D&D 4E Showing the Math: Proving that 4e’s Skill Challenge system is broken (math heavy)

[Spatula]

That was my point. All the math and interpretations of books aside, the conclusions made by some people in this thread have two logical results: either several months of detailed design and playtesting of one of the most innovative and important mechanical elements of 4E is obviously flawed, or the people making conclusions are grossly misinterpreting things.

...
That is one reason I am more inclined to ditch the +5 that is being mentioned here, simply because removing it creates a reasonable result.

So, taking what's written in black and white on the printed page and using it as written = gross misinterpretation. Gotcha.

I fully expect that there might have been some typos, or stuff accidentally left out the book. Regardless, people who buy the DMG only have what's written there to go by, and what's written there is extremely counter-intuitive (which is putting it charitably). If there's some hidden, mysterious bonus that we're supposed to be adding to the player's skill checks, the DMG does not see fit to point it out to the reader.

And if you're supposed to ditch the +5, no one told Bill Slavicsek that before he wrote the KotS web enhancement.

 

[Spatula]

Using aid another to help someone else's check does not prevent you making your own skill checks. I think that's where our disagreement lies; if each aid another counted as a success or failure, then I'd completely agree with you. But they don't, and you can roll as many d20s as you ike in the course of a skill challenge as long as you don't don't rack up too many failures. the importan tpoint is that the aid another doesn't rack up a success or a failure, so is simply an additional action.

Well, I'm curious as to what you're basing that on. Everyone is supposed to roll initiative and, when their turn comes around, make a skill check based on what their character is doing. Aid another, by my reading of the PHB, is a normal use of a skill, not a free action or anything.

 

[Tervin]

That is one reason I am more inclined to ditch the +5 that is being mentioned here, simply because removing it creates a reasonable result. Regardless of whether or not some line of rules text implies it should be included or not, if it works without the +5, then that is probably the way it was intended to be done. However, if there is some other way to modify the results, then those should be factored in as well. Morrus mentioned something about racial stats, skill kits, small boosts (like the Elf's boost to team Perception), and such.

Also, another thing to consider is that, since there are rules for Easy checks, then there must be some way for a player to take advantage of that, even for just a single check in a whole challenge. Would a mix of Easy, Normal, and Hard checks affect the math, compared to nothing but Normal checks? It seems like it would affect it greatly, especially for very complex challenges, since the Easy or Hard modifier might push the success rate above or below the line where added complexity (and thus the law of large numbers) favors or hurts the PCs, and the mix would do so in ways the current models might not predict.

If you use any of the tools made by people in this thread you will note that the problem is only partly with challenges seeming to be too difficult if run like in the examples offered in the rules. Even if you fix that (which is easy enough) you have another problem left, which can't be shrugged off so easily.

A complexity 5 challenge is supposed to give 5 times as much XP as a complexity 1 challenge. However, if you use managable difficulty levels a complexity 5 challenge will actually end up quite a bit easier than a complexity 1 challenge. This is not a problem that has to do with us reading the difficulties wrong or forgetting about skill bonuses. It is a problem of bad choice of math in the underlying system.

I will still try to run the game with these systems, but knowing that the system is flawed, I will put focus on player creativity being needed to get through the challenges. That way complex challenges need a lot more creativity, and I don't feel so bad about giving out XP. And yes, that is something that the DMG section on skill challenges could have included - but it didn't. As it is it feels like the DM needs to have all the math info from this thread and test out where to put DCs beforehand in order to get a satisfying experience.

I like 4th Edition a lot, but that doesn't mean I can't see flaws in it. For some reason it seems all good RPGs have to have one or two subsystems using really bad math for simulations. (Storyteller comes to mind...)

 

[Ulthwithian]

Well, on p.74, it lists in narrative form your options during a skill challenge, it seems. Either primary skill, secondary skill (once), or Aid Another. I perfectly agree that if you fail an Aid Another check, it doesn't increment the failure count. OTOH, this is what people seem to be envisioning:

Player 1: I Aid Another Player 4's Diplomacy check.
Player 2: I Aid Another Player 4's Diplomacy check.
Player 3: [same]
Player 4: I make a Diplomacy check.

In Initiative order. Is it your argument that it should be:

Player 1: I make an x check.
Players 2,3,4: I Aid Another him
Player 2: I make a y check.
Players 1,3,4: I Aid Another him.
Player 3: I make a z check.
Players 1,2,4: I Aid Another him.
Player 4: I make a Diplomacy check.
Players 1,2,3: I Aid Another him.

