Where do you land on the “natural 20s are always successes/natural 1s are always failures” debate? Thanks so much to OnlyCrits for sponsoring this video! Visit www.onlycrits.com/supergeekmike and use the promo code SUPERGEEK at checkout! www.onlycrits.com/supergeekmike
If success or failure is impossible, why are you making them roll (unless you're solely rolling to determine the degree of success/failue) - At the same time... Success and failure can mean different things in different contexts. In your 'I ask the king to adopt me' example, that's always going to _fail_ but I might ask for a roll to see how badly this goes for you, where a success against whatever DC I set in my head leading to the King finding it amusing rather than leading to a prison break episode next session (Though I'd probably have something in between those outcomes.)
I don't like it because it does imply that there's always a chance of success, and I think it leads to players declaring rolls instead of actions. But I'm a weirdo that doesn't even like it as a rule for attack rolls and saving throws. So what do I know?
I think the lowest roll always fails the highest roll always succeeds would be better if dnd wasn't 1d20 based. It is just too probable to roll crits with a 1d20, making them way less special in my eyes.
Sometimes, a natural 20 isn't success but simply the best possible result. In a case where the stated goal is impossible, such as the classic Persuasion to get the king to give up his throne, a natural 20 or other high roll may have him laugh it off or like your ambition whereas a low roll may have him interpret it as a threat or otherwise take offense.
This has been my stance as well. There's Also the degree of failure that can be important as if a task is impossible but a player wants to do it then how well do they fail becomes important especially in dangerous attempts.
Conversely, I like using DC 1 checks for tasks where there is no chance of failure. A 1 means they succeed, but almost fumble and a 20 means they totally nail it with some extra flair on top.
I will sometimes tell my players "roll to see how cool you look" when they're doing something that shouldn't be able to fail at. Like if my rogue wants to do some parkour and I've already established that these rooftops are pretty closely spaced, I'll still let him roll, because he WANTS to roll. He took that acrobatics proficiency for a reason and he wants to use it, even just to show off. That way I make it clear to the player that I'm respecting his character build, I'm not going to let him fail at something his character should be an expert at, but at the same time the roll isn't meaningless. If he rolls high I describe him looking like a gymnast with flowery language, and if he gets a 1 I can describe how he mistimes his jump and hits the next roof harder than he meant to, causing him to stumble before he catches himself. He still 100% makes it to the next roof, but it looks way less cool. Another version I'll do sometimes is "roll to see how long it takes." This can actually work for something the players shouldn't be able to fail, and something they shouldn't be able to succeed. If they're doing research in a library, a high roll means they find the answer in ten minutes, or within ten minutes they become certain this library doesn't have the answer at all. A low roll means they spend hours there before finally finding the right book, or they waste hours searching before realizing they can't find the answer here. That way I can give them the answer I want to give either way and thus progress the story, but I'm still rewarding someone who put a proficiency in investigation by letting them use that skill. Because we all know how frustrating it is to have a skill and never get to use it.
What's funny is that the "roll to see how long it takes" is the opposite of taking 10 or taking 20. If you roll a 20, it goes quickly, but if you take 20, it takes a long time.
It seems so simple, but I actually never thought about using the dice roll as outside forces as a base. I had a player with a +15 Athletics try to climb a destroyed house, and he rolled a 1, I just narrated him not being able to find firm ground, and he felt really bad. Looking back I think that if I just narrated him trying to climb, but the wreckage just kept falling and he manage to dodge out of the way or break it before it hit him would feel like a descent outcome. Even in failure his character would still manage to do something cool.
This also works well as a way to justify not allowing skill dogpiling. I first hit on the concept in my own games when I had to articulate why a player couldn't make multiple attempts at picking a lock: "Because you failed that first roll, this is an uncommonly hard lock."
There have definitely been times I've asked for a roll not because an outcome was uncertain, but because not asking for it gave the players information that I thought would hurt the scene. The most obvious would be a character sneaking around and scouting out unknown areas that end up being uninhabited; I know there's no one around to hear them, but I still say, "Roll stealth" because if I don't, I'm functionally saying, "The coast is clear" and removing most of the tension. (And, by extension, it means every time I do ask for stealth, they know there's something around that could detect them, even if they haven't detected it yet.)
Exactly. A lot of people making this discussion tend to miss the harmful effect of this rule on the essential restricted flow of information. You gave a great common example for this, and I'll add to this revealing difficulty levels and ACs, when you say to your players: "well, may I have the paladin and the wizard roll stealth, but not the monk and the ranger, you'll be OK.". Same goes for looking for traps or taking a watch at night. A similar problematic case, much rarer but very crucial, is when an important thing is hidden or disguised. If the BBEG is trying to escape while leaving a decoy behind for the party to mess with, you don't want to reveal it right away with "no need to roll, he is dead."; Similarly, if he is spying on the king looking like a simple maid - you won't want the barbarian to accidentally stumble on it while trying to push through her. Unless you do want to; And this is the point, the DM should have the absolute power regarding what information to reveal, when and in what rate and timing, and not be forced to it by the rules. Only by the actions of the players.
