And that's why I find it interesting. A 0.25% chance of failure is negligible. That, to me, is a very strong indication that this has nothing to do with power or balance, but just style preferences; a desire for dice rolling, no matter how skewed, than it is about someone succeeding or failing.
That is exactly my perspective, a matter of style and preference. I should have probably started by saying that from the beginning, shouldn't I?
Rollback Post to RevisionRollBack
Born in Italy, moved a bunch, living in Spain, my heart always belonged to Roleplaying Games
I'm sorry, but, again, there is no "variant rule" that makes 1s auto-fail on ability checks. There's a variant rule that makes 1s that are also failures into "critial failures" (and 20s that are also successes into "critical successes").
Some of the text, trying to avoid posting it all because its part of the DMG which is a paid book, so I don't want to have my post removed.
Rolling a 20 or a 1 on an ability check or a saving throw doesn’t normally have any special effect. However, you can choose to take such an exceptional roll into account when adjudicating the outcome. It’s up to you to determine how this manifests in the game.
It then gives you examples saying you can reward players for critical success or failures, but those are examples of what you can do. You could also just treat the roll as a failure. if you have the DMG you can click on the following link to see the exact text Critical Success or Failure
Please visit the link you posted, and read the last sentence there. "Rolling 1 on a failed attempt..." If the attempt didn't fail, the 1 does nothing. It does not say "rolling 1 on an attempt...", it specifically says rolling a 1 on a failed attempt. That rule variant gives you options for what to do with "spectacular" failures or successes, not for turning successes into failures.
My apologies, you are correct, it appears I am slightly modifying the variant rule making it a house role.
My point still stands, I don't think anyone should be able to auto-succeed at anything. I have no problem with them practically never failing, but I do believe they should always be able to fail.
And that's why I find it interesting. A 0.25% chance of failure is negligible. That, to me, is a very strong indication that this has nothing to do with power or balance, but just style preferences; a desire for dice rolling, no matter how skewed, than it is about someone succeeding or failing.
Honestly perfectly stated. A 0.25% chance of failure is negligible, but having to roll for it just to avoid it is a style I prefer. My players have yet to have an issue with it.
That is my point, they practically never fail, having them fail on a 1 is them practically never failing, but still the possibility exists.
What I'm disagreeing with is the notion that 5% is "practically never". A magician that fails basic tricks on stage 5% of the time is incompetent. A singer that screws up the lyrics to their own songs 5% of the time is incompetent. A trapeze artist that misses 5% of their jumps is incompetent. An easy task (DC 10) is easy enough that an untrained commoner will still get it right half the time. 5% failure rate on is ridiculous for an 11th level Rogue with Expertise and Reliable Talent.
By 11th level "characters have reached a level of power that sets them high above the ordinary populace and makes them special even among adventurers." They should be extremely good at what they do. Again, monks are literally running up walls by then, and they don't fail that 5% of the time.
My point still stands, I don't think anyone should be able to auto-succeed at anything. I have no problem with them practically never failing, but I do believe they should always be able to fail.
Sure, and that's your prerogative as a DM. But, as has been mentioned before, bear in mind that "1s as auto-failures" is not "practically never failing", it's failing 5% of the time. Consider, for example, a master chef. Would you say it is reasonable that they failed to properly cook one out of every 20 plates? Now, lets move a bit closer to the game... should a level 20 bard fail to entertain a crowd once out of every 20 performances? Note, we're not talking about influencing them any way, just basic entertaining. Or, consider a brutish half-orc, level 20 Barbarian, with 20 STR, maybe even higher due to magic items or whatever, and trained in Athletics, arm-wrestling a puny wizened old commoner, with 4 STR, and no Athletics training whatsoever. Is it reasonable that once out of every 20 times, the old guy wins? Now consider a master thief, who's spent his entire life honing the art of lockpicking, and who has near super-human(oid) dexterity. Is it reasonable he fails one out of every 20 attempts at picking a simple, practice lock? Or, let's say you're using the variant rule, and "1s are spectacular failures": is it reasonable that he opens entry-level practice locks 19 out of every 20 times, and breaks the lockpicks the rest of the time?
