I have. My group used that rule extensively during the early years of 5E. It felt horrible to fail on a nat 1 when your modifier was high enough to succeed if it was not for the auto fail. It is a 5% chance of your build not mattering.
Having a 95% chance of success is the opposite of "your build not mattering"; if your DM didn't want there to be a chance of failure, then they can simply not ask you for a check at all.
It's also kind of a weird assumption that it should be possible to build a character that can never fail at something; fighters can always miss with an attack, so why shouldn't a bard always be able to accidentally insult someone, or a healer be unable to foresee every potential complication in an emergency surgery? If an ancient red dragon can fail to intimidate someone, why shouldn't a humanoid adventurer?
Automatic successes are the more complex case, and one where DM's may wish to adjust the results for an unexpected success; while in a game of dragonchess you might plausibly play so erratically that you throw off your opponent and somehow blunder into a victory, if we go back to the emergency surgery example it's unlikely that a well meaning adventurer might luck onto saving someone, but you might rule that they've bought more time instead. It's reasonable for a DM to rule that characters with and without proficiency in a skill may achieve different outcomes for the same check with the same DC; this is why we have a DM, to make these kinds of determinations.
The DM also plays a crucial role in what a failure actually means; for example, if you build a character to be really good at climbing, and they're properly equipped etc., then failing an Athletics check to climb shouldn't result in a fall except in extreme circumstances, more likely you just don't climb as fast until you can recover. Not ever failure (even a natural 1) needs to involve a character soiling themselves before landing flat on their ass, though there is an unfortunate tendency for some DM's to delight in humiliating players for rolling poorly, that's not really something that can be fixed by the rules.
It is a 1/20 chance that your build does not matter. If a nat 1 is a failure regardless of you having a -5 or a +30, then it is a case of your build not mattering because if your build always mattered, then there would not be an auto fail chance. The nat 1 auto fail, is a 5% chance of negating any bonus your build offers you. If your build always mattered, then if you have a +9, you should always be succeeding at a DC10 or lower not you fail on a DC10 with a nat 1.
I honestly don't like that nat 1's are auto miss and nat 20's being auto hits; however, it is heavily mitigated by the fact that against level appropriate enemies, the nat 1's and nat 20's auto miss and hit rarely ever make a difference as you would normally miss on those rolls anyway. Furthermore, rolling a nat 1 on an attack generally has far lesser consequences than failing a Saving Throw.
Also, it is not weird making a character that can succeed on a nat 1; note that is different from building a character that can never fail. If the DC is high enough, you can still fail. An commoner should not have a chance at succeeding against an an ancient red dragon's frightful presence as the DC for it is 21 and the commoner should not have any bonuses to wisdom saves, meaning it should be impossible for them to succeed that DC21 wisdom save.
Also, complications for certain surgeries are well below the 5% mark. If every surgery had a 5% chance of failure, that would mean 1 in every 20 surgeries, no matter how major or minor, would fail. However, that is not the case in reality. If there is a chance of failure, it should simply be represented by a higher DC.
The rule shouldn't require the DM to think that they shouldn't need to ask for the check. Having a nat 1/20 auto fail/success will imply to DM that they should always ask for a roll because there is always a chance of success or failure.
If a rule relies on a DM being good, it is not a good rule. The current rule in 5E is more ideal when considering a variety of DM's, especially the inexperienced ones. They can introduce the nat 1/20 autofail later as a variant/optional rule later, but it shouldn't be the default RAW.
I have. My group used that rule extensively during the early years of 5E. It felt horrible to fail on a nat 1 when your modifier was high enough to succeed if it was not for the auto fail. It is a 5% chance of your build not mattering.
Having a 95% chance of success is the opposite of "your build not mattering"; if your DM didn't want there to be a chance of failure, then they can simply not ask you for a check at all.
It's also kind of a weird assumption that it should be possible to build a character that can never fail at something; fighters can always miss with an attack, so why shouldn't a bard always be able to accidentally insult someone, or a healer be unable to foresee every potential complication in an emergency surgery? If an ancient red dragon can fail to intimidate someone, why shouldn't a humanoid adventurer?
Automatic successes are the more complex case, and one where DM's may wish to adjust the results for an unexpected success; while in a game of dragonchess you might plausibly play so erratically that you throw off your opponent and somehow blunder into a victory, if we go back to the emergency surgery example it's unlikely that a well meaning adventurer might luck onto saving someone, but you might rule that they've bought more time instead. It's reasonable for a DM to rule that characters with and without proficiency in a skill may achieve different outcomes for the same check with the same DC; this is why we have a DM, to make these kinds of determinations.
The DM also plays a crucial role in what a failure actually means; for example, if you build a character to be really good at climbing, and they're properly equipped etc., then failing an Athletics check to climb shouldn't result in a fall except in extreme circumstances, more likely you just don't climb as fast until you can recover. Not ever failure (even a natural 1) needs to involve a character soiling themselves before landing flat on their ass, though there is an unfortunate tendency for some DM's to delight in humiliating players for rolling poorly, that's not really something that can be fixed by the rules.
It is a 1/20 chance that your build does not matter. If a nat 1 is a failure regardless of you having a -5 or a +30, then it is a case of your build not mattering because if your build always mattered, then there would not be an auto fail chance. The nat 1 auto fail, is a 5% chance of negating any bonus your build offers you. If your build always mattered, then if you have a +9, you should always be succeeding at a DC10 or lower not you fail on a DC10 with a nat 1.
I honestly don't like that nat 1's are auto miss and nat 20's being auto hits; however, it is heavily mitigated by the fact that against level appropriate enemies, the nat 1's and nat 20's auto miss and hit rarely ever make a difference as you would normally miss on those rolls anyway. Furthermore, rolling a nat 1 on an attack generally has far lesser consequences than failing a Saving Throw.
Also, it is not weird making a character that can succeed on a nat 1; note that is different from building a character that can never fail. If the DC is high enough, you can still fail. An commoner should not have a chance at succeeding against an an ancient red dragon's frightful presence as the DC for it is 21 and the commoner should not have any bonuses to wisdom saves, meaning it should be impossible for them to succeed that DC21 wisdom save.
Also, complications for certain surgeries are well below the 5% mark. If every surgery had a 5% chance of failure, that would mean 1 in every 20 surgeries, no matter how major or minor, would fail. However, that is not the case in reality. If there is a chance of failure, it should simply be represented by a higher DC.
The rule shouldn't require the DM to think that they shouldn't need to ask for the check. Having a nat 1/20 auto fail/success will imply to DM that they should always ask for a roll because there is always a chance of success or failure.
If a rule relies on a DM being good, it is not a good rule. The current rule in 5E is more ideal when considering a variety of DM's, especially the inexperienced ones. They can introduce the nat 1/20 autofail later as a variant/optional rule later, but it shouldn't be the default RAW.
Except for the fact that 30 is the highest DC in the rules. 30 is nearly impossible. So if you have a build that gives you a plus 30 you are always succeeding. Even the DMG says that someone level 20 with proficiency and an ability of 20 should need to role a 19 or 20 on a DC 30. Now if they want to adjust the rule of DC we can do that.
As for the real world. Do you have a 100% success rate? Do you never make a mistake? According to you I should be fired becase my accuracy rate is 98%. Your 5% error is too high, note I remember you saying last time we discussed this even 1% was too high, is still 95% of the time a success. Your surgery example still has the mistake you think they relate to each other. Each one has a 95% chance of success. If they have 19 in a row succeed doesn't mean the 20th fails.
