumbling up ahead)
¦
Failure - rocks fall.

Ok, so my example is a bit silly, but the general idea is there. Interestingly, if you run the maths then if there were say, 5 success stages before completion instead of 2 success stages, the overall chance of success is only slightly (<1%) reduced, unless you are near 50/50 every dice roll. There are also many ways to expand on the basic principle without destroying the challenge. You can add 'anchor' stages, where failure won't send you down a stage. You can modify the system so that you only move down a stage when you fail to reach the DC by 5 or more. You can vary the DC of checks at different stages (the earlier ones obviously impacting the overall chance of success the most, but then you can have difficult finales without hurting the party). Most importantly, the addition of a time limit is crucial, since these chains can go on forever, and when there are only four rounds to win, the party usually all get involved. I think a number of rounds equal to the complexity works out best, but again, I've only lightly run the numbers.

So, I guess I'd like to open myself up to being torn down, so to speak. Have I got anything wrong with my maths? Is the principle just awful? Are there any more interesting additions that could be made, or suggestions I've made that won't work? Is it too complicated/time-consuming/lame? Comments welcome !

 

[77IM]

This works for certain types of challenges. For example, a chase scene -- you're either pulling away from pursuers, or catching up with pursuees.

But in general, I don't think it works, because the closer you get to success the further you get from failure. Thus, the tension and drama actually decrease over time. There's no possibility of snatching victory from the jaws of defeat, and no fear of the opposite occurring. So as we get closer to victory, we slack off and it becomes a foregone conclusion, and as we get closer to defeat, things look hopeless and we give up.

One of the things about the core system (and Stalker0's system) that I like is that, depending on how the rolls go, you could simultaneously be very close to succeeding and very close to failing. This is similar to combat; it's possible that the players are all bloodied and starting to drop... but so are the monsters. Do you run or stand your ground? It's tense, and the skill challenge system should be the same way. I don't know how to achieve that with a disease-track system.

-- 77IM

 

[Chris_Nightwing]

I think the time-limit of the challenge is what adds to tension, without the need for absolute failure at the end of it. You may only get partway through the challenge, but you have achieved something - Stalker0's system includes partial success in the same way. If you want to add the make-or-break aspect then you can have critical stages, at which point all the previous 'positive' stages are forgotten and you're two bad rolls away from failure, one/two good rolls away from success. A modification to my anchor idea (which I realise doesn't work unless you want there to be no chance of failure, only a measure of success).

This system covers situations that don't include make-or-break tense moments too though. If you're conducting diplomacy then a bad roll has to represent such a bad gaffe that it counters your many excellently put debate points. You can also use this to track NPC attitude the long-term of a campaign, with or without skill checks to adjust how close to success/failure you are (completing a quest may be +1). I think it's nice to unite and formalise systems people already use for this.

I will try to flesh it out some more into a set of formal rules and then lay out the matrix maths to explain my assertion that the system is more robust than those that just count success/failure absolutely. I think it has potential!

 

[77IM]

A formalized time limit adds a lot. It basically produces two ways of failing: One that draws ever nearer (the time limit), and one that you can get away from (the bottom of the track). Better yet, combating one puts the other at risk: making your check can get you away from the bottom of the track but puts you closer to the time limit. In some ways, this is a rephrasing of Stalker0's system but using the track as a fine-grained way of specifying partial success.

-- 77IM

 

[Stalker0]

I would give you some feedback, but you need a few more numbers. Right now you have the basic outline. It makes plenty of sense, but without seeing any DCs or anything its hard to gauge how well it works.