Having PCs that can always fail at even mundane tasks can be terribly fun, definitely. But it's a different type of campaign, for different types of players. Many (maybe most?) players would rather have the option of ensuring success for certain parameters, giving them something to work towards, and a feeling of achievement when they get there. I imagine playing a rogue with expertise in thieve's tools must be a fun and paradigm shifting experience: locks are no longer obstacles, but advantages: a locked door is an open door for the party, but not for the enemy. Or with expertise in Persuasion: when everybody's willing to do my bidding, how do I know who's really with me? (See: Marvel's Purple Man)
Yes, it creates more challenges for the DM (and that alone could be reason to avoid it by houseruling), but it's not less reasonable than "no matter how good you get, you'll always fail 5% of the time", and certainly no less fun than "ooh, let's see if you fail to walk and chew bubblegum simultaneously this time! Ha ha!".
What I'm disagreeing with is the notion that 5% is "practically never".
I think this is one of those things where perspective can be flawed by not understanding how dice probabilities work, because yes, a 5% chance per roll is not even remotely "practically never" - it's more like "a few times a night."
The last session I ran was 4 hours in length. I rolled maybe 20-30 d20 rolls, and 4 of them were natural 1s. A player of mine rolled 3 natural 1s of his own, even though he only rolled about 12 or so d20 rolls. The other 3 players at the table didn't even roll a single natural 1 between them. That's a relatively typical sampling of how frequently a particular 5% chance per roll thing happens. (Note: I've got an accurate count of how many natural 1s came up because the system we were playing has mechanics that only come into question when a natural 1 is rolled, so I remember how many times we had to see if something extra happened - things like if you roll a 1 on your defense roll and your total doesn't at least match the attack roll against you, the attack gets to make an extra attack against you, which makes the chance that something extra happening actually matters a lot lower than 5% per roll.)
"In the fragile reality of Discworld, and with the gods who like to play games, a million-to-one chance succeeds nine times out of ten." - Terry Pratchett.
Sure, and that's your prerogative as a DM. But, as has been mentioned before, bear in mind that "1s as auto-failures" is not "practically never failing", it's failing 5% of the time. Consider, for example, a master chef. Would you say it is reasonable that they failed to properly cook one out of every 20 plates?
Once of of every 20 times the master chef tries to cook, someone interrupts them at a crucial point in the process, ruining the dish.
Now, lets move a bit closer to the game... should a level 20 bard fail to entertain a crowd once out of every 20 performances? Note, we're not talking about influencing them any way, just basic entertaining.
Once out of every 20 times the 20th-level bard tries to entertain a crowd, he comes down with strep throat, his instrument breaks (possibly sabotage), he annoys a bad-tempered old wizard and gets silenced, the crowd isn't in the mood to be entertained, or the crowd hates his instrument/style/face/whatever.
Now consider a master thief, who's spent his entire life honing the art of lockpicking, and who has near super-human(oid) dexterity. Is it reasonable he fails one out of every 20 attempts at picking a simple, practice lock? Or, let's say you're using the variant rule, and "1s are spectacular failures": is it reasonable that he opens entry-level practice locks 19 out of every 20 times, and breaks the lockpicks the rest of the time?
Once out of every 20 times the master thief tries to pick a practice lock, he gets interrupted, he gets overconfident, or it wasn't a practice lock after all.
Or, consider a brutish half-orc, level 20 Barbarian, with 24 STR, maybe even higher due to magic items or whatever, and trained in Athletics, arm-wrestling a puny wizened old commoner, with 4 STR, and no Athletics training whatsoever. Is it reasonable that once out of every 20 times, the old guy wins?
I don't even remember whether it was a D&D book, a GURPS book, or someone on this forum that said that dice are only for when there's a chance of success and a chance of (meaningful) failure. Applying that would lead to not even rolling for any of your examples. I did enjoy coming up with justifications for failing 1/20 times in those circumstances, though.
Rollback Post to RevisionRollBack
"Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
What I'm disagreeing with is the notion that 5% is "practically never".
I think this is one of those things where perspective can be flawed by not understanding how dice probabilities work, because yes, a 5% chance per roll is not even remotely "practically never" - it's more like "a few times a night."
The last session I ran was 4 hours in length. I rolled maybe 20-30 d20 rolls, and 4 of them were natural 1s. A player of mine rolled 3 natural 1s of his own, even though he only rolled about 12 or so d20 rolls. The other 3 players at the table didn't even roll a single natural 1 between them. That's a relatively typical sampling of how frequently a particular 5% chance per roll thing happens. (Note: I've got an accurate count of how many natural 1s came up because the system we were playing has mechanics that only come into question when a natural 1 is rolled, so I remember how many times we had to see if something extra happened - things like if you roll a 1 on your defense roll and your total doesn't at least match the attack roll against you, the attack gets to make an extra attack against you, which makes the chance that something extra happening actually matters a lot lower than 5% per roll.)