It is a 1/20 chance that your build does not matter.
It's a 1/20 chance that you fail. That's not the same as your build not mattering. It means that once you reach 95% any further bonuses don't do anything, but reaching 95% is certainly part of your build, and pretty important.
It is a 1/20 chance that your build does not matter.
It's a 1/20 chance that you fail. That's not the same as your build not mattering. It means that once you reach 95% any further bonuses don't do anything, but reaching 95% is certainly part of your build, and pretty important.
Right. You can even alter your build to counter this. For example, the feats that let your reroll ones, racial abilities, etc. Providing a 5% chance of failure (which, through tactics and/or build, can be mitigated down to 0.0125% (I'm not sure of the rulings, but possibly even further) doesn't negate a build.
Rollback Post to RevisionRollBack
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
Part of the problem in discussing this is how different people view a mistake or failure.
DnD doesn't make you roll for every note in a song, tumbler in a lock, word in a conversation, or step on a balance beam. A roll represents the sum total of your efforts. So yes, a real world expert in a skill will still make some small mistakes. But they are rarely common enough to fail the entire exercise.
Look at that lock picking guy on youtube. He consistently picks the most difficult modern locks in the world in seconds. While he doesn't always get it first try, he always gets it fast. He does not fail to pick 1 in 20 locks entirely. He's an expert.
One could argue that, in the world of DnD, he might have Expertise in thieves tools, and a high Dexterity. But he's not the equivalent of a level 20 rogue. Level 20 characters are superhuman. A high level rogue should absolutely be able to pick any normal lock with ease. Skills are practically all they have to make them superhuman. It's why Expert is a category of class now. A wizard at that level could just disintegrate the door. Let the rogue have an auto success. The cost was one action of their time.
If they are under extreme stress, like trying to pick a lock while being attacked by orcs, then it might matter. That might be the time to introduce a chance of failure. Not because they should always fail 5% of the time. But because right now, with axes coming down on them, they might fail to do it in 6 seconds. Even then, they should get to try again the next round.
I have. My group used that rule extensively during the early years of 5E. It felt horrible to fail on a nat 1 when your modifier was high enough to succeed if it was not for the auto fail. It is a 5% chance of your build not mattering.
Having a 95% chance of success is the opposite of "your build not mattering"; if your DM didn't want there to be a chance of failure, then they can simply not ask you for a check at all.
It's also kind of a weird assumption that it should be possible to build a character that can never fail at something; fighters can always miss with an attack, so why shouldn't a bard always be able to accidentally insult someone, or a healer be unable to foresee every potential complication in an emergency surgery? If an ancient red dragon can fail to intimidate someone, why shouldn't a humanoid adventurer?
Automatic successes are the more complex case, and one where DM's may wish to adjust the results for an unexpected success; while in a game of dragonchess you might plausibly play so erratically that you throw off your opponent and somehow blunder into a victory, if we go back to the emergency surgery example it's unlikely that a well meaning adventurer might luck onto saving someone, but you might rule that they've bought more time instead. It's reasonable for a DM to rule that characters with and without proficiency in a skill may achieve different outcomes for the same check with the same DC; this is why we have a DM, to make these kinds of determinations.
The DM also plays a crucial role in what a failure actually means; for example, if you build a character to be really good at climbing, and they're properly equipped etc., then failing an Athletics check to climb shouldn't result in a fall except in extreme circumstances, more likely you just don't climb as fast until you can recover. Not ever failure (even a natural 1) needs to involve a character soiling themselves before landing flat on their ass, though there is an unfortunate tendency for some DM's to delight in humiliating players for rolling poorly, that's not really something that can be fixed by the rules.
It is a 1/20 chance that your build does not matter. If a nat 1 is a failure regardless of you having a -5 or a +30, then it is a case of your build not mattering because if your build always mattered, then there would not be an auto fail chance. The nat 1 auto fail, is a 5% chance of negating any bonus your build offers you. If your build always mattered, then if you have a +9, you should always be succeeding at a DC10 or lower not you fail on a DC10 with a nat 1.
I honestly don't like that nat 1's are auto miss and nat 20's being auto hits; however, it is heavily mitigated by the fact that against level appropriate enemies, the nat 1's and nat 20's auto miss and hit rarely ever make a difference as you would normally miss on those rolls anyway. Furthermore, rolling a nat 1 on an attack generally has far lesser consequences than failing a Saving Throw.
Also, it is not weird making a character that can succeed on a nat 1; note that is different from building a character that can never fail. If the DC is high enough, you can still fail. An commoner should not have a chance at succeeding against an an ancient red dragon's frightful presence as the DC for it is 21 and the commoner should not have any bonuses to wisdom saves, meaning it should be impossible for them to succeed that DC21 wisdom save.
Also, complications for certain surgeries are well below the 5% mark. If every surgery had a 5% chance of failure, that would mean 1 in every 20 surgeries, no matter how major or minor, would fail. However, that is not the case in reality. If there is a chance of failure, it should simply be represented by a higher DC.
The rule shouldn't require the DM to think that they shouldn't need to ask for the check. Having a nat 1/20 auto fail/success will imply to DM that they should always ask for a roll because there is always a chance of success or failure.
If a rule relies on a DM being good, it is not a good rule. The current rule in 5E is more ideal when considering a variety of DM's, especially the inexperienced ones. They can introduce the nat 1/20 autofail later as a variant/optional rule later, but it shouldn't be the default RAW.
Except for the fact that 30 is the highest DC in the rules. 30 is nearly impossible. So if you have a build that gives you a plus 30 you are always succeeding. Even the DMG says that someone level 20 with proficiency and an ability of 20 should need to role a 19 or 20 on a DC 30. Now if they want to adjust the rule of DC we can do that.
As for the real world. Do you have a 100% success rate? Do you never make a mistake? According to you I should be fired becase my accuracy rate is 98%. Your 5% error is too high, note I remember you saying last time we discussed this even 1% was too high, is still 95% of the time a success. Your surgery example still has the mistake you think they relate to each other. Each one has a 95% chance of success. If they have 19 in a row succeed doesn't mean the 20th fails.
I was using +30 as an extreme example. Highest I ever got without onetime boosts like Flash of Genius or Bardic Inspiration is +21 on Charisma Saves in particular. Also if the task is simple enough, then yes I do have a 100% success rate on doing doing something. Basic addition for example is something I never make a mistake on nowadays. To have a modifier that lets you succeed on a nat 1 is akin to the task in hand being absolutely trivial to you.
Furthermore, the latest UA removed the limit of a DC30 from the d20 tests. So we should be able to go higher than a DC30 now.
I am also not wrong in my surgery example. I never said that after 19th successes that the 20th one will be a failure. I said 1 in 20; that is very different from saying every 20th. Saying 1 in 20 means over a neigh infinite number of attempts, 1 in 20 of them will be failures. If you did a task repeatedly, the number of failures over successes will begin to resemble the average as you increase the number of attempts. The failures won't be perfectly spread apart but that was never what I claimed.
Also, if DC30 was the absolute highest a DC could be without being ouright impossible, I would argue that no one in real life would have a +30 to anything. At a certain point, we are going beyond what it means to be human. If DC30 is the max, then someone with a +30 is very much superhuman which is something that is possible in D&D.
It is a 1/20 chance that your build does not matter.
It's a 1/20 chance that you fail. That's not the same as your build not mattering. It means that once you reach 95% any further bonuses don't do anything, but reaching 95% is certainly part of your build, and pretty important.
