There's something here to be said about LBA (Level Based Advancement) and why that system is important. The "game" of D&D is getting our squishy 1st-level characters as far up the "level mountain" as possible. If you've played BECMI and AD&D, you understand this aint easy. But, it can be the best fun any game has ever offered, to include the sense of reward that comes from EARNING 36 levels of experience (RC).
Makes sense! Thanks for responding!
My reply is that I think you're conceiving of playing a campaign where you A) intend to go from 1st - 36th level and B) things like picking locks and opening the doors will be both equally challenging and equally important throughout the campaign this makes sense. C) You're using the base rules provided without making significant changes. I tend not to see games like this often, and want rules/procedures that don't only work for this sort of nostalgically idealized Gygaxian play.
In my experience it takes about a solid year of playing weekly to get PCs to around 5th level, at least accounting for deaths and replacement PCs, and by that time, or sooner, most players will either have dropped out due to living grown up lives (job changes, relationship changes, having babies, developing a crippling alcohol or powerlifting addiction etc.) or simply want to try something else. Longest game I ran on G+ lasted five years off and on and had two 7th level PCs and a number of 3rd - 5th level ones. Longest campaign I've played in since 2010 lasted three years and I ended up with a 7th level thief.
Of course this wasn't the case back in Jr. High when we could play 10 hours sessions all summer long each year and did end up near 36th level by the time High School rolled around ... but that's not how I see most people play OSR games today. So I tend to find having characters start somewhat capable (especially a specialist like a thief) is a good idea - the players are unlikely to ever see "high level play" in the Gygax approved manner. I often doubt how many did even back in the 70's and 80's outside of tournament scenarios. I think most campaigns I know of got to Expert levels at best before enough players moved schools or got in a fight of combo pizza rolls and quit.
So why not update systems for what appear to be the reality of play? Earning 5 levels should be rewarding if one's campaign is 5 levels long (I could even take 15 years if the referee is extra stingy with advancement...) and in such a campaign it makes sense not lock "high level" things (abilities and challenges) away for levels the party will never reach.
IF a designer shifts high-level capability onto 1st-level characters, the structure challenge of a game depending on LBA begins to weaken. Consequently, the obstacles that once challenged the PC DON'T and the D/GM will struggle to provide interesting adventures for that overpowered character. This was the (very widespread) problem with 3e-5e: GMs were/are having a hard time challenging the overpowered PCs because the structure of D&D had been broken (player capability had increased but the capability of most traps and monsters either remained the same as AD&D or they had been nerfed).
While I understand this argument in the context of playing AD&D RAW up to high levels, it seems a bit of a tautology. The old rules tend to make all thief related challenges equal regardless of level ... a flat percentage to do x or y ... but that just doesn't make sense to me.
Sticking to thieves skills, picking the "rusting lockbox of the goatmen bandits" at level one is going to be a challenge perhaps (but doable) while picking the lock on "The Burning Gates of the 95th Abyss" is a much bigger challenge and should be tricky even for a high level PC and impossible at first level. Even higher level classic adventures often go to (annoying I think and emblematic of poor design) efforts limit the usefulness of certain "earned" advantages for spell casters by saying X and Y kind of spell doesn't work. There's no huge problem with this for skills - some challenges are just harder - it makes sense even - and I think only becomes an issue if the player is expecting the abilities to work the same on everything all the time - more a product of the organized and somewhat adversarial play of 3.5E - 5E culture then a mechanics issue.
Moreso, with leveling -- to me it's a given that styles of adventure change with level and so do styles of obstacle though. A high level thief is far more likely to be doing negotiations and running spy networks or pulling a con on the Gods then plundering a ruined chateau full of locked doors etc. They may not even use their thief skills very much as the PC MU will have greater access to easier ways or the party will have magic items that do the work. 27% or 86% to find small traps isn't going to matter much at 30th level - these traps aren't a common danger anymore, but the thief character will also have grown into something beyond a dungeon trap finder and scout.
Challenges should grow and change to maintain freshness - which indeed is hard, but doesn't feel like a solved problem for base BECMI or AD&D - because as you point out it's a referee and adventure design based problem.
Of course this stands right in the middle of "play experience", and yeah, it's fun to run through a game slamming every obstacle with little or no effort. Until it isn't (see
LFQW). So, when I say that a LI Thief being able to crack locks with 83% ability is broken, I'm looking at what that kind of capability will do to a campaign as that PC levels.
For me it makes sense that the a character will have some competence in their specialties even at level one. 5 in 6 at one skill is a bit much perhaps, but if the player wants to make a PC that is only good at one thing and terrible at all else ... who am I to deny them that play experience? One can referee around knowing the PCs will thwart almost every spear or pit trap...
In base rules one can of course do something similar - by adjusting challenges - giving say a base +40% or whatever to picking rusty goat-forged locks or by making the gates of hell have a -40%. The X in 6 method just makes things more granular, but I don't really see the difference between a lock that is +2 in 6 to open and one that has a +40% chance, and if it's something that 1st level PCs should have a good chance to get at - I don't see why I wouldn't include either? Old adventures tend not to very often, but it doesn't feel either unfair or damaging to the play experience. No one gets to smash ever obstacle with ease at any level if the design and refereeing is well done?
That's a bit of a cop-out though on my part perhaps? What I should say is that knowing how one intends to run one's campaign (how long and to what levels even), using or designing adventures that change challenges to evolve with the players and characters, and making challenges challenging but manageable for the level of play throughout the campaign is a core set of RPG referee skills.