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

Reply to thread

Message: <blockquote data-quote="clearstream" data-source="post: 7209646" data-attributes="member: 71699">Possibly it could help to hear what the figuring out is about. Through having targets we can tune toward our intent. So a statement such as "I want to see no more than one death per 10 character-engagements in deadly encounters is useful because when we're running such encounters we know what we're looking out for. In the same vein, considering the cumulative probabilities lets us understand what the long-term consequences are of such targets. For me, they suggest that revivification magic is potentially underutilised - in part because deaths are very infrequent. If we're increasing the game difficulty that's going to make deaths more frequent and the cumulative probabilities suggest we may want to allow such spells to be used more. How much more? Again, we can set goals for that and know that - relative to our framing statement - they are likely to turn out to make sense.Essentially, the maths is done to make the empirical testing more efficient, by homing in on targets that are more likely about right. In estimation "accuracy" is always relative to your goals and contrasted with "precision". The calculating is accurate enough provided we remember that the apparently precise values are approximations or suggested centering values.Agreed, my sense is that it is efficient to modify monster CRs downward, but that cannot go very far due to the propensity to one-shot, which is warned about in the DMG of course. I'd suggest a range of -1 to -3 CR steps.Unless we are after all deadly, all alpha combats. And narrative purposes only easy and medium encounters i.e. not mechanically meaningful. The adventuring day matters.We need to respect a difference in the cumulative lethality of one deadly versus N warm-up fights. The lethality of the latter needs to be very far below the lethality of the former, because they exponentiate. X^1 vs Y^N.</blockquote>

Verification