I disagree. It makes any bonus that would exceed success on a roll of a 2 pointless. Those bonuses are still part of the build and are negated by the nat 1. If someone actually made the investment to be able to get a high enough modifier to succeed on a nat 1, that modifier should always matter; however it will not matter if you roll a nat 1 and autofail. The nat 1 autofail disregards any modifiers and thus it negates the build because it does not matter if you have a -5 or a +30, a nat 1 will autofail. For it to not negate the build, the modifier must affect the outcome regardless of what is rolled.
It is a 1/20 chance that your build does not matter.
It's a 1/20 chance that you fail. That's not the same as your build not mattering. It means that once you reach 95% any further bonuses don't do anything, but reaching 95% is certainly part of your build, and pretty important.
Right. You can even alter your build to counter this. For example, the feats that let your reroll ones, racial abilities, etc. Providing a 5% chance of failure (which, through tactics and/or build, can be mitigated down to 0.0125% (I'm not sure of the rulings, but possibly even further) doesn't negate a build.
Yes, because you should have to be a halfling on top of all the investment you are already making to get a modifier that high. If someone is already putting in the investment to get a modifier that high, they shouldn't have it negated by a nat 1.
==========================
Here is the other thing I don't see too many people considering: From what I see, the majority of players that would actually make builds that would let them succeed on a nat 1 do not enjoy nat 1 autofails. Meanwhile the group that is in favor of the autofail generally do not raise their modifiers that high. For the latter group, the autofail will never affect them because they would fail on a nat 1 regardless of it being an autofail or not. Meanwhile, the group that enjoys being able to succeed on a nat 1 would have their fun hindered by the nat 1 auto fail.
The nat 1 autofail rule actively hinders the fun for people who do not like it while its inclusion would not affect the vast majority of people who are for it. It harms far more than it does good because the vast majority of the time it does not affect the group that is for it while always adversely affecting the group that is against it.
I disagree. It makes any bonus that would exceed success on a roll of a 2 pointless. Those bonuses are still part of the build and are negated by the nat 1.
And without the nat 1 rule, any bonus that would exceed success on a roll of 1 is pointless. Why is that not a problem?
I disagree. It makes any bonus that would exceed success on a roll of a 2 pointless. Those bonuses are still part of the build and are negated by the nat 1.
And without the nat 1 rule, any bonus that would exceed success on a roll of 1 is pointless. Why is that not a problem?
Because (and this is the part no one wants us to spot and a large number wont admit even to themselves) Without the Nat1=fail they get to showoff their system mastery. And establish their "Alpha Geek" status at the table.
if the task is simple enough, then yes I do have a 100% success rate on doing doing something.
If you are doing something so simple then your DM should not be asking you for a check; a DM only asks for a check when there is a chance of failure.
That said, just because something is simple does not mean you cannot fail to do it under pressure, or at high speed etc. It doesn't matter if your rogue cannot fail to pick a simple lock under ideal circumstances, if you need to pick that lock within 6 seconds. Even in ideal circumstances the lock could be poorly maintained and seize up, a lock-pick could break etc. There is no such thing as something that cannot go wrong, and that uncertainty is a big part of what is fun because adapting to the unexpected is the challenge in any good roleplaying game.
There seems to be some bewildering expectation here that it should be possible for a character to become infallible at something, but even the gods in D&D are not infallible. And again, if your DM agrees that a task is so simple, and in suitable conditions that you cannot reasonably fail, then they can simply not ask you for a check at all, because doing so is wasting everybody's time; if your DM demands rolls that are unnecessary, then get annoyed at that rather than the fact that in One D&D all checks will actually serve a purpose.
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
I disagree. It makes any bonus that would exceed success on a roll of a 2 pointless. Those bonuses are still part of the build and are negated by the nat 1.
And without the nat 1 rule, any bonus that would exceed success on a roll of 1 is pointless. Why is that not a problem?
Not necessary meaningless if degrees of success is a thing. However, the biggest difference is that autofail results in a failure and just feels awful when it is the sole reason of failure while the other feels great when you can succeed on a nat 1.
Because (and this is the part no one wants us to spot and a large number wont admit even to themselves) Without the Nat1=fail they get to showoff their system mastery. And establish their "Alpha Geek" status at the table.
It is not about showing off. It is about making the build investment actually feel good. It feels punishing when you fail simply due to the autofail; meanwhile succeeding on a nat 1 brings a sense of satisfaction as your efforts paid off. Not everyone who enjoys system mastery is there to show off or be the "alpha geek". There is nothing wrong with enjoying system mastery either. I am not forcing others to min/max or optimize and 5E allows for an intermixing of playstyles due to how low the optimization ceiling is.
if the task is simple enough, then yes I do have a 100% success rate on doing doing something.
If you are doing something so simple then your DM should not be asking you for a check; a DM only asks for a check when there is a chance of failure.
That said, just because something is simple does not mean you cannot fail to do it under pressure, or at high speed etc. It doesn't matter if your rogue cannot fail to pick a simple lock under ideal circumstances, if you need to pick that lock within 6 seconds. Even in ideal circumstances the lock could be poorly maintained and seize up, a lock-pick could break etc. There is no such thing as something that cannot go wrong, and that uncertainty is a big part of what is fun because adapting to the unexpected is the challenge in any good roleplaying game.
There seems to be some bewildering expectation here that it should be possible for a character to become infallible at something, but even the gods in D&D are not infallible. And again, if your DM agrees that a task is so simple, and in suitable conditions that you cannot reasonably fail, then they can simply not ask you for a check at all, because doing so is wasting everybody's time; if your DM demands rolls that are unnecessary, then get annoyed at that rather than the fact that in One D&D all checks will actually serve a purpose.
That's the point, if someone can succeed on a nat 1, they shouldn't need to roll. However the nat 1 autofail will be interpretted in a manner in which people will still have to roll because there is always that 5% chance of autofailure. If a situation adds pressure and makes a roll harder, that shouldn't be represented by a 5% autofail; it should be represented by an increased DC. If a lock is normally a DC10, then well maybe it is a DC15 when doing it midcombat. Alternatively, it could mean an additional modifier is lowering your roll, thus bringing your roll into potentially failable ranges.
If you are able to succeed on a nat 1, that should be a sign of the task being absolutely trivial to the point where it is like breathing to you. It does not mean your character is infalliable, because you can still fail at higher DC's.
The Nat 1 Autofail relies far too much on the DM being a good DM. A good rule never relies on the DM being a good DM. It hinders the fun of a subset of players while its inclusion does nothing to the subset that is for it (as that subset is highly unlikely to ever build characters that can succeed on a nat 1).
And as others have stated before, it is far easier to add a rule than to remove a rule. Nat 1 autofail is fine as an optional/variant rule, but not as the default RAW.
What Mana is saying, I believe, is that if 1DD incorporates a rule in which ones are automatic failure and twenties are automatic success, then many, many, many (extremely bad) DMs will interpret that rule as overriding/nullifying the rule of "only roll when the outcome is in doubt and failure has consequences". (Extremely bad) DMs will see the dumbass autosuccess/failure rule and say "the outcome is ALWAYS in doubt!", and they will also work backwards to infer that failure should have consequences even when any sane person would know that's not the case.
It's the sort of rule that promotes and encourages bad play, both on the part of (extremely bad) DMs and on the part of (really not good) players. And yes, I know - "bad players don't make a bad rule!" True. But rules that actively foster bad play and hinder the learning of better habits are indeed bad rules, and automatic success and failure is a bad rule that fosters bad play and hinders the learning of better play.