You do know that as a party you guys hard an 8% chance of a 1? That is also assuming your other players rolled about the same amount of times, I adjusted for rolls you could say you had a 6-11% chance of getting a 1 between four players.
Out of curiosity, how many of those ones were limited to just skill checks?
You do know that as a party you guys hard an 8% chance of a 1? That is also assuming your other players rolled about the same amount of times, I adjusted for rolls you could say you had a 6-11% chance of getting a 1 between four players.
I know that in a span of 70 d20 rolls (which is my estimate of how many we had as a group last session), us getting the 7 total that we did is a thing that has a 3.7% chance of happening, but rolling at least 7 natural 1s in a set of 70 d20 rolls is a 6.04% chance. (unless my calculator is broken.)
Out of curiosity, how many of those ones were limited to just skill checks?
Not that it is actually relevant, but zero. Skills in that system are rolled with a d100, rather than a d20. Whether a d20 is being rolled for an attack roll or for a skill check doesn't change the probability a 1 will get rolled - the point is that for a set of 70 actions that were rolled, 1 in 10 being a natural 1 is a typical result despite each roll having only a 5% chance of being a natural 1.
You do know that as a party you guys hard an 8% chance of a 1? That is also assuming your other players rolled about the same amount of times, I adjusted for rolls you could say you had a 6-11% chance of getting a 1 between four players.
I know that in a span of 70 d20 rolls (which is my estimate of how many we had as a group last session), us getting the 7 total that we did is a thing that has a 3.7% chance of happening, but rolling at least 7 natural 1s in a set of 70 d20 rolls is a 6.04% chance. (unless my calculator is broken.)
Out of curiosity, how many of those ones were limited to just skill checks?
Not that it is actually relevant, but zero. Skills in that system are rolled with a d100, rather than a d20. Whether a d20 is being rolled for an attack roll or for a skill check doesn't change the probability a 1 will get rolled - the point is that for a set of 70 actions that were rolled, 1 in 10 being a natural 1 is a typical result despite each roll having only a 5% chance of being a natural 1.
My math might be bad, but I was also trying to be fairly forgiving when I was doing it hence my very forgiving percentages
As for the d100, can I ask how the skill system in that game works? These last two questions are fairly curiosity based and have little to nothing with the rest of the conversation.
You do know that as a party you guys hard an 8% chance of a 1? That is also assuming your other players rolled about the same amount of times, I adjusted for rolls you could say you had a 6-11% chance of getting a 1 between four players.
I know that in a span of 70 d20 rolls (which is my estimate of how many we had as a group last session), us getting the 7 total that we did is a thing that has a 3.7% chance of happening, but rolling at least 7 natural 1s in a set of 70 d20 rolls is a 6.04% chance. (unless my calculator is broken.)
Out of curiosity, how many of those ones were limited to just skill checks?
Not that it is actually relevant, but zero. Skills in that system are rolled with a d100, rather than a d20. Whether a d20 is being rolled for an attack roll or for a skill check doesn't change the probability a 1 will get rolled - the point is that for a set of 70 actions that were rolled, 1 in 10 being a natural 1 is a typical result despite each roll having only a 5% chance of being a natural 1.
My math might be bad, but I was also trying to be fairly forgiving when I was doing it hence my very forgiving percentages
As for the d100, can I ask how the skill system in that game works? These last two questions are fairly curiosity based and have little to nothing with the rest of the conversation.
I'm curious, too.
Rollback Post to RevisionRollBack
"Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
As for the d100, can I ask how the skill system in that game works? These last two questions are fairly curiosity based and have little to nothing with the rest of the conversation.
From a high-level overview, the skill system is ratings ranging from not being able to roll for lack of training (which doesn't apply to things that everyone can reasonably do well enough without particular training, like listening and observing or making an attempt at diplomacy) to 100. Those ratings are derived from ability scores and investment spending build points to roll an improvement die (with the cost of a roll weighted based on the influence of the skill over a typical campaign, so a single roll of something really useful like lock picking costs 6 times as much as something an adventurer isn't likely to benefit from like agricultural skill), with the size of the improvement die growing smaller incrementally as the rating being improved upon grows higher, representing the difficulty of continuing to make significant leaps forward in knowledge/ability as you've already learned so much.
