Lyxen
Great Old One
Poor question for game design: we don't "need" anything. The real question is, "How is a formal structure useful?" And the answer is, "
It might be useful, or it might be cumbersome and unneeded.
Why not? Skill challenges as a framework do not have a single built-in time scale. You could quite easily have a skill challenge that factors in things like travelling between locations, spending resources now to avoid rolling (or compensate for bad rolls) vs. saving them for later in that challenge or other challenges down the road, or leveraging past successes (or failures!) to shape the future points of difficulty or conflict.
In fact, I could absolutely see something like a month of warfare being structured as a Skill Challenge--including elements like battle strategy (a History roll), foraging off the land to keep up your non-perishable supplies (Nature), keeping troop morale up with religious ceremonies (Religion) or entertainment (Diplomacy), disguising the movements of your forces or distracting the enemy with false efforts (Bluff or Stealth, depending), executing feints or ballsy gambits like Hannibal executing a pincer maneuver on a larger force during the Battle of Cannae (again, Bluff or Stealth), or various other valid paths--and each decision would rationally build on the previous, changing the theater of war and shifting local opinions, the opinions of civilian leadership backing the PCs and their army/armies, and overall troop numbers and morale/fitness.
Yes, you could do something like that with no formal system whatsoever. But with a formal system, it becomes possible to seek leverage, which almost always adds interest to the process. Further, a formal system can create both dread and excitement as the final result nears, while keeping things clearly "fair" because the fundamental rules-objective was clear and defined in advance. (Plus, let's be honest, it's not like the Skill Challenge system is THAT formal! It's literally just, "get X successes before Y failures," any further embellishments being something players did to improve it, not the underlying system.)
This does not take into account the quality of the successes and failures, the drama and the story, and the way the players react and participate to it. So either you make the count visible and I can guarantee that the players will just try to use skills to increase the count of successes, or you hide it, and in that case it really serves no purpose except to constrain the DM into declaring an overall failure or success when the count reaches a pre-determined value.
As a DM, I don't want the players to hunt successes and skill rolls, and I value the story and my players reaction to it more than counting things arbitrarily because when I created the scenario I certainly did not envision all the ways to succeed and fail. So why bind myself arbitrarily to these?
Again, it's a question of playing style mostly, I'm not denigrating another way of playing, just explaining the one that we prefer. And it's linked to preparation, for example I usually don't try and find the ways by which the PCs might succeed. I present a situation, just trying to make it so that it does not look impossible to get out of, and then I go with the flow and the PCs ideas.
This avoid any sort of railroading towards pre-determined solutions - and again I'm not accusing anyone of railroading or saying that a bit is bad or not, just the fact that listing the solutions and where the PCs get points is already creating some structure that will guide the results.
Okay. Do you think D&D should be designed primarily to get out of the way of experienced players as much as possible, or primarily to help and support new players as much as possible?
Well, please consider the fact that 5e, which does not have these formal structures, has been what, ten times more successful than 4e ever was in introducing new players to the hobby.
For me, it's because it's easy to play overall, and does not have much formalism or actually even a jargon and technical vocabulary.
So for me, there are better ways to support new players.
Because I think it's pretty trivial for very experienced players to look at the system and say, "Nah, that's not how I want to do things." On the flipside, it's often very hard for a band-new, inexperienced player (especially if they're DMing) to get a foothold. Supporting new players, and especially new DMs, to the utmost is critical for keeping the hobby alive and sustainable. We cannot rely solely on veteran players bringing new players into the fold; we must specifically support groups made up of exclusively new folks, even those who have never played at all before but now must step into the DM's role.
See above.
What do Skill Challenges have to do with using a grid? The two are unrelated. You can have a grid situation (whether or not it's combat) without having a Skill Challenge, and the sizable majority (I'd say 2/3 or more) of SCs I've done as a 4e player had nothing to do with running on a grid. Many were about managing crowds, persuading an important NPC, securing a resource, stopping/mitigating a disaster, or infiltrating(/exfiltrating) a location. Some were precipitated by finishing a combat, others set the stage for a subsequent combat, most had nothing to do with combat at all. Only those that specifically occurred during combat itself actually used a grid, and those were uncommon (usually having to deal with traps or environmental hazards during the fight).
It was just a tangent, to say that although it can help some people, formalising things (skill challenges, grid) has a cost in terms of ease of use of the system, that's all.