If you know what you're doing and want to use the rule anyway, per OP, then do it up. But there's a point to be made in that it's already brutally difficult sometimes to wean newer players off of "can I roll for..." and teach them how to play properly. This rule doesn't help.
the nat 1 autofail will be interpretted in a manner in which people will still have to roll because there is always that 5% chance of autofailure.
That is not how checks in D&D work; you tell your DM what you want to do, and they decide if you need to roll a check and what that check should be, they decide if you get advantage or disadvantage, and any other relevant bonuses to the check.
If your DM is asking you for checks for everything then you have a problem with your DM.
If a lock is normally a DC10, then well maybe it is a DC15 when doing it midcombat.
Why? The difficulty of picking the lock hasn't changed, what's changed is that now you want to do it on the first try and might not, whereas out of combat it very likely didn't matter if it took an extra 18 seconds.
If you are able to succeed on a nat 1, that should be a sign of the task being absolutely trivial to the point where it is like breathing to you. It does not mean your character is infalliable, because you can still fail at higher DC's.
That's not experience works. You might have picked the same lock a hundred times every day for the past twenty years, that doesn't mean that today won't be the day that something unforeseen causes it to seize up, or the lock-pick breaks etc.
Even for the simplest of tasks it is possible to fail; a big part of why we have a DM is to decide when we should roll and when we shouldn't for things that are not important to the story, or which just unnecessarily slow things down.
If they're asking for a check then it should be because it matters, and it only matters if you can fail in some way, otherwise it's an irrelevance.
The Nat 1 Autofail relies far too much on the DM being a good DM.
No it doesn't, all it does is reinforce the core of what an ability check is; something that can be failed. This is literally the first paragraph on ability checks:
An ability check tests a character's or monster's innate talent and training in an effort to overcome a challenge. The DM calls for an ability check when a character or monster attempts an action (other than an attack) that has a chance of failure. When the outcome is uncertain, the dice determine the results.
All One D&D is doing is making this statement always true. If your DM isn't using ability checks correctly, tell them.
What Mana is saying, I believe, is that if 1DD incorporates a rule in which ones are automatic failure and twenties are automatic success, then many, many, many (extremely bad) DMs will interpret that rule as overriding/nullifying the rule of "only roll when the outcome is in doubt and failure has consequences".
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Bad dms can still read the rules, and follow them, but yurei has a solid point and so does mana, Many dms will now look at this and say "well shit, I know it says only when its something uncertain they can fail/succeed at, but now that they can nat one or nat 20 anything then i need them to roll
you very much should write rules so that they wont be misinterpreted or abused by a dm, or a player, otherwise your game is gonna kinda suck.
How is a "new" DM supposed to know that making a player roll to order breakfast at the inn is stupid and bad if the rules tell them "if the player rolls a 1 their action fails no matter what"? How's a "new" DM supposed to differentiate between two conflicting rules - one that says "actions that can't fail (or succeed) don't merit a roll" and one that says "literally every single possible imaginable action has a 5% minimum chance of succeeding where all rational reality says it shouldn't or failing despite all sanity saying it can't"?
the nat 1 autofail will be interpretted in a manner in which people will still have to roll because there is always that 5% chance of autofailure.
That is not how checks in D&D work; you tell your DM what you want to do, and they decide if you need to roll a check and what that check should be, they decide if you get advantage or disadvantage, and any other relevant bonuses to the check.
If your DM is asking you for checks for everything then you have a problem with your DM.
This rule will make inexperienced DM's think players have to roll each time because there is always a 5% chance of failure. The problem starts with the rules because DM's will interpret it that way.
If a lock is normally a DC10, then well maybe it is a DC15 when doing it midcombat.
Why? The difficulty of picking the lock hasn't changed, what's changed is that now you want to do it on the first try and might not, whereas out of combat it very likely didn't matter if it took an extra 18 seconds.
Because the DC accounts for the situation, not just the lock itself. The lock may be a DC10 lock, but the actual check is higher because of the additional pressure. That additional pressure should not and currently is not represented by a nat 1 auto fail.
If you are able to succeed on a nat 1, that should be a sign of the task being absolutely trivial to the point where it is like breathing to you. It does not mean your character is infalliable, because you can still fail at higher DC's.
That's not experience works. You might have picked the same lock a hundred times every day for the past twenty years, that doesn't mean that today won't be the day that something unforeseen causes it to seize up, or the lock-pick breaks etc.
Even for the simplest of tasks it is possible to fail; a big part of why we have a DM is to decide when we should roll and when we shouldn't for things that are not important to the story, or which just unnecessarily slow things down.
If they're asking for a check then it should be because it matters, and it only matters if you can fail in some way, otherwise it's an irrelevance.
That is how experience works. As you get better at something, the chance of failure decreases. At a certain point, if you get good enough at something, you stop failing at it. In some of the games I play, I have never failed at clearing certain levels because they are that easy to me. Some math questions I will never get wrong because I am good enough with math that the problem is trivial to me. However, I know there are other people who will have trouble with those things and can fail at them. So the task is not necessary trivial or simple for everyone. Yet, I do not have a 5% or even a sub 1% chance of failing at them, because I am good enough at them. And again, the nat 1 auto fail rule will make inexperienced DM's ask for checks when they shouldn't because they will think the 5% autofail chance means there is always a chance to fail; so even if someone has a modifier higher than the DC, those DMs will ask for rolls.
Also, the rule doesn't apply just to ability checks, but saving throws as well, which are essentially forced.
The Nat 1 Autofail relies far too much on the DM being a good DM.
No it doesn't, all it does is reinforce the core of what an ability check is; something that can be failed. This is literally the first paragraph on ability checks:
An ability check tests a character's or monster's innate talent and training in an effort to overcome a challenge. The DM calls for an ability check when a character or monster attempts an action (other than an attack) that has a chance of failure. When the outcome is uncertain, the dice determine the results.
All One D&D is doing is making this statement always true. If your DM isn't using ability checks correctly, tell them.
If the rules make it so that the player has to tell the DM not to do something, then it is a bad rule. A DM should not be asking for a roll if the player has the modifier to succeed on a nat 1 and currently in 5E that is the case. However, with the nat 1 autofail, everything has a 5% chance of failure now. Someone with a +20 has a 5% chance of failing a DC10 and there will be DMs that will ask for a roll. It very much relies on a good DM because a bad DM can interpret it in a manner where the players always have to roll even if their modifier alone would succeed twice over.
What Mana is saying, I believe, is that if 1DD incorporates a rule in which ones are automatic failure and twenties are automatic success, then many, many, many (extremely bad) DMs will interpret that rule as overriding/nullifying the rule of "only roll when the outcome is in doubt and failure has consequences".
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
It has and it does. My group originally used nat 1 auto fails because we originally thought they were the rules until one of us actually read the rules more clearly regarding Nat 1/20 and found out that in 5E they only applied to Attacks and Death Saves. So the rule actually stopped us from interpreting it incorrectly. Sure there will be people who will get it wrong, but we shouldn't change a rule because some people are interpreting it incorrectly.
If someone has the modifier to succeed on a nat 1, the rules should support them in succeeding on a nat 1.
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Bad dms can still read the rules, and follow them, but yurei has a solid point and so does mana, Many dms will now look at this and say "well shit, I know it says only when its something uncertain they can fail/succeed at, but now that they can nat one or nat 20 anything then i need them to roll
you very much should write rules so that they wont be misinterpreted or abused by a dm, or a player, otherwise your game is gonna kinda suck.