Then when it comes to rolls, there are two types of skill check: standard checks, which are rolled as 1d100 modified by difficulty of the task at hand (ranging from -90 for trivial tasks to +10 for the most difficult task) and succeed if the roll result is less than or equal to the skill rating (example: my wife's character attempting to treat a wound with first aid, an average difficulty check, would roll 1d100-40, and succeeds on a result that is 68 or less - which means she only fails to patch up a wound correctly if there are other modifiers at play); and opposed checks, which are rolled as 1d100+your relevant skill rating against your opposition's 1d100+relevant skill rating, with the higher roll winning (example: an elf sneaking through the forest would roll 1d100+62 (or more, as that's just average elf sneaking with no-investment given their racial bonus to sneaking in natural settings) and my wife's character would roll observation of 1d100+13 (because she's not very observant and wearing a helmet doesn't help your hearing) and likely have no idea there is an elf nearby unless said elf decided to say "hello").
Sure, and that's your prerogative as a DM. But, as has been mentioned before, bear in mind that "1s as auto-failures" is not "practically never failing", it's failing 5% of the time. Consider, for example, a master chef. Would you say it is reasonable that they failed to properly cook one out of every 20 plates?
Once of of every 20 times the master chef tries to cook, someone interrupts them at a crucial point in the process, ruining the dish.
Now, lets move a bit closer to the game... should a level 20 bard fail to entertain a crowd once out of every 20 performances? Note, we're not talking about influencing them any way, just basic entertaining.
Once out of every 20 times the 20th-level bard tries to entertain a crowd, he comes down with strep throat, his instrument breaks (possibly sabotage), he annoys a bad-tempered old wizard and gets silenced, the crowd isn't in the mood to be entertained, or the crowd hates his instrument/style/face/whatever.
Now consider a master thief, who's spent his entire life honing the art of lockpicking, and who has near super-human(oid) dexterity. Is it reasonable he fails one out of every 20 attempts at picking a simple, practice lock? Or, let's say you're using the variant rule, and "1s are spectacular failures": is it reasonable that he opens entry-level practice locks 19 out of every 20 times, and breaks the lockpicks the rest of the time?
Once out of every 20 times the master thief tries to pick a practice lock, he gets interrupted, he gets overconfident, or it wasn't a practice lock after all.
Or, consider a brutish half-orc, level 20 Barbarian, with 24 STR, maybe even higher due to magic items or whatever, and trained in Athletics, arm-wrestling a puny wizened old commoner, with 4 STR, and no Athletics training whatsoever. Is it reasonable that once out of every 20 times, the old guy wins?
I don't even remember whether it was a D&D book, a GURPS book, or someone on this forum that said that dice are only for when there's a chance of success and a chance of (meaningful) failure. Applying that would lead to not even rolling for any of your examples. I did enjoy coming up with justifications for failing 1/20 times in those circumstances, though.
Those justifications are examples of penalties to the dice roll, I would think (being interrupted while doing something either raises the DC, gives Disadvantage, or gives a penalty). Except the silence one, which is just an example of preventing the skill check from even occurring, and the "it wasn't a practice lock after all", which is an example of "that wasn't a DC 10 roll, it was really a DC 25 roll". In any case, even then, were the math on that right, restaurants and music bands would be much less successful than what we see IRL. Five percent is too high a failure rate for the most skilled, from a realism perspective. Again, if that makes the game more fun or interesting for you, then you can certainly houserule it that way, and you definitely should! (And please read no sarcasm into that, honestly.)
The last part, regarding "only rolling when there's a chance of success and a chance of meaningful failure" is precisely what ultra-high bonuses from heavy investment are meant to model. If you've invested enough time and energy into honing your skill at something, it's reasonable that some activities in which another character would have a chance of success/meaningful failure are trivial for you, and there is no chance of meaningful failure. Put another way, we don't roll for things like building a campfire in normal settings, it's just something we do. But in a setting where it's always raining heavily, for example, rolling a, say Survival check to light a campfire might make sense.
Sure, and that's your prerogative as a DM. But, as has been mentioned before, bear in mind that "1s as auto-failures" is not "practically never failing", it's failing 5% of the time. Consider, for example, a master chef. Would you say it is reasonable that they failed to properly cook one out of every 20 plates?
Once of of every 20 times the master chef tries to cook, someone interrupts them at a crucial point in the process, ruining the dish.
