D&D 5E Reliable Talent. What the what?

[Ovinomancer]

That has nothing to do with bounded accuracy but 1 and 20 being criticals.

Pretty sure that's part of bounded accuracy.

The best archer in the world may have rolled an 18 for his dex, meaning (Ranger or Fighter in this case) he only needs his first ASI (LvL4) to always hit a commoner unless he failed critically.
As for the best Wizard, he will be impossible to resists for a commoner for the best Wizard in the World certainly gets something like the Robe. Given there is no critical for Saves the Commoner has NO chance to ever succeed.

No, the fighter ALWAYS has a chance of missing. You cannot excise a core part of the mechanics that purport to follow bounded accuracy and say 'well, surely they didn't mean this part, too.'

As for the wizard, I'm totally right. The wizard needs help outside of his ability to achieve 100%. But you're splitting inconsequential hairs -- the wizard, at no tier, receives class abilities that obviate bounded accuracy. To even get the lowest levels of such, he must seek help outside of his class abilities.

As you see: There is plenty of precedence where bounded accuracy is worth nothing for it is completely ignored as being relevant for trash encounters.

I disagree, as those require cooperation between multiple game rule niches to get to just the lowest levels of floor, and they're gradual and hard earned. No other class receives benefits that obviate bounded accuracy anywhere near the same level. And, when those do, they obviate the lowest DCs, not the middle of the pack. The Wizard's DC and the archer's attack role never suddenly turn every number the bottom half of the d20 into a 10, for instance. Reliable talent does.

Again, this isn't sufficient to make reliable talent bad. I roll with it just fine, even as I dislike the design elements it embodies. I can point out what I consider the very flawed design behind reliable talent and still say that it doesn't break the game. This is part of why I suggest an alternative path that achieves the same design goals while not making a sudden jump that tosses bounded accuracy for all of the rogue's proficient skills (which is sizeable). It's a disruptive design element because it radically shifts the necessary thought process behind adventure design and doesn't offer a significant benefit to the game that cannot be achieved through other means. I find design elements that run directly counter to core conceits to be bad, especially when mechanics already exist that suffice for the design goal, like offering advantage instead of an arbitrary floor. This is a holdover from previous editions I find unsuited to the core mechanics of 5e. It's further exacerbated by it's pairing with expertise, suddenly shifting DCs that were in questions (DC 20, for instance, has a 35% chance of failure at 10th level. At 11th, it shifts to 0%. A DC 23 has a 45% chance of failure at 10th. At 11th, it's 0%. No other class ability makes as big a shift, and those that are even remotely close don't do so for such a wide range of checks. Instead offering advantage still works, as it dramatically reduces failure chances for easy checks to near zero, but not to zero, and would change the failure chance for that DC 23 to ~20%, The chance to make a DC 30 check goes from 20% to ~32%. That's a benefit across the board while maintaining a small chance of failure and not just autosuccess for the vast majority of DCs encountered.

 

[redrick]

Another way to think about Reliable Talent is the options it opens up for your adventures. Reliable Talent allows your Rogue to do difficult, and potentially dangerous, things (depending on where their expertise and ability scores are distributed) without risk of killing the character. Instead of saying, "Damn. The Rogue can walk a tightrope across a bottomless chasm with no risk of slipping and falling off. This sucks," why not say, "Damn, the Rogue can run a tightrope across a bottomless chasm and leap through a narrow window without risk of falling. We can do some awesome stuff!" Suddenly, some serious Ethan Hunt shenanigans are on the table. And once you say, "As you are racing along the tightrope, suddenly a hurricane force gust of wind starts whipping through the canyon — what do you do?" The players know that if you call for a roll, the situation is intense and the risk of failure is high.

Level 11. We on some high-tier now.

-----

Meanwhile, the 5th level caster is flying the rest of the party and their donkey across the chasm with a 3rd level spell. But the Rogue looks way cooler.

 

[schnee]

Well, the best archer in the world misses the easiest target with frightening regularity. The best wizard in the world can employ his most powerful magicks and fail to affect a commoner. This is what a game based on bounded accuracy means.

And the best Rogue skill check lets them talk to someone, or to climb a wall, or pick a pocket.

Meanwhile, that Wizard transforms into a Black Dragon.

Scope of a given ability is as important an axis to measure as reliability.

 

[Saeviomagy]

Well there you go. 11-16 is third tier, where characters "are special even among adventures."

And yet they still can't pull off moderate skill checks in their bailiwicks with reasonable reliability.

 

[The Crimson Binome]

As you see: There is plenty of precedence where bounded accuracy is worth nothing for it is completely ignored as being relevant for trash encounters.

That could be taken as a failure of the system, then. One of the stated goals behind Bounded Accuracy was that you could still use regular orcs as enemies against high-level PCs, and the math wouldn't fall apart the way it does in 4E.

Although, to be fair to the designers, it was usually stated in the form of "everyone has a chance to succeed" rather than "everyone has a chance to fail"; it wasn't as important that an epic fighter miss against an orc, as that the orc can still hit the fighter without needing to roll a 20. And that mostly remains true, as long as the fighter doesn't have magic armor.

 

[Stalker0]

I feel that at this point every reasonable argument FOR and AGAINST have been made.

So OP what's your decision....keeping it or not?

 

[FieserMoep]

True, though there are als plenty of builds with near “constant“ absurd ac that prove that bounded accuracy has its limits, by design. You may call that failure, icall that the nature of.the beast.
In the end, what i am saying is: bounded accuracy is a design philosophy, not a dogma one has to recite for every homebrew to “balance“ a core mechanic.

Rogues are skill monkeys and they invested resources to get reliable. If one is dependant on a simple heist going wrong for a plot hook at that level you may just reroll lvl 1 characters. God forbid any decently played wizard starts to vast in your group.

 

[Xetheral]

It seems that many of the respondents here are insistent that I'm out to make sure that this rogue fails, and I'm just as insistent that that is just not the case. What I want is tension when faced with a skill check that could have dangerous consequences, even if it's just the tension of uncertainty; the possibility, even an extremely small one, of failure. As I said in an earlier reply, I think even the nat 1 failure rate for skills with which the rogue has Reliable Talent is too high at 5%. It is not that I want him to fail, but neither do I want challenges to become total non issues.

But, as someone pointed out, they've got mage hand and all sorts of other clever ways to mitigate traps, so maybe I'm over thinking it. The traps are just a small portion of an already combat heavy and fairly challenging adventure.

You have a different definition of "challenge" than I do. To me, succeeding at a die roll isn't a challenge, it's luck. The challenge comes from figuring out how to approach an entire task to maximize the chance of success. At the point where success or failure hinges on a roll of the die, the challenge is already over before the roll.

Under that interpretation, Reliable Talent only has an indirect impact on the level challenge. The task of robbing a house doesn't change significantly: the odds of failing due to chance simply go down. The character still has the challenge of figuring out when occupants will be home, whether there will be sufficient cover inside the house to be able to hide once inside, whether there is a dog, whether the neighbors have dogs (who wound be honor-bound by the canine code to bark if the dog in the targeted home starts barking), what tools will be needed to access any valuables (e.g. a shovel for digging up the hearth, a crowbar to move the hearth flagstones), whether light inside the house would be visible to the neighbors (for thieves without darkvision), whether alerted neighbors will rally to attack a thief or mind their own business, how to approach the house without being seen (strangers will be memorable in a small town), escape routes from the neighborhood, etc.

Ultimately, under my interpretation, the challenge comes from figuring out the approach to the task, such as coming up with the above list of information needed to maximize the odds of a successful burglary. Reliable Talent may make certain approaches better than others, or open up new approaches that would previously have been gated behind too many die rolls, or even make a horrible approach feasible, but mostly it just reduces the chance that a given approach will simply fail due to luck.

 

[Ovinomancer]

You have a different definition of "challenge" than I do. To me, succeeding at a die roll isn't a challenge, it's luck. The challenge comes from figuring out how to approach an entire task to maximize the chance of success. At the point where success or failure hinges on a roll of the die, the challenge is already over before the roll.

Under that interpretation, Reliable Talent only has an indirect impact on the level challenge. The task of robbing a house doesn't change significantly: the odds of failing due to chance simply go down. The character still has the challenge of figuring out when occupants will be home, whether there will be sufficient cover inside the house to be able to hide once inside, whether there is a dog, whether the neighbors have dogs (who wound be honor-bound by the canine code to bark if the dog in the targeted home starts barking), what tools will be needed to access any valuables (e.g. a shovel for digging up the hearth, a crowbar to move the hearth flagstones), whether light inside the house would be visible to the neighbors (for thieves without darkvision), whether alerted neighbors will rally to attack a thief or mind their own business, how to approach the house without being seen (strangers will be memorable in a small town), escape routes from the neighborhood, etc.

Ultimately, under my interpretation, the challenge comes from figuring out the approach to the task, such as coming up with the above list of information needed to maximize the odds of a successful burglary. Reliable Talent may make certain approaches better than others, or open up new approaches that would previously have been gated behind too many die rolls, or even make a horrible approach feasible, but mostly it just reduces the chance that a given approach will simply fail due to luck.

I'd actually argue that by making certain approaches known automatic successes, it significantly reduces the enjoyment of devising a suitable approach. It trivializes the challenge by confirming the outcome without effort. As others have said, the best way to adjust adventure design to reliable talent isn't to continue to provide the same kinds of challenge, because reliable talent renders them moot, but to switch to new challenges that don't engage reliable talent or assume reliable talent to gain access to the new challenge. In effect, the advice is to stop using those kinds of challenges that reliable talent trivializes and instead use different challenges and turn reliable talent into a narration device.

And that's exactly how I've dealt with it. I have an issue though that the best method of dealing with a player investing into skills means that those skills so invested become narration devices except when a very difficult challenge comes up, and then that investment is of limited use (failure rates at DCs over the reliable talent threshold revert to 50/50). This doesn't really help me understand how being super duper good at a skill means that I either never fail or fail more than half the time. Let me pimp the advantage replacement for reliable talent again as addressing both of these issues with an already extant game mechanic.

 

[ad_hoc]

That has nothing to do with bounded accuracy but 1 and 20 being criticals.
The best archer in the world may have rolled an 18 for his dex, meaning (Ranger or Fighter in this case) he only needs his first ASI (LvL4) to always hit a commoner unless he failed critically.
As for the best Wizard, he will be impossible to resists for a commoner for the best Wizard in the World certainly gets something like the Robe. Given there is no critical for Saves the Commoner has NO chance to ever succeed.

As you see: There is plenty of precedence where bounded accuracy is worth nothing for it is completely ignored as being relevant for trash encounters.

It's not even that. Under non-stressful conditions the best archer in the world will hit a target 100% of the time. There is no check, they just succeed.

Combat is a stressful situation with usually moving targets (and if the target isn't moving the chance to succeed is much higher).

Now, a character like Marvel's Hawkeye will succeed at ridiculous shots even when under stress.

And that's what we're talking about really.

The Rogue is Hawkeye. The Barbarian is The Hulk. The Wizard is Iron Man.

Sure, Hawkeye has a great power to never miss, but compared to The Hulk and Iron Man this ability pales in comparison and the propensity for breaking the adventure is much much lower.