I'd argue that people say you shouldn't roll there are the ones misinterpreting the rules. If you rule meant just don't ask for a roll when a person would never succeed on a 20 or never fail on a 1 the rule you always succeed or fail on those rolls would be nonsensical.
The intent is clearly to allow you to succeed even when it would by the numbers be impossible for you and to fail when by the numbers you couldn't fail, otherwise the rule would be if the DC is such that a 1 would succeed they auto succeed no roll needed. The rule also sets up a grey area where yes by the numbers it would be impossible for them but its so absurd that they could succeed or fail there don't ask for a roll, the closest we got to guidelines was TN 5/30. This isn't a really bad DM issue, its a vague grey area where DMs will be making bad calls all the time.
It is why I would suggest if they wanted a 1/20 roll is special they could do something like if you roll a 20 add your proficiency bonus on to the roll again for your final result.
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Bad dms can still read the rules, and follow them, but yurei has a solid point and so does mana, Many dms will now look at this and say "well shit, I know it says only when its something uncertain they can fail/succeed at, but now that they can nat one or nat 20 anything then i need them to roll
you very much should write rules so that they wont be misinterpreted or abused by a dm, or a player, otherwise your game is gonna kinda suck.
I'd argue that people say you shouldn't roll there are the ones misinterpreting the rules. If you rule meant just don't ask for a roll when a person would never succeed on a 20 or never fail on a 1 the rule you always succeed or fail on those rolls would be nonsensical.
The intent is clearly to allow you to succeed even when it would by the numbers be impossible for you and to fail when by the numbers you couldn't fail, otherwise the rule would be if the DC is such that a 1 would succeed they auto succeed no roll needed. The rule also sets up a grey area where yes by the numbers it would be impossible for them but its so absurd that they could succeed or fail there don't ask for a roll, the closest we got to guidelines was TN 5/30. This isn't a really bad DM issue, its a vague grey area where DMs will be making bad calls all the time.
It is why I would suggest if they wanted a 1/20 roll is special they could do something like if you roll a 20 add your proficiency bonus on to the roll again for your final result.
Actually, the intent was to codify some people mistaking nat 1/20 were auto fail/success. Like I remember Jeremy Crawford saying the reason behind the Nat 1/20 auto fail/success was because people mistaking it as RAW.
Also, it isn't nonsensical for a task being impossible or being absolutely trivial for someone. I will always know what each of the variables in E = mc^2 means; there is no way if someone asks me that question, I will get it wrong. However, your average elementary school student who hasn't ever taken a proper physics class will never figure out that question if asked on the spot. In this situation, it would be nonsensical for me to somehow roll a nat 1 and fail on the check; same with the elementary school student rolling a nat 20 and succeed on the check.
A lot of the issue here is that the d20, while simple, is otherwise a rather poor means of deciding outcomes, because realistic success chance is generally some form of sigmoid -- for example, in many competitive games (such as chess) some variant of the ELO system is used -- often, a variant that uses the logistic curve.
It's actually pretty easy to match a logistic curve to d20 -- use the logistic function with L=100%, K=10. For low skill differences, this does a fine job of matching a d20, but it takes a very long time to actually drop to zero. For example, here are the chance of success with 1-5 levels of superior skill (equal skill = 50%)
55.0%
59.9%
64.6%
69.0%
73.1%
Not quite 5% per level, but pretty close. However, with a skill difference of 10 (100% chance on a d20) the actual odds are only 88.1%; you need +15 to get to 95% and +23 to get to 99%.
However, this is not an easy function to implement with dice, so a common hack is to use a simple die metric (such as a d20 or 3d6) and then hard cap success/fail chances.
How is a "new" DM supposed to know that making a player roll to order breakfast at the inn is stupid and bad if the rules tell them "if the player rolls a 1 their action fails no matter what"? How's a "new" DM supposed to differentiate between two conflicting rules - one that says "actions that can't fail (or succeed) don't merit a roll" and one that says "literally every single possible imaginable action has a 5% minimum chance of succeeding where all rational reality says it shouldn't or failing despite all sanity saying it can't"?
Because the current rules tell you to only roll checks when the outcome of an action is uncertain, and we have been given no reason to expect that this part is being removed from the OneD&D Ability Check section (as they've only been giving us the significant changes). So the rules essentially state:
DO NOT ROLL CHECKS UNLESS THE OUTCOME IS UNCERTAIN.
IF AND ONLY IF AN ACTION IS UNCERTAIN roll a d20 test using a skill or whatever.
If you really care about making it clear to new players, then all that's required is a few clear examples:
"Big Jimmy the Jam Eating Giant decides to put some jam on a piece of bread. He has been doing this literally every hour of every day of his entire life so no check is required for him to succeed."
"Clutzy McFallsalot decides to jump a 500 foot gap across a canyon. No check is required because he's already died in the attempt."
"Jenneric Hugh-Mann wants to climb a ladder in combat, so must pass a DC 10 Athletics check" (DC 15 if he or the ladder is covered in butter).
It feels like problems are being invented to justify hatred of a change that is incredibly benign; especially if "all new DM's and players are braindead idiots" is the current leading argument against, especially considering in my experience this is how most new players and DM's ran skill checks anyway, not realising that natural 1's and 20's were for attacks only. 😂
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Bad dms can still read the rules, and follow them, but yurei has a solid point and so does mana, Many dms will now look at this and say "well shit, I know it says only when its something uncertain they can fail/succeed at, but now that they can nat one or nat 20 anything then i need them to roll
you very much should write rules so that they wont be misinterpreted or abused by a dm, or a player, otherwise your game is gonna kinda suck.
I'd argue that people say you shouldn't roll there are the ones misinterpreting the rules. If you rule meant just don't ask for a roll when a person would never succeed on a 20 or never fail on a 1 the rule you always succeed or fail on those rolls would be nonsensical.
The intent is clearly to allow you to succeed even when it would by the numbers be impossible for you and to fail when by the numbers you couldn't fail, otherwise the rule would be if the DC is such that a 1 would succeed they auto succeed no roll needed. The rule also sets up a grey area where yes by the numbers it would be impossible for them but its so absurd that they could succeed or fail there don't ask for a roll, the closest we got to guidelines was TN 5/30. This isn't a really bad DM issue, its a vague grey area where DMs will be making bad calls all the time.
It is why I would suggest if they wanted a 1/20 roll is special they could do something like if you roll a 20 add your proficiency bonus on to the roll again for your final result.
Actually, the intent was to codify some people mistaking nat 1/20 were auto fail/success. Like I remember Jeremy Crawford saying the reason behind the Nat 1/20 auto fail/success was because people mistaking it as RAW.
Also, it isn't nonsensical for a task being impossible or being absolutely trivial for someone. I will always know what each of the variables in E = mc^2 means; there is no way if someone asks me that question, I will get it wrong. However, your average elementary school student who hasn't ever taken a proper physics class will never figure out that question if asked on the spot. In this situation, it would be nonsensical for me to somehow roll a nat 1 and fail on the check; same with the elementary school student rolling a nat 20 and succeed on the check.