Now, lets move a bit closer to the game... should a level 20 bard fail to entertain a crowd once out of every 20 performances? Note, we're not talking about influencing them any way, just basic entertaining.
Once out of every 20 times the 20th-level bard tries to entertain a crowd, he comes down with strep throat, his instrument breaks (possibly sabotage), he annoys a bad-tempered old wizard and gets silenced, the crowd isn't in the mood to be entertained, or the crowd hates his instrument/style/face/whatever.
Now consider a master thief, who's spent his entire life honing the art of lockpicking, and who has near super-human(oid) dexterity. Is it reasonable he fails one out of every 20 attempts at picking a simple, practice lock? Or, let's say you're using the variant rule, and "1s are spectacular failures": is it reasonable that he opens entry-level practice locks 19 out of every 20 times, and breaks the lockpicks the rest of the time?
Once out of every 20 times the master thief tries to pick a practice lock, he gets interrupted, he gets overconfident, or it wasn't a practice lock after all.
Or, consider a brutish half-orc, level 20 Barbarian, with 24 STR, maybe even higher due to magic items or whatever, and trained in Athletics, arm-wrestling a puny wizened old commoner, with 4 STR, and no Athletics training whatsoever. Is it reasonable that once out of every 20 times, the old guy wins?
I don't even remember whether it was a D&D book, a GURPS book, or someone on this forum that said that dice are only for when there's a chance of success and a chance of (meaningful) failure. Applying that would lead to not even rolling for any of your examples. I did enjoy coming up with justifications for failing 1/20 times in those circumstances, though.
The last part, regarding "only rolling when there's a chance of success and a chance of meaningful failure" is precisely what ultra-high bonuses from heavy investment are meant to model. If you've invested enough time and energy into honing your skill at something, it's reasonable that some activities in which another character would have a chance of success/meaningful failure are trivial for you, and there is no chance of meaningful failure. Put another way, we don't roll for things like building a campfire in normal settings, it's just something we do. But in a setting where it's always raining heavily, for example, rolling a, say Survival check to light a campfire might make sense.
Exactly. (That last sentence of mine was meant to convey/indicate/suggest that I came up with those justifications purely for my own amusement and that I wouldn't actually make my players roll in those situations. Though, now that I think about it, I would probably make the bard and the chef roll to see how good a job they did).
Rollback Post to RevisionRollBack
"Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
Born in Italy, moved a bunch, living in Spain, my heart always belonged to Roleplaying Games
I think this is one of those things where perspective can be flawed by not understanding how dice probabilities work, because yes, a 5% chance per roll is not even remotely "practically never" - it's more like "a few times a night."
The last session I ran was 4 hours in length. I rolled maybe 20-30 d20 rolls, and 4 of them were natural 1s. A player of mine rolled 3 natural 1s of his own, even though he only rolled about 12 or so d20 rolls. The other 3 players at the table didn't even roll a single natural 1 between them. That's a relatively typical sampling of how frequently a particular 5% chance per roll thing happens. (Note: I've got an accurate count of how many natural 1s came up because the system we were playing has mechanics that only come into question when a natural 1 is rolled, so I remember how many times we had to see if something extra happened - things like if you roll a 1 on your defense roll and your total doesn't at least match the attack roll against you, the attack gets to make an extra attack against you, which makes the chance that something extra happening actually matters a lot lower than 5% per roll.)
" In the fragile reality of Discworld, and with the gods who like to play games, a million-to-one chance succeeds nine times out of ten." - Terry Pratchett.
Roleplaying since Runequest.
"Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
Tooltips (Help/aid)
I know that in a span of 70 d20 rolls (which is my estimate of how many we had as a group last session), us getting the 7 total that we did is a thing that has a 3.7% chance of happening, but rolling at least 7 natural 1s in a set of 70 d20 rolls is a 6.04% chance. (unless my calculator is broken.)
Not that it is actually relevant, but zero. Skills in that system are rolled with a d100, rather than a d20. Whether a d20 is being rolled for an attack roll or for a skill check doesn't change the probability a 1 will get rolled - the point is that for a set of 70 actions that were rolled, 1 in 10 being a natural 1 is a typical result despite each roll having only a 5% chance of being a natural 1."Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
Tooltips (Help/aid)
"Any society that would give up a little liberty to gain a little security will deserve neither and lose both" -- allegedly Benjamin Franklin
Tooltips (Help/aid)