It's worth considering that, even if the DM knows that what the player is trying to accomplish is impossible (or impossible for them to fail at), it might still be worth calling for a roll if there's a range of possible outcomes beyond simple success or failure. Take your entertainer bard, for example: she doesn't need to roll to get free room and board, but if she wants to play for tips on top of that, then the performance roll would determine how much she gets that night. After all, that's going to be affected by factors outside the character's control, like how rich the crowd is, and what kind of mood they were in before she started playing. Or, to use the "players demand the king hand over all his riches" example: he's never going to agree to something that ridiculous, but a persuasion roll could determine whether the king is in a good enough mood that day to be amused/impressed by the party's audacity, or if he has them thrown out on their ears for their insolence. I'm fairly certain something like that has happened at least once on Critical Role. In a later C1 episode, two PCs compete against each other to try and seduce the same NPC, and after some roleplay Matt has them both roll persuasion checks. Now although Matt doesn't say this explicitly, I'm fairly certain that there was no roll that would have resulted in the NPC saying yes, if only because I don't think that's behavior Matt wants to encourage in his players. But the PC who rolled extremely high got a "I'm flattered but no" in response, and the NPC remained friendly towards him afterwards, going to so far as to later introduce him to a friend of hers who was more open to the idea of a one night stand. Meanwhile, the PC who rolled low was essentially told to eff off, and got the cold shoulder for the rest of the night.
I’ve also cut down on the amount of rolls I call for. A lot of the time I use the degrees of success or failure and use their roll to determine how long something takes. Roll a 1:”It takes you a few hours, but you find that cave you were looking for.”
I'm currently playing a rogue with an entertainer background, and what my DM and I decided on is that the performance roll is not for how well she performs, but how well the audiences receives the message she's trying to convey with her performance. This has worked out really well and has culminated in her using her music to start a gladiator revolution.
I was thinking a similar thing. For the bard example, let them roll. That's how much gold they receive. Nat1: "The audience is appreciative, but not particularly wealthy. You get a handful of copper and silver, that comes out to about 1 gold." Nat20: Roll again. Multiply the results together. So you roll a 20 and a 12. "You see a number of very finely dressed patrons in the audience, who are tapping their feet appreciatively. At the end of the performance, they come forward and drop several coins into your hat. You find a total of 240 gold after everyone drifts away."
I am definitely going to rethink when to call for rolls going forward; I hate locking essentially story stuff behind a roll, and I want to reward my players for their choices. I *really* liked your player's argument about Performance checks; maybe if the crowd was hostile or very different culture should that sort of thing require a roll.
As long as you don’t write yourself into a dead end it is perfectly fine to put things behind roles. and put out consequences for failing them. Your heroes have to go through a certain amount of personal risk and the dice simulate that. If you want to reward good choices you should do it in the dice roll. Advantage and disadvantage are one thing. But you’re the DM. If you only think they have a slight advantage you can just give them a plus to aura +3 it doesn’t have to be +5. Which is effectively what rolling the d20 two times gives you is essentially statistically a +5
I've always seen natural 20s as not necessarily "success," but just the best possible outcome. Likewise, natural 1s would be the worst. This comes into play mostly when it comes to contested rolls or interactions where there can be varying degrees of success. Then there is combat in which 20s always hit, and 1s always miss.
Yeah agree. In the 'roll a nat 20 on the persuasion check to ask the king to give up his throne' example; the Nat 20 doesn' t mean the king agrees and gives the players what they want, the Nat 20 just means that the king isn't totally furious and executes them on the spot !
One thing I've thought about a lot on the issue of whether a DM should or shouldn't have a player roll for something that a Natural 20 can't accomplish is the possibility of having them roll to demonstrate that a thing is possible to accomplish even if a character can't right now. If the party artificer or wizard wants to understand the magic bound to an artefact, for instance, but they don't the requisite skill in Arcana yet, just shutting them down might tell them that this is an unsolvable problem when it's actually a problem that you want them to solve in the future. Asking them to roll and then telling them that there are elements of magical theory implicit to the item's design that are beyond them implies that those elements can be solved, if not at their current level of skill. Thus, they have the motivation to find out more about the artefact or the magic and come back to solve it several levels later when they have a +16 bonus (complete with the dip into rogue for Expertise, Flash of Genius, Enhance Ability for advantage and Bardic Inspiration just in case).
love this, just had this discussion last week with a friend about a sailor climbing a mast while not in combat and they had lots of time. they cant fail that, and if they did it would be no fun so theres no roll needed. unless you want to have the roll be degrees of success like a 1 is it takes them 2 tries and someone helping them. is that fun? go for it is it not? then dont. they shouldn't be able to fail that action they want to do if theres no risks or story involved.