Let's check the implications of this. If the average player has, say, a +2 to his un-trained skills, then he has a 65% chance of succeeding, and thus the expected number of successes from 4 people doing Aid Another is 0.65 * 4, or 2.6. So you would get an average of a 5.2 point bonus on your skill check from Aid Another checks... which offsets the +5 DC penalty. Hm. Personally, I would prefer if you simply drop the 5 DC increase in that event and stress individual actions.

Both ways have their drawbacks. If Aid Another takes your action, but doesn't contribute successes or failures inherently, you have, in essence, a long but essentially solo skill check system, at an extreme. However, if it doesn't take your action, and you do not allow players to 'do nothing', then you have a better chance of succeeding at any given skill check but the distinctness of the characters blurs almost to the point of nonexistance. I'm currently thinking about how to resolve this issue.

 

[Morrus]

Well, I'm curious as to what you're basing that on. Everyone is supposed to roll initiative and, when their turn comes around, make a skill check based on what their character is doing. Aid another, by my reading of the PHB, is a normal use of a skill, not a free action or anything.

Sure, but you just do something the next round then. I guess in some time-critical challenges it might matter (thus a time-critical challenge would inherently be more difficult- but I guess that's what time-critical things are like); but, hey, climb the cliff a bit more slowly, or take your time about persuading the duke. If there's no expiry date on the challenge, what difference does it make?

 

[Morrus]

Well, I'm curious as to what you're basing that on. Everyone is supposed to roll initiative and, when their turn comes around, make a skill check based on what their character is doing. Aid another, by my reading of the PHB, is a normal use of a skill, not a free action or anything.

No, but there's no limit to the number of actions you can take to resolve a skill challenge. So do 'em one at a time instead of simulataneously. Not everyone has to do everything in the same round - X does something, aided by others; then Y does something aided by A and B while Z is doing something else on his own. Then A and B both do their own thing, B aided by X.

Like I said above, the only time this would be a problem is if there was a time-critical element ot the challenge which, in essence, is one way of making it harder.

 

[silentounce]

That was my point. All the math and interpretations of books aside, the conclusions made by some people in this thread have two logical results: either several months of detailed design and playtesting of one of the most innovative and important mechanical elements of 4E is obviously flawed, or the people making conclusions are grossly misinterpreting things.

I will be blunt here. It is a LOT easier for me to believe that a few fans running the numbers for fun have made a mistake than professional game developers who devoted a lot of energy making a product they intend to sell.

Considering that I know there are skilled mathematicians at WotC (pretty much the entire development team exists for this one purpose), there is no way to argue that this is a matter of fans knowing more about math than the WotC guys (particularly since the math behind this is so simple).

I don't have the 4E books, so I can't look at the rules and numbers myself, but I am going to say that all of the people making bold claims about WotC's failings should take a minute to re-evaluate their assumptions and try to interpret the RAW in a different light. Instead of interpreting the text as you think it is supposed to be read and reaching a broken result, try to find a way to get a good result out of the math without changing much and see how that affects the interpretation of the text. If nothing else, it is more constructive and useful to your fellow gamers to see how to modify the RAW to get a good result without scrapping the whole system and creating a new one from scratch.

That is one reason I am more inclined to ditch the +5 that is being mentioned here, simply because removing it creates a reasonable result. Regardless of whether or not some line of rules text implies it should be included or not, if it works without the +5, then that is probably the way it was intended to be done. However, if there is some other way to modify the results, then those should be factored in as well. Morrus mentioned something about racial stats, skill kits, small boosts (like the Elf's boost to team Perception), and such.

Also, another thing to consider is that, since there are rules for Easy checks, then there must be some way for a player to take advantage of that, even for just a single check in a whole challenge. Would a mix of Easy, Normal, and Hard checks affect the math, compared to nothing but Normal checks? It seems like it would affect it greatly, especially for very complex challenges, since the Easy or Hard modifier might push the success rate above or below the line where added complexity (and thus the law of large numbers) favors or hurts the PCs, and the mix would do so in ways the current models might not predict.

You know, there are professional game designers and skilled mathematicians here, too. And just because something is easier for you to believe, doesn't make it the truth. I was going to cut you some slack, but by your own admission you don't even have the books and haven't even read the relevant sections. If you had stopped after your fourth paragraph you would have been fine. But everything you say after that is pretty much due to you not having read the book so you don't know better. Especially, your last paragraph. Because all of the things you mention in it do, in fact, exist in the system as written, and it still doesn't work right.

