Upgrade your account to a Community Supporter account and remove most of the site ads.

Reply to thread

Message: <blockquote data-quote="Kraydak" data-source="post: 4384801" data-attributes="member: 12306">The math isn't trivial (college level statistics, and I could easily have made minor errors).  Would the following addition be useful to you?The question of the +5 to DC for skill checks is unrelated to skill challenges being broken as a system.  4e's skill challenge system can be thought of as having a target skill check successes to total skill check attempts ratio (in this case, 2/3).  If you plot the probability of the outcome, as the number of skill checks increases (in 4e's case, the complexity of the challenge), the result is a spike at the most likely result, that becomes higher and narrower as the number of skill checks increases.  Eventually (and it doesn't take long) the entire spike will be either above or below the targeted ratio, and you have virtually guaranteed success or failure at the challenge.You can imagine managing to "balance" the probabilities so that the spike lies almost exactly on the target ratio.  There are two (in reality identical) problems that makes this "solution" illusory.  Firstly, as the number of checks increases, the width of the spike becomes narrower and so "balancing" the challenges becomes harder.  Secondly, as the spike gets higher and narrower, the sides of the spike get steeper (what I referred to earlier as the gradient).  This has the effect of magnifying small changes in the individual check DCs.  Even very small DC modifications at the individual check level can move the spike substantially.  This magnification of small modifications makes Skill Challenges very hard to use because the effect of small modifications becomes unpredictable: keeping the spike "balanced" in actual play is functionally impossible.If you are having problems with the results of Skill Challenges being all over the place, it isn't your fault: the system is working against you.  The fix to the above is to move to a system that takes into account the fact that the range that the total number of successes falls into scales differently than the average total number of successes (for example, you might go from 4 +/- 2 to 9 +/- 3).</blockquote>

Verification