While I'm not sure that I agree on the bottlenecks, the discussion is probably tangential enough to deserve it's own thread. In short: while I agree bottlenecks and railroads are best avoided where possible, I do think that good and even great DMs sometimes make use of them. Not everyone is good with spontaneity, so there will be times when they have to resort back to what they have prepared for, or the session comes to a screeching halt.
There's also the simple fact that just because players choose to pursue a particular option (negotiating, for instance), doesn't mean it's always going to work, even if the DM was perfectly willing to give them a reasonable chance at success.
"Focused more" is quite vague. One player can hear that as "this is what we'll spend all of our time doing", while another player hears it as "we'll do this a bit more often than the other things". The DM might have meant something altogether.
Campaigns, and sessions within them, can quite easily drift away from where people originally intended them to go. Especially, if, as you say, they avoid bottlenecks and railroads.
I also categorically reject the idea that paladins and bards should be strictly better than fighters and barbarians in social interaction play. Different, yes. Better, not automatically.
The fighter class has to cover a lot of ground, to be sure, but a couple of the areas it is quite frequently expected to cover are knights and samurai. Both of which were expected to have some degree of culture, and be able to comport themselves well in society. Forcing fighters to be socially inept because they wanted to be "a knight without being tied to religion" is not something I'm terribly fond of.
Should a bard or a paladin be better at comporting themselves in a social situation that takes place in high society than a barbarian is? Certainly. What about a social situation in a tribal culture? What about negotiating with other "uncivilized" people. Surely the barbarian has a better grasp of how to negotiate with that above-mentioned orc chieftain than someone who expects to have the right cutlery at the diplomatic reception?
I definitely agree that viable and equal are not the same thing. They can be, but only if one prefers boring and flavourless solutions, which neither of us are arguing for.
Math-wise, I think I prefer a narrower range than what you're suggesting. However, the math misses the point of what I'm trying to say.
I don't accept that all social situations or all exploration situations or all combat situations fall into the same category, and thus don't feel any class should be or even could be described as universally "better" in one pillar than another.
Why SHOULD a fighter have a higher rate of success in combat than a wizard?
A wizard didn't learn to throw massive fireballs at people because it's a great trick at parties, or a decent way to start a campfire. He learned it to be good in combat. A rather significant percentage of the traditional spell-list is made up of things that clearly are intended for combat usage.
A fighter should be better at doing fighter things in combat than a wizard. The wizard should be better at doing wizard things in combat than a fighter. The same holds true in the other pillars as well.
A barbarian might have rage issues, and come from a tribal society. That doesn't mean they're automatically non-verbal and asocial. They're from a different culture, they're not sub-human cavemen. There should be situations in the social pillar where they have advantages, especially when it comes to interacting with others of the same or similar cultures. They should even have the occasional trick up their sleeve that makes them effective in the situations where yes, a bard would be an advantage most of the time. Sometimes, only outsiders can spot a flaw in an arrangement that others have come to accept as normal.
Success levels can and should be measurably different. However, not all situations in a pillar should be lumped into the same category of "Class A is good at ALL of this pillar, Class B stinks at all of this pillar".
A wilderness-based ranger is better at tracking people in the wild than pretty much anyone else. Put him in a city, and why wouldn't someone else be better off? Why not the rogue, who knows the place like the back of his hand, and has learned to spot who doesn't fit in from learning to spot guards when choosing marks? Why not the bard, who has an easy in at all the local taverns, and can pump the crowd for rumours?
Even out in the wild, a ranger might be better at finding and following tracks, but why shouldn't the fighter (or warlord, or both) be better at determining details like force size and composition from the tracks that are found?
Every class can and should have reasonable strengths and weakness across all pillars.
Writing a class name on my sheet because I like some aspect of the flavour of a class shouldn't commit me to being less relevant than the next player, regardless of the mix of pillars.
There's no reason except tradition that says fighters have to be socially inept unless they sacrifice combat proficiency for it. Tradition is rarely a sufficient justification for anything, in my books. Especially not when other options exist that still allow for the traditional approach if people opt into it, instead of being forced into it.
There's no reason that wizards should have to be bad at combat (and they never have been other than at low levels) in exchange for being good at exploration and decent at social scenarios.
As I've said before, I have absolutely no issue with those who want to play the socially inept characters, or the combat inept characters, or the exploration inept characters. Just don't make me have to be one of them.