I call for extra rolls pretty often cuz my table is in the "rolling dice is more fun than not" camp, but I take my cues from Dungeon World and usually include degrees of success, and sometimes cost trade-offs - eg if they roll a 14 on a DC15 Arcana I'll give them the same info about a magic object, but remove a key detail... Or show hints of a key detail they might be able to uncover if they would like to clutch it in two hands and line up with some finger-holes (beating a 15 would've explained the finger holes. Not beating it means they don't know and will have to take a leap of faith). I'm also fond of making low rolls in tense situations "success at a meaningful time cost," and the good old "you COULD easily do that, normally, but this place is not normal. Roll to make sure you DON'T roll a 1." Basically leaving a 5% chance for shenanigans and muckery. And yeah I don't like when players say "can I roll X on it?" I will ALWAYS turn that around and ask "what information are you trying to discover/what outcome are you trying to set up for?"
I think your tangent on failure being the stopping point of the adventure gets to the heart of this problem. Once rolling has a more nuanced outlook, this issue dissolves pretty quickly. Taking your example of the Entertainer background, I would still have the player roll for their performance just to see what the audience reaction is. It wouldn't take anything away (the character has invested in that Background and we're not going to punish that) but I would provide upsides if the character rolls well (say, that room being one level above their current lifestyle as compensation for drawing a large crowd). Just because there's no room for failure doesn't mean there isn't room for a roll; just because there's set conditions for a pass and fail mechanically doesn't mean those conditions have to be good and bad! The other things I'd say is that success or failure should be a mixture of both character and outside factors. Having technical proficiency is only one measurement of success. A performer can play the most exquisitely composed piece, but if it's to a bunch of commoners who can't appreciate those details then it may fall flat. Both knowing the audience you're performing for and choosing what would be best to perform for that audience and choices that the character is at fault for _and_ are effected by the outside world. A great way to find that balance, I've found, is to let the player tell you why the check didn't go over as planned. If you let them decide the narration for the outcome they'll often be much more receptive to letting their character fall flat or find a middle ground between character and outside factors.
I just realized that I've technically always done the "nat 20 always success, nat 1 always fail" because if the DC is so high they cannot succeed, I'll let them know that what they're tying to do seems impossible. And if the DC is so low they cannot fail, I just describe them succeeding. That way my players know that when they're rolling that it will actually mean something about what happens. Only if a player still wants to try something that I said is impossible, which in that case it will be a "how much to they get hurt" check or a check to see if they can make the challenge a little easier for the next person depending on context.
Yeah, it seems surprising to me that there are tables out there were that isnt the case were DMs allow their players to rol even if success or failure is imposible
my thing is, sometimes the extra bonuses get you to the DC 30 (labeled Impossible in the DM's Guide) before you hit 20 and sometimes only if you hit 20, but how would you know if they didn't roll vs...yeah, I'm not gonna make you roll because there is no chance in hell you'll persuade the guy who doesn't speak Elvish whatever you're persuading. Idc how high you roll, it's not happening
There are a few circumstances when the auto-succeed/auto-fail rule causes problems: opposed rolls, graduated rolls, and variable bonuses. Opposed rolls I think is pretty clear why it could be a problem. Graduated rolls, or rolls with different grades of success, invalidate the idea of a success/failure binary. And with variable bonus, like those from Guidance and Bardic Inspiration, there's a chance for results significantly higher than a DC which makes a task normally impossible for a character, but it's only a chance. If someone can't succeed without Bardic Inspiration, do you refuse the roll? If they get a 20 but don't get the DC, do you give it to them? I didn't like the rule. I do encourage the use of the passive score or auto-average result in unstressful situations, though. Like the entertainer bard, if a task is something you have enough experience with you can reasonably allow the action to take place with the assumption of a 10 on the roll if there's nothing happening that would disrupt them.
I really like Fabula Ultima's approach where you can invoke your "Traits and Bonds" (who you are and the strong relationships you have) in order to spend a limited resource to ease related check (this guy killed my parents so I can punch him in the face even better) and to get easier access to equipment (a herbalist is more likely to have a specific flower on them than a princess who just ran away from home). I have no idea how to put something similar into DnD
On the "Entertainer performance check" - I like to instead make the check relate to "extras" they might be able to get. After considering they usually have really good +'s to performance, on something like a 16+, or 20+, the owner of the tavern themselves is so entertained maybe they shout the whole tavern a round of drinks and make it an event (that the rest of the patrons would then be extra friendly towards the performer), or even giving their best room(s) to the party. It doesn't necessarily have to have a gameplay impact, sometimes those extras are enough for people to have a more memorable moment. But it otherwise can still have more impact: better rapport with the owner which may help down the line, or inspiration or some other bonus for spending the night in relative luxury to what they were expecting. Also doesn't have to be every time if you need to move the game along slightly faster!
I believe in degrees of failure and degrees of success. If someone wants to persuade someone else of something that will always fail, they can still roll to see if the person takes offense or politely declines. If you try to do something that you cannot fail, you might do it in such an amazing way that the group gains some additional benefit. You might be so good at tracking that you would never be able to lose the tracks you are following, but if you roll well enough you might gain additional insight about the creature you are tracking.
My DM does a little bit of a mix of both re: whether success or failure is related to the character or external effects, depending on the circumstances and even the individual character. Like, our warforged DMPC is extremely awkward and physically and mentally out of place because of her backstory, so rolling low on stealth checks is almost always a mistake on her part, whereas in combat the DM tends towards low rolls and especially nat1s being the opponent