I think you are misunderstanding me. Maybe the reason why they decided to make this rule is because they got a lot of questions about it in sage advice, sure. But the intent of the rule clearly is to allow people to fail on a 1 when 1s would succeed or succeed on a 20 when a 20 would normally fail. If the rule was based around people knowing not to call for rolls in those circumstances, the rule would just say that. The very fact the rule exists that a 20 always succeeds means it is meant for times when a roll of a 20 normally wouldn't by just the die roll and the modifier. So the people claiming its just bad DMs who would make that call are wrong. As the rule would not exist if it was not for times when a 20 on the roll would not succeed.
the idea that the rule is meant not to be used when the TN would be impossible for the player is what I called nonsensical. If I have a TN of 25 +4 to the modifiers and roll a 20, I only got a 24. I'd fail. This rule is clearly for things like that. If I had a +5 modifier I'd of succeeded by the numbers anyways so therefore a rule that I always succeed on a 20 is not needed. To say a good DM knows not to call for the roll when it would be impossible for that player based on their modifiers, is nonsensical as the rule clearly meant to let a person do what would be impossible for them, basically my first example they only got a 24 on that TN 25 test. The guidelines on when it would be impossible and you should not call for a rule are at best suggested at absolute TNs of 5 and 30. That is the guidance DMs are given in the rule as given, not if that players modifiers would make a 20 a failure don't ask for a roll.
IMO if they want a 20/1 to always be an event they should just add a modifier to the result, i suggested prof bonus but that may be too much, make it half proficiency bonus. So basically in my example the 24 would still succeed as even at level 1 1/2 prof bonus is 1 as the final result would be 25. This would mean if I was heavily trained I'd also get a potential perk out of a 20 as at level 20 my result would be 3 higher, so lets say 20 stat, skill, expertise, luck stone, I'm +18, I roll a 20 get a 38 but add +3 for a total result of 41. And on the plus side it would allow a TN 20 test for almost any starting player as the min stat is 8 in d&d normally or a -1, so on a 20 they'd normally get a 19, but now they would add 1/2 prof bonus and get a 20.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
It is a 1/20 chance that your build does not matter. If a nat 1 is a failure regardless of you having a -5 or a +30, then it is a case of your build not mattering because if your build always mattered, then there would not be an auto fail chance. The nat 1 auto fail, is a 5% chance of negating any bonus your build offers you. If your build always mattered, then if you have a +9, you should always be succeeding at a DC10 or lower not you fail on a DC10 with a nat 1.
I honestly don't like that nat 1's are auto miss and nat 20's being auto hits; however, it is heavily mitigated by the fact that against level appropriate enemies, the nat 1's and nat 20's auto miss and hit rarely ever make a difference as you would normally miss on those rolls anyway. Furthermore, rolling a nat 1 on an attack generally has far lesser consequences than failing a Saving Throw.
Also, it is not weird making a character that can succeed on a nat 1; note that is different from building a character that can never fail. If the DC is high enough, you can still fail. An commoner should not have a chance at succeeding against an an ancient red dragon's frightful presence as the DC for it is 21 and the commoner should not have any bonuses to wisdom saves, meaning it should be impossible for them to succeed that DC21 wisdom save.
Also, complications for certain surgeries are well below the 5% mark. If every surgery had a 5% chance of failure, that would mean 1 in every 20 surgeries, no matter how major or minor, would fail. However, that is not the case in reality. If there is a chance of failure, it should simply be represented by a higher DC.
The rule shouldn't require the DM to think that they shouldn't need to ask for the check. Having a nat 1/20 auto fail/success will imply to DM that they should always ask for a roll because there is always a chance of success or failure.
If a rule relies on a DM being good, it is not a good rule. The current rule in 5E is more ideal when considering a variety of DM's, especially the inexperienced ones. They can introduce the nat 1/20 autofail later as a variant/optional rule later, but it shouldn't be the default RAW.
Except for the fact that 30 is the highest DC in the rules. 30 is nearly impossible. So if you have a build that gives you a plus 30 you are always succeeding.
Even the DMG says that someone level 20 with proficiency and an ability of 20 should need to role a 19 or 20 on a DC 30. Now if they want to adjust the rule of DC we can do that.
As for the real world. Do you have a 100% success rate? Do you never make a mistake? According to you I should be fired becase my accuracy rate is 98%. Your 5% error is too high, note I remember you saying last time we discussed this even 1% was too high, is still 95% of the time a success. Your surgery example still has the mistake you think they relate to each other. Each one has a 95% chance of success. If they have 19 in a row succeed doesn't mean the 20th fails.
It's a 1/20 chance that you fail. That's not the same as your build not mattering. It means that once you reach 95% any further bonuses don't do anything, but reaching 95% is certainly part of your build, and pretty important.
Right. You can even alter your build to counter this. For example, the feats that let your reroll ones, racial abilities, etc. Providing a 5% chance of failure (which, through tactics and/or build, can be mitigated down to 0.0125% (I'm not sure of the rulings, but possibly even further) doesn't negate a build.
If you're not willing or able to to discuss in good faith, then don't be surprised if I don't respond, there are better things in life for me to do than humour you. This signature is that response.
Part of the problem in discussing this is how different people view a mistake or failure.
DnD doesn't make you roll for every note in a song, tumbler in a lock, word in a conversation, or step on a balance beam. A roll represents the sum total of your efforts. So yes, a real world expert in a skill will still make some small mistakes. But they are rarely common enough to fail the entire exercise.
Look at that lock picking guy on youtube. He consistently picks the most difficult modern locks in the world in seconds. While he doesn't always get it first try, he always gets it fast. He does not fail to pick 1 in 20 locks entirely. He's an expert.
One could argue that, in the world of DnD, he might have Expertise in thieves tools, and a high Dexterity. But he's not the equivalent of a level 20 rogue. Level 20 characters are superhuman. A high level rogue should absolutely be able to pick any normal lock with ease. Skills are practically all they have to make them superhuman. It's why Expert is a category of class now. A wizard at that level could just disintegrate the door. Let the rogue have an auto success. The cost was one action of their time.
If they are under extreme stress, like trying to pick a lock while being attacked by orcs, then it might matter. That might be the time to introduce a chance of failure. Not because they should always fail 5% of the time. But because right now, with axes coming down on them, they might fail to do it in 6 seconds. Even then, they should get to try again the next round.
I was using +30 as an extreme example. Highest I ever got without onetime boosts like Flash of Genius or Bardic Inspiration is +21 on Charisma Saves in particular. Also if the task is simple enough, then yes I do have a 100% success rate on doing doing something. Basic addition for example is something I never make a mistake on nowadays. To have a modifier that lets you succeed on a nat 1 is akin to the task in hand being absolutely trivial to you.
Furthermore, the latest UA removed the limit of a DC30 from the d20 tests. So we should be able to go higher than a DC30 now.
I am also not wrong in my surgery example. I never said that after 19th successes that the 20th one will be a failure. I said 1 in 20; that is very different from saying every 20th. Saying 1 in 20 means over a neigh infinite number of attempts, 1 in 20 of them will be failures. If you did a task repeatedly, the number of failures over successes will begin to resemble the average as you increase the number of attempts. The failures won't be perfectly spread apart but that was never what I claimed.
Also, if DC30 was the absolute highest a DC could be without being ouright impossible, I would argue that no one in real life would have a +30 to anything. At a certain point, we are going beyond what it means to be human. If DC30 is the max, then someone with a +30 is very much superhuman which is something that is possible in D&D.
I disagree. It makes any bonus that would exceed success on a roll of a 2 pointless. Those bonuses are still part of the build and are negated by the nat 1. If someone actually made the investment to be able to get a high enough modifier to succeed on a nat 1, that modifier should always matter; however it will not matter if you roll a nat 1 and autofail. The nat 1 autofail disregards any modifiers and thus it negates the build because it does not matter if you have a -5 or a +30, a nat 1 will autofail. For it to not negate the build, the modifier must affect the outcome regardless of what is rolled.
