1) Yes but you are not factoring in resources such as spells, magic items, class abilities, etc that will increase and even in certain instances guarantee success. You're assuming the use a of a specific skill in a specific way and that's not necessarily a given.
2) No but there are abilities like expertise, there are classes like the Bard who can give inspiration to other PC's, there are simple spells like Bless and guidance that increase the chances of whoever they are cast upon, there are feats
3) But this isn't necessary to run a good heist scenario. this is what I was worried about and why I clarified my position... it's not that I am asking how does BitD play... I am asking specifically what makes it a better choice for a heist scenario than D&D... You're contrasting gameplay but that in and of itself is just showing differences. Also doesn't the GM in BitD have final say on what the position and effect are?
4) Okay a few things you may not be aware of... in 5e, Invisibility is a 2nd level spell (and outside of invisibility there are spells like guidance that will also help). Why does every member of the party need to sprint across the bridge as opposed to say the invisible, expertise in stealth rogue while the others either provide distractions (throw something in the water to distract the half squid/half humans) act as lookouts for any other cult members that might be coming and so on?
5) Funnily enough, what you're describing here sounds more like the show Leverage than everyone being an expert in stealth and all sneaking into some place the same way. In Leverage one character is an expert in stealth while another is an expert in hacking another is a bruiser and so on. They aren't all able to slip in some where like a ghost and get out undetected.
6) Why would they defeat it with stealth alone as opposed to using all of the resources at their disposal? Again very few if any heist movies have the entire ensemble as masters of stealth. And what I see, in the case of 5e, is a narrowly defined set of actions and a DM whose DM'ing has been purposefully set up to push a particular outcome so it supports your scenario.
Numerically broke these out and worked up a big post, but I think I don't want to do this. The very first thing you did here was reverse field on your typical Fighter vs Wizard position (where you lament that everyone is using "Shrodinger's Spells" and "Shrodinger's Loadout" and "Shordinger's Adventuring Day" to explain how Wizards have the potency and breadth and staying power to defeat the overwhelming % of obstacles before them in contrast with a Fighter who does not) and used "Shrodinger's Builds" and "Shrodinger's Group Makeup" and "Shrodinger's Adventuring Day" to have all kinds of resources available to ensure a Stealth Score is enabled. And honestly, I don't even think you need to do that here.
What I wanted to do was discuss something concrete and focused so we could get somewhere (more on that below):
1) Zoom in on a singular obstacle. Your proposition put that initial obstacle as a stealth challenge; a MASSIVE cavern + scary bridge that is very adverse to balance/stealth + 2 active sentinels + 1 active sentinel as relic. The first challenge would be getting across the considerable length of that precarious bridge without being detected.
2) Discuss what defeating that initial stealth obstacle (long, precarious bridge that is stealth adverse with multiple sentinels), via stealth and trickery, would look like under normal circumstances (meaning...not novaing it and dumping a huge amount of resources in it...because there are possible obstacles before and a lot of obstacles left to contend with).
3) Examine the levers that can be used to ENSURE that if you instantiated exact scenario (say) 100 times that the defeating obstacle via stealth and trickery scenario could be pulled off at an extremely high % (say 90%).
4) Discuss the differences between the two games and how this would look. My contention is that this scenario in D&D would just not reproduce the "group as Spec Ops squad that is broadly capable of defeating stealth obstacles." Its not a group of broadly capable scoundrels who can all pretty much do Rogue stuff. A "Stealth Score" is not going to be a thing in D&D. That group is not going to break into that mansion > navigate the mansion > navigate the catacombs > get the relic > get back out...like they were never there. Its just not going to happen.
If they did somehow manage to make it through the mansion down to the catacombs without fully raising the alarm (and razing the mansion), this next bit is going to be full on niche protection. Its not going to go as I proposed above (which, I still hold 100 % that is the correct adjudication for each of those things). Its going to end up as "let the Rogue do it...lets buff him...lets stand back and support...Heavy Fighter...you aren't infantry...get your bow out and be artillery w/ your crappy Dex from the steps...Cleric/Bard, you're buff-guy w/ Bless...Wizard, you're Arcane deception guy - Invis the Rogue to lower the DC (and so you don't have to upcast it to spend your 4th level spell slot which gets all of us) or spend a nothingburger Cantrip for Int vs Save DC.
Then (Sneak * 3 for Invis Rogue vs DC 10 will defeat that obstacle automatically w/ probably a +10 or 11 check). Hopefully the Rogue's Passive Perception or Perception is good enough to figure out the relic is sentient before they pull the thing from the altar.
So it would look about like you would expect. Heavy Fighter hoping things go tits-up so they can actually fight things and be useful (and being frustrated they didn't choose a Dex/Bow build). Cleric relegated to buffer. Wizard doing awesome Wizardy stuff. Rogue being the star of the show.
This is just a totally different deal than a full-on, broadly capable Spec-Ops group where (a) you can defeat stealth obstacles with reliability AS A GROUP because (b) even heavy Fighters have the activatable resources and the ability to negotiate Position/Effect to be sneak-thiefs (or at least not be liabilities in the sneak-thief aspect of play) and (c) support characters are more than buffers in this situation and (d) arcane characters can both do their arcane shtick and sneak across the bridge.
AND, Blades won't develop rote power-plays (as above in the 5e example) to ensure success on stealth obstacle. Niche protection isn't a thing. Characters who reliably incompetent in Stealth or Deception or Social aren't a thing in Blades (like the significant bulk of Fighters in D&D will be...by design). Characters won't be sitting out Stealth Scores/obstacles. Characters will be helping on Engaging a Supernatural Power Score. This is because (a) the way the classes are constructed, (b) the way the action resolution mechanics work, (c) the way the conflict resolution mechanics work, (d) the broad resource structure of play (that isn't class-specific), and (e) the Act Now, Plan Later ethos and the Be Bold/Daring incentive structures that undergird play.
(a) - (e) creates a fundamentally different play experience of Spec Ops teams who are all broadly capable, who are all aggressive and bold, and who can reliably pull of Stealth Scores without engaging in combat...pull off Social Scores without engaging in combat...pull off Smuggling Scores without engaging in combat...pull of Short/Long Cons without engaging in combat...pull off Occult Scores (excorcism or summoning/binding etc) without engaging in combat.
D&D is basically
Plan Now (to develop rote power plays and reduce risk as much as possible...preferably to 0) > Act Later "A Team" where discrete characters with focused niches do VERY specific things when "plans come together" (eg rote power plays inevitably develop) > all hell breaks loose > they fight their way out.
Its awesome. I love it.
But it is absolutely not the same thing as the sort of
hyper-daring, seat-of-your-pants, Spec Ops Caper-fest which somehow holds together through scoundrel guile/skill/panache that Blades produces. Not remotely in its inputs. Not remotely in the cognitive space each player is occupying during play (wholly and from decision-point to decision-point). Not remotely in the dynamism and emergent nature of "OP CHOICE + OP APPROACH". Not in feel. And certainly not in outputs (outputs here meaning - we were
victorious on this entire Score without detection and without body count...that happens in D&D at a rate so ridiculously low that its pointless to even discuss...and this is by design).
Can we at least agree the two bolded things juxtaposed in the above 3 paragraphs are not remotely the same?
They're not the same in conception of genre, not in design, not in execution, and not in what ultimately emerges out of play.