Really, I get your point. And if there was an easy fix, I'm sure everyone would be all over it. Also, the system that the OP created is by no means a scrapping of the entire concept at all. Yeah, coming in here and bashing the creators isn't constructive. But the majority of this thread isn't about bashing the creators, it's about trying to find a fix for the problem and trying to understand why they did the things they did.

I read the sections in the DMG several times before I decided to come into this post and make my own suggestions/comments. I don't understand how someone can come in and suggest something without having read it all. And to use your own words against you, it seems to me like you are making bold claims about the posters in this thread and the rules you haven't read yet, maybe you should should take a minute to re-evaluate your assumptions.

Sorry if I offended, but I didn't much care for the tone of your comment, even if I imagined it.

 

[Aegir]

I honestly haven't read the entirety of this thread, but to my knowledge, this hasn't been brought up yet. There are two things that, when taken together, suggest that the assumption ("For skill checks: increase DCs by 5") is the main flaw in this entire argument.

DMG, p73: "Set a level for the challenge and DCs for the checks involved. As a starting point (...) use moderate DCs for the skill checks (see the Difficulty Class and Damage by Level table on page 42)."

Keep on the Shadowfell, p52 (Talking with Keegan): "In order to change his mind, the PCs have to converse with him and, in the process of doing so, succeed on four separate DC 15 skill checks before they fail four such checks."

Now, for an assumption: this encounter was in fact designed using their own rules for Skill Challenges. If this is the case, then the call for DC 15 checks is in fact a "moderate" challenge, which places it squarely in the 10/15/20 range that the p42 table presents, without the +5 applied.

Its been mentioned before in this thread: whats the more likely scenario? That the 4E designers let something this horribly glaring slip by and published a system - a system they themselves called one of the major innovations of 4E - that was so flawed that PCs had virtually zero chance of success, or that we as a community are simply misinterpreting that the +5 comment applies to Skill Challenges, when it in fact only applies to singular skill checks?

Given the fact that the removal of the +5 puts things far more in line with the success range of the rest of the system, I'm personally assuming the latter.

 

[Morrus]

I honestly haven't read the entirety of this thread, but to my knowledge, this hasn't been brought up yet. There are two things that, when taken together, suggest that the assumption ("For skill checks: increase DCs by 5") is the main flaw in this entire argument.

DMG, p73: "Set a level for the challenge and DCs for the checks involved. As a starting point (...) use moderate DCs for the skill checks (see the Difficulty Class and Damage by Level table on page 42)."

Keep on the Shadowfell, p52 (Talking with Keegan): "In order to change his mind, the PCs have to converse with him and, in the process of doing so, succeed on four separate DC 15 skill checks before they fail four such checks."

Now, for an assumption: this encounter was in fact designed using their own rules for Skill Challenges. If this is the case, then the call for DC 15 checks is in fact a "moderate" challenge, which places it squarely in the 10/15/20 range that the p42 table presents, without the +5 applied.

Its been mentioned before in this thread: whats the more likely scenario? That the 4E designers let something this horribly glaring slip by and published a system - a system they themselves called one of the major innovations of 4E - that was so flawed that PCs had virtually zero chance of success, or that we as a community are simply misinterpreting that the +5 comment applies to Skill Challenges, when it in fact only applies to singular skill checks?

Given the fact that the removal of the +5 puts things far more in line with the success range of the rest of the system, I'm personally assuming the latter.

It can certainly be read both ways (which is why I haven't weighed in on the +5 - it really is ambiguous), but the example certainly does nt include the +5.

I myself would have naturally interpreted that as "without the +5", until poeple here suggested otherwise and put the seed of doubt in my mind.

 

[The Hitcher]

I've started my own thread on this topic on the House Rules forum:

http://www.enworld.org/showthread.php?p=4283258#post4283258

Two things seem clear to me:

1. Succeeding an average Skill Challenge should be about as hard as succeeding an average Skill Check. The only difference should be in the complexity of the system.

2. The complexity of the task should be irrelevant to the difficulty of the challenge. That's what DC is for.

Therefore, all Skill challenges should have a success/failure ration of 1:1. Then it is only the DCs that matter as a measure of difficulty. Complexity is only a measure of complexity. Problem solved.

The other other change necessary is to tie XP rewards to DC, rather than Complexity rating.