Yes, because you should have to be a halfling on top of all the investment you are already making to get a modifier that high. If someone is already putting in the investment to get a modifier that high, they shouldn't have it negated by a nat 1.
==========================
Here is the other thing I don't see too many people considering: From what I see, the majority of players that would actually make builds that would let them succeed on a nat 1 do not enjoy nat 1 autofails. Meanwhile the group that is in favor of the autofail generally do not raise their modifiers that high. For the latter group, the autofail will never affect them because they would fail on a nat 1 regardless of it being an autofail or not. Meanwhile, the group that enjoys being able to succeed on a nat 1 would have their fun hindered by the nat 1 auto fail.
The nat 1 autofail rule actively hinders the fun for people who do not like it while its inclusion would not affect the vast majority of people who are for it. It harms far more than it does good because the vast majority of the time it does not affect the group that is for it while always adversely affecting the group that is against it.
And without the nat 1 rule, any bonus that would exceed success on a roll of 1 is pointless. Why is that not a problem?
Because (and this is the part no one wants us to spot and a large number wont admit even to themselves) Without the Nat1=fail they get to showoff their system mastery. And establish their "Alpha Geek" status at the table.
If you are doing something so simple then your DM should not be asking you for a check; a DM only asks for a check when there is a chance of failure.
That said, just because something is simple does not mean you cannot fail to do it under pressure, or at high speed etc. It doesn't matter if your rogue cannot fail to pick a simple lock under ideal circumstances, if you need to pick that lock within 6 seconds. Even in ideal circumstances the lock could be poorly maintained and seize up, a lock-pick could break etc. There is no such thing as something that cannot go wrong, and that uncertainty is a big part of what is fun because adapting to the unexpected is the challenge in any good roleplaying game.
There seems to be some bewildering expectation here that it should be possible for a character to become infallible at something, but even the gods in D&D are not infallible. And again, if your DM agrees that a task is so simple, and in suitable conditions that you cannot reasonably fail, then they can simply not ask you for a check at all, because doing so is wasting everybody's time; if your DM demands rolls that are unnecessary, then get annoyed at that rather than the fact that in One D&D all checks will actually serve a purpose.
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
Not necessary meaningless if degrees of success is a thing. However, the biggest difference is that autofail results in a failure and just feels awful when it is the sole reason of failure while the other feels great when you can succeed on a nat 1.
It is not about showing off. It is about making the build investment actually feel good. It feels punishing when you fail simply due to the autofail; meanwhile succeeding on a nat 1 brings a sense of satisfaction as your efforts paid off. Not everyone who enjoys system mastery is there to show off or be the "alpha geek". There is nothing wrong with enjoying system mastery either. I am not forcing others to min/max or optimize and 5E allows for an intermixing of playstyles due to how low the optimization ceiling is.
That's the point, if someone can succeed on a nat 1, they shouldn't need to roll. However the nat 1 autofail will be interpretted in a manner in which people will still have to roll because there is always that 5% chance of autofailure. If a situation adds pressure and makes a roll harder, that shouldn't be represented by a 5% autofail; it should be represented by an increased DC. If a lock is normally a DC10, then well maybe it is a DC15 when doing it midcombat. Alternatively, it could mean an additional modifier is lowering your roll, thus bringing your roll into potentially failable ranges.
If you are able to succeed on a nat 1, that should be a sign of the task being absolutely trivial to the point where it is like breathing to you. It does not mean your character is infalliable, because you can still fail at higher DC's.
The Nat 1 Autofail relies far too much on the DM being a good DM. A good rule never relies on the DM being a good DM. It hinders the fun of a subset of players while its inclusion does nothing to the subset that is for it (as that subset is highly unlikely to ever build characters that can succeed on a nat 1).
And as others have stated before, it is far easier to add a rule than to remove a rule. Nat 1 autofail is fine as an optional/variant rule, but not as the default RAW.
What Mana is saying, I believe, is that if 1DD incorporates a rule in which ones are automatic failure and twenties are automatic success, then many, many, many (extremely bad) DMs will interpret that rule as overriding/nullifying the rule of "only roll when the outcome is in doubt and failure has consequences". (Extremely bad) DMs will see the dumbass autosuccess/failure rule and say "the outcome is ALWAYS in doubt!", and they will also work backwards to infer that failure should have consequences even when any sane person would know that's not the case.
It's the sort of rule that promotes and encourages bad play, both on the part of (extremely bad) DMs and on the part of (really not good) players. And yes, I know - "bad players don't make a bad rule!" True. But rules that actively foster bad play and hinder the learning of better habits are indeed bad rules, and automatic success and failure is a bad rule that fosters bad play and hinders the learning of better play.
If you know what you're doing and want to use the rule anyway, per OP, then do it up. But there's a point to be made in that it's already brutally difficult sometimes to wean newer players off of "can I roll for..." and teach them how to play properly. This rule doesn't help.
Please do not contact or message me.
That is not how checks in D&D work; you tell your DM what you want to do, and they decide if you need to roll a check and what that check should be, they decide if you get advantage or disadvantage, and any other relevant bonuses to the check.
If your DM is asking you for checks for everything then you have a problem with your DM.
Why? The difficulty of picking the lock hasn't changed, what's changed is that now you want to do it on the first try and might not, whereas out of combat it very likely didn't matter if it took an extra 18 seconds.
That's not experience works. You might have picked the same lock a hundred times every day for the past twenty years, that doesn't mean that today won't be the day that something unforeseen causes it to seize up, or the lock-pick breaks etc.
Even for the simplest of tasks it is possible to fail; a big part of why we have a DM is to decide when we should roll and when we shouldn't for things that are not important to the story, or which just unnecessarily slow things down.
If they're asking for a check then it should be because it matters, and it only matters if you can fail in some way, otherwise it's an irrelevance.
No it doesn't, all it does is reinforce the core of what an ability check is; something that can be failed. This is literally the first paragraph on ability checks:
All One D&D is doing is making this statement always true. If your DM isn't using ability checks correctly, tell them.
The rules can't prevent a bad DM from interpreting something incorrectly, especially if they purposefully ignore a section that tells them the opposite; ability checks in D&D 5e have always been optional, you only roll a check when the outcome is uncertain. With nat 1's and 20's now the outcome is always uncertain, that's it. That is all that has changed.
And you don't solve bad DM's or players by trying to rule around them; the whole problem of bad DM's is they're not reading the rules, or they're misunderstanding them (not all "bad" DM's are malicious). If anything, making all d20 tests function in essentially the same way will make it less likely they'll skip the parts that matter, because instead of three similar yet slightly different rolls, we now have one that's used for different things but otherwise behaves the same.
Maybe now they'll read the paragraph that tells them when to roll checks, maybe they won't; that's not a problem that can be solved, but less reading, or less complexity, is almost always going to help.
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
Bad dms can still read the rules, and follow them, but yurei has a solid point and so does mana, Many dms will now look at this and say "well shit, I know it says only when its something uncertain they can fail/succeed at, but now that they can nat one or nat 20 anything then i need them to roll
you very much should write rules so that they wont be misinterpreted or abused by a dm, or a player, otherwise your game is gonna kinda suck.
How is a "new" DM supposed to know that making a player roll to order breakfast at the inn is stupid and bad if the rules tell them "if the player rolls a 1 their action fails no matter what"? How's a "new" DM supposed to differentiate between two conflicting rules - one that says "actions that can't fail (or succeed) don't merit a roll" and one that says "literally every single possible imaginable action has a 5% minimum chance of succeeding where all rational reality says it shouldn't or failing despite all sanity saying it can't"?
