Reply to thread

Message: <blockquote data-quote="Stalker0" data-source="post: 4285013" data-attributes="member: 5889">Let me first say can we all calm down a little. I realize that anytime you challenge an established body and say its wrong, that's going to cause tension, but we can all discuss this rationally. Let me address some points:1) You cannot determine if a system is broken based on pure math, you need playtesting and experience.In general, I wholeheartidely agree with this. In fact, normally I champion it. I've done a lot of houserule tweaks in the past, and I have learned the painful lesson that some times theory and reality don't mix.So...why not this time right? How come after a few calculations (and by few I mean several thousand, and no I'm not kidding) I'm ready to call it and say the system is broken?The reason is because the math is so very low. If I had run my calculations and found that a party had a 50% win rate, or heck even 40%, I would have been very skeptical of the system, but I would have given it a shot. I would have thought, maybe aid another makes a big difference, or maybe those utility powers really add up, etc. But I didn't find 40%, I found 7%. So I started investigating mathematically, and I didn't just find bad win rates, I found incredible variation with different skill modifiers, and an inversion in complexity, I found strong problems with the system. And I thought I could do better. That's why I started the skill challenge system in the house rules thread, and why I alerted the community to this problem.2) You can't model the system with pure math.Actually, 4e has made this easier than ever before. In 3e, you had to account for a wide array of variables, numbers, conditions, etc etc. 4e has fulfilled its promise of streamlining the math, and as a result the system is much easier to model mathematically. Of course I can't account for every thing a player might do under the sun, but I can certainly account for an average party.3) When the math was done, was the fact that some people will have over 50% and less than 50% taken into account?The answer is yes. I've run countless scenarios with mixed party skill bonuses. Some I do in depth, meaning I run all of the actual numbers. However, in many cases I will run an averaged value. This is important, because my averaged values actually give better results than the real ones.For example, take 3 people with 50% to succeed. Then take a party with a 60%, 50%, and 40%. The average of both parties is 50%, but in actuality the 1st group has a better chance to succeed. So when I say a group at 50% will have a 7% chance to succeed, that means that if that group has 1 person at 60% and another at 40%, the chance is even LOWER than 7%!4) The skill challenge system will work if you use aid another all the time, or use these utility powers, or X Y and Z.If you will look on page 3 of this thread, I already addressed one of the utility powers, as well as a skill bumping item (for those who don't want to look back, the answer is those things help, but not near enough) But here's the thing, if the system requires skill utility powers and everyone have +X bonus from a magic item to their skills, then the system doesn't work, because not everyone will have those utility powers, and won't have those bonuses.Others are saying aid another is the key. If that's true, then WOTC needs to make that clearer. If WOTC came in and said, "guys we made this system assuming everyone would aid one person every round" and those numbers worked out, then I would throw up my hands and agree that the system works. I wouldn't like such a system, but I would agree that the system does what they intended it to do.But what the poor DM who just wants to give his party a level 1 complexity 5 challenge, and is walking into a 93% failure chance for his party? Should he be forced to tell his party, everyone pick one guy to roll for real, the rest of you aid, or you will fail this challenge? If that's the case, then what's the point of the skill challenge system in the first place?</blockquote>

Verification