But a lot of non-combat does matter.
"Say yes or roll the dice" is one thing, and tends to work fairy well. "Say no without a die roll" tends to be trickier, in a context where an important part of character building is assigning bonuses (via skill choice, stat allocation and boosting, feats etc) that are meant to be useful against DCs.
This doesn't strike me as special to 5e, though. It's an aspect of 2nd ed AD&D that I didn't like. After all, a common label for the GM creating the fiction that s/he wants to is "railroading"!
I've never played 3E and have GMed a handful of sessions of it. It's had no impact on my play preferences.
The trad systems that influence my conception of how out-of-combat is resolved are AD&D Oriental Adventures, Traveller, Runequest and most of all Rolemaster. The "modern" systems that have influence me in this respect are HeroWars/Quest, Burning Wheel, Marvel Heroic RP and 4e.
Five of these systems use "objective", "world set" DCs: AD&D, Traveller, RQ, RM and BW. The others use "subjective" DCs (this is true at least for HQ in its revised version) - but the "subjective" DCs are not set by GM fiat, but by reference to encounter difficulty procedures (MHRP has probably the most complex version of this, in the form of the Doom Pool). The aspiration of these systems is that player resources interact with the GM-side mechanics to produce dynamic fiction that is not under the direct control of either GM or the players.
"Subjective" DCs with no constraints on the GM side is not something I've ever encountered, outside of the context of a certain approach to 2nd ed AD&D. And "rules lawyering" - a phrase that I first came across before 2nd ed was published - has no direct bearing on the working of either sort of system, at least in my experience.
But there
is guidelines. A
lot of them, and the extremely loose guidelines outside combat was always one of the main critics of 4th edition, and one of the things that the fanbase has always defended. It would be very naive to call purely DM fiat the DCs of certain tasks in 5th edition. The DC settings, because Bounded Accuracy and linear, non exponential growth of power, are fixed to the world and
not levels, hence the difficulties chart:
5 for very easy task, (75% of probability without being trained),
10 for easy tasks (50% chances without being trained),
15 for average (25% chances whitout training),
20 for hard (5% chance without training),
25 for very hard (0% chance with or without training if you have not a significant Stat Mod -at least +3 with training), 30 for nearly impossible (0% chance if you have not Expertise, magical aid, and a significant modifier -at least for a level 4 character. With Expertise -+4-, exceptional stat -+4/+5- and Guidance -1d4- you have a slight chance, but you are an exceptional expert magically aided).
So, for example, a 20th level 12 Strenght Wizard cannot perform any Very Hard feature of Athletics unless is trained (only achievable via Sailor/Soldier background, variant human or months of training outside adventures, all of them reasonable enough) -like swimming against the flow-, unless he handwaves it with a spell (that always have a limited duration), while a 1st level, 18/20 str half orc champion with athletics (via class features, you probably choose that to enhance your strenghts) has between 10/15% chance (better than a hard situation for the same wizard), and that same 20th level warrior has a 55% chance, better than an Easy task to a non proficient, non exceptionally Stat. Also:
DIFFICULTY CLASS
It's your job to establish the Difficulty Class for an ability check or a saving throw when a rule or an adventure doesn't give you one. Sometimes you'll even want to change such established DCs. When you do so, think of how difficult a task is and then pick the associated DC from the Typical DCs table.
TYPICAL DCs
Task DC Task DC
Very easy 5 Hard 20
Easy 10 Very hard 25
Moderate 15 Nearly impossible 30
The numbers associated with these categories of difficulty are meant to be easy to keep in your head, so that you don't have to refer to this book every time you decide on a DC. Here are some tips for using DC categories at the gaming table.
If you've decided that an ability check is called for, then most likely the task at hand isn't a very easy one. Most people can accomplish a DC 5 task with little chance of failure. Unless circumstances are unusual, let characters succeed at such a task without making a check.
Then ask yourself, "Is this task's difficulty easy, moderate, or hard?" If the only DCs you ever use are 10, 15, and 20, your game will run just fine. Keep in mind that a character with a 10 in the associated ability and no proficiency will succeed at an easy task around 50 percent of the time. A moderate task requires a higher score or proficiency for success, whereas a hard task typically requires both. A big dose of luck with the d20 also doesn't hurt.
If you find yourself thinking, "This task is especially hard," you can use a higher DC, but do so with caution and consider the level of the characters. A DC 25 task is very hard for low-level characters to accomplish, but becomes more reasonable after 10th level or so. A DC 30 check is nearly impossible for most low-level characters. A 20th-level character with proficiency and -relevant ability score of 20 still needs a 19 or 20 on the die roll to succeed at a task of this difficulty.
VARIANT: AUTOMATIC SUCCESS
Sometimes the randomness of a d20 roll leads to ludicrous results. Let's say a door requires a successful DC 15 Strength check to be battered down. A fighter with a Strength of 20 might helplessly flail against the door because of bad die rolls. Meanwhile, the rogue with a Strength of 10 rolls a 20 and knocks the door from its hinges.
If such results bother you, consider allowing automatic success on certain checks. Under this optional rule, a character automatically succeeds on any ability check with a DC less than or equal to the relevant ability score minus 5. So in the example above, the fighter would automatically kick in the door. This rule doesn't apply to contests, saving throws, or attack rolls.
Having proficiency with a skill or tool can also grant automatic success. If a character's proficiency bonus applies to his or her ability check, the character automatically succeeds if the DC is 10 or less. If that character is 11th level or higher, the check succeeds if the DC is 15 or less.
The downside of this whole approach is its predictability. For example, once a character's ability score reaches 20, checks of DC 15 and lower using that ability become automatic successes. Smart players will then always match the character with the highest ability score against any given check. If you want some risk of failure, you need to set higher DCs. Doing this, though, can aggravate the problem you're trying to solve: higher DCs require higher die rolls, and thus rely even more on luck.
The Variant rule keeps the game in the level that you want. It's up to the DM to set the difficulty of a task, or to the module/ adventure you seek. It's expectable that the DM isn't a stupid or evil person, but a person who try to fit the expectations of his group. But there is a guideline in a trap of how the difficulty is set:
PITS
Mechanical trap
Four basic pit traps are presented here.
Simple Pit. A simple pit trap is a hole dug in the ground. The hole is covered by a large cloth anchored on the pit's edge and camouflaged with dirt and debris. The DC to spot the pit is 10. Anyone stepping on the cloth falls through and pulls the cloth down into the pit, taking damage based on the pit's depth (usually 10 feet, but some pits are deeper).
Hidden Pit. This pit has a cover constructed from material identical to the floor around it. A successful DC 15 Wisdom (Perception) check discerns an absence of foot traffic over the section of floor that forms the pit's cover. A successful DC 15 Intelligence (Investigation) check is necessary to confirm that the trapped section of floor is actually the cover of a pit.
When a creature steps on the cover, it swings open like a trapdoor, causing the intruder to spill into the pit below. The pit is usually 10 or 20 feet deep but can be deeper. Once the pit trap is detected, an iron spike or similar object can be wedged between the pit's cover and the surrounding floor in such a way as to prevent the cover from opening, thereby making it safe to cross. The cover can also be magically held shut using the arcane lock spell or similar magic.
Locking Pit. This pit trap is identical to a hidden pit trap, with one key exception: the trap door that covers the pit is spring-loaded. After a creature falls into the pit, the cover snaps shut to trap its victim inside. A successful DC 20 Strength check is necessary to pry the cover open. The cover can also be smashed open (determine the cover's statistics using the guidelines in chapter 8). A character in the pit can also attempt to disable the spring mechanism from the inside with a DC 15 Dexterity check using thieves' tools, provided that the mechanism can be reached and the character can see.
In some cases, a mechanism (usually hidden behind a secret door nearby) opens the pit.
Spiked Pit. This pit trap is a simple, hidden, or locking pit trap with sharpened wooden or iron spikes at the bottom. A creature falling into the pit takes 11 (2d10) piercing damage from the spikes, in addition to any falling damage. Even nastier versions have poison smeared on the spikes. In that case, anyone taking piercing damage from the spikes must also make a DC 13 Constitution saving throw, taking an 22 (4d10) poison damage on a failed save, or half as much damage on a successful one.
Under this rules, the difficulty isn't set
by level, but by means of who can make the things needed, and level is only significant if you already have some training. Level matters in the chance you have to accomplish certain tasks. And with the variant rule of 5, the level 20 warrior can automaticly succeed where the wizard with or without spells can't, and vice versa: the warrior (unless trained in Arcana, has significatively high Intelligence, and magical aid never could identify a spell in the lips of a rival) don't.
There is some niche protection, but the protection isn't graved in stone, so class & level are not the only factors that comes around in the resolution of a certain task. You can also take account on the Difficulty Class between Hide and the general, untrained Passive Perception, like a peasant (PP 10- easy task), a Guard (PP 12- in between the Average and easy task) or a trained 1st level ranger (PP 14- almost average difficulty). Passing a guarding dog is arguably difficultier, as he has Advantage on Hearing and Smell checks (14 +5 to their Passive Perception if it doesn't rely on sight, but hearing, almost the Hard check, only 10% chance if the hiding character isn't trained nor exceptionally agile).
As you can see, the DCs are world-given, and the characters will have a better or lesser chance of success given not only his level, but their stats and training. This keeps the minor modifiers relevant, as it isn't the same to have a +1 or don't have it.
But at this time, I sense that you have fear to give
any control to the DM. There are loose guidelines, yes and the DM always have the ultimate word, but the DM is not an evil dictator, nor it's a stupid person: is the narrator and always keep the best interest of the table, and presumably has common sense to determine if anything is hard or not, and the players have their modifiers and abilities to surpass this. If you don't trust your DM, why play any game with him? Why don't give him a published adventure if he is learning to determine difficulties? That way, the DCs of the several tasks are going to be
in the module. And there are numerous examples of difficulty sets among the books, and plenty of unofficial DM screens to aid setting the DC.
But they are flexible enough to give some liberty to the DM. This have been said countless times: the rules can't possibly consider every single aspect of the game, and it's a waste of time to list every possible example of DCs, because A) it would be a pain in the ass, and B) there is no chance to cover every single possibility.