Please do not contact or message me.
This rule will make inexperienced DM's think players have to roll each time because there is always a 5% chance of failure. The problem starts with the rules because DM's will interpret it that way.
Because the DC accounts for the situation, not just the lock itself. The lock may be a DC10 lock, but the actual check is higher because of the additional pressure. That additional pressure should not and currently is not represented by a nat 1 auto fail.
That is how experience works. As you get better at something, the chance of failure decreases. At a certain point, if you get good enough at something, you stop failing at it. In some of the games I play, I have never failed at clearing certain levels because they are that easy to me. Some math questions I will never get wrong because I am good enough with math that the problem is trivial to me. However, I know there are other people who will have trouble with those things and can fail at them. So the task is not necessary trivial or simple for everyone. Yet, I do not have a 5% or even a sub 1% chance of failing at them, because I am good enough at them. And again, the nat 1 auto fail rule will make inexperienced DM's ask for checks when they shouldn't because they will think the 5% autofail chance means there is always a chance to fail; so even if someone has a modifier higher than the DC, those DMs will ask for rolls.
Also, the rule doesn't apply just to ability checks, but saving throws as well, which are essentially forced.
If the rules make it so that the player has to tell the DM not to do something, then it is a bad rule. A DM should not be asking for a roll if the player has the modifier to succeed on a nat 1 and currently in 5E that is the case. However, with the nat 1 autofail, everything has a 5% chance of failure now. Someone with a +20 has a 5% chance of failing a DC10 and there will be DMs that will ask for a roll. It very much relies on a good DM because a bad DM can interpret it in a manner where the players always have to roll even if their modifier alone would succeed twice over.
It has and it does. My group originally used nat 1 auto fails because we originally thought they were the rules until one of us actually read the rules more clearly regarding Nat 1/20 and found out that in 5E they only applied to Attacks and Death Saves. So the rule actually stopped us from interpreting it incorrectly. Sure there will be people who will get it wrong, but we shouldn't change a rule because some people are interpreting it incorrectly.
If someone has the modifier to succeed on a nat 1, the rules should support them in succeeding on a nat 1.
I'd argue that people say you shouldn't roll there are the ones misinterpreting the rules. If you rule meant just don't ask for a roll when a person would never succeed on a 20 or never fail on a 1 the rule you always succeed or fail on those rolls would be nonsensical.
The intent is clearly to allow you to succeed even when it would by the numbers be impossible for you and to fail when by the numbers you couldn't fail, otherwise the rule would be if the DC is such that a 1 would succeed they auto succeed no roll needed. The rule also sets up a grey area where yes by the numbers it would be impossible for them but its so absurd that they could succeed or fail there don't ask for a roll, the closest we got to guidelines was TN 5/30. This isn't a really bad DM issue, its a vague grey area where DMs will be making bad calls all the time.
It is why I would suggest if they wanted a 1/20 roll is special they could do something like if you roll a 20 add your proficiency bonus on to the roll again for your final result.
Actually, the intent was to codify some people mistaking nat 1/20 were auto fail/success. Like I remember Jeremy Crawford saying the reason behind the Nat 1/20 auto fail/success was because people mistaking it as RAW.
Also, it isn't nonsensical for a task being impossible or being absolutely trivial for someone. I will always know what each of the variables in E = mc^2 means; there is no way if someone asks me that question, I will get it wrong. However, your average elementary school student who hasn't ever taken a proper physics class will never figure out that question if asked on the spot. In this situation, it would be nonsensical for me to somehow roll a nat 1 and fail on the check; same with the elementary school student rolling a nat 20 and succeed on the check.
A lot of the issue here is that the d20, while simple, is otherwise a rather poor means of deciding outcomes, because realistic success chance is generally some form of sigmoid -- for example, in many competitive games (such as chess) some variant of the ELO system is used -- often, a variant that uses the logistic curve.
It's actually pretty easy to match a logistic curve to d20 -- use the logistic function with L=100%, K=10. For low skill differences, this does a fine job of matching a d20, but it takes a very long time to actually drop to zero. For example, here are the chance of success with 1-5 levels of superior skill (equal skill = 50%)
Not quite 5% per level, but pretty close. However, with a skill difference of 10 (100% chance on a d20) the actual odds are only 88.1%; you need +15 to get to 95% and +23 to get to 99%.
However, this is not an easy function to implement with dice, so a common hack is to use a simple die metric (such as a d20 or 3d6) and then hard cap success/fail chances.
Because the current rules tell you to only roll checks when the outcome of an action is uncertain, and we have been given no reason to expect that this part is being removed from the OneD&D Ability Check section (as they've only been giving us the significant changes). So the rules essentially state:
If you really care about making it clear to new players, then all that's required is a few clear examples:
It feels like problems are being invented to justify hatred of a change that is incredibly benign; especially if "all new DM's and players are braindead idiots" is the current leading argument against, especially considering in my experience this is how most new players and DM's ran skill checks anyway, not realising that natural 1's and 20's were for attacks only. 😂
Former D&D Beyond Customer of six years: With the axing of piecemeal purchasing, lack of meaningful development, and toxic moderation the site isn't worth paying for anymore. I remain a free user only until my groups are done migrating from DDB, and if necessary D&D, after which I'm done. There are better systems owned by better companies out there.
I have unsubscribed from all topics and will not reply to messages. My homebrew is now 100% unsupported.
I think you are misunderstanding me. Maybe the reason why they decided to make this rule is because they got a lot of questions about it in sage advice, sure. But the intent of the rule clearly is to allow people to fail on a 1 when 1s would succeed or succeed on a 20 when a 20 would normally fail. If the rule was based around people knowing not to call for rolls in those circumstances, the rule would just say that. The very fact the rule exists that a 20 always succeeds means it is meant for times when a roll of a 20 normally wouldn't by just the die roll and the modifier. So the people claiming its just bad DMs who would make that call are wrong. As the rule would not exist if it was not for times when a 20 on the roll would not succeed.
the idea that the rule is meant not to be used when the TN would be impossible for the player is what I called nonsensical. If I have a TN of 25 +4 to the modifiers and roll a 20, I only got a 24. I'd fail. This rule is clearly for things like that. If I had a +5 modifier I'd of succeeded by the numbers anyways so therefore a rule that I always succeed on a 20 is not needed. To say a good DM knows not to call for the roll when it would be impossible for that player based on their modifiers, is nonsensical as the rule clearly meant to let a person do what would be impossible for them, basically my first example they only got a 24 on that TN 25 test. The guidelines on when it would be impossible and you should not call for a rule are at best suggested at absolute TNs of 5 and 30. That is the guidance DMs are given in the rule as given, not if that players modifiers would make a 20 a failure don't ask for a roll.
IMO if they want a 20/1 to always be an event they should just add a modifier to the result, i suggested prof bonus but that may be too much, make it half proficiency bonus. So basically in my example the 24 would still succeed as even at level 1 1/2 prof bonus is 1 as the final result would be 25. This would mean if I was heavily trained I'd also get a potential perk out of a 20 as at level 20 my result would be 3 higher, so lets say 20 stat, skill, expertise, luck stone, I'm +18, I roll a 20 get a 38 but add +3 for a total result of 41. And on the plus side it would allow a TN 20 test for almost any starting player as the min stat is 8 in d&d normally or a -1, so on a 20 they'd normally get a 19, but now they would add 1/2 prof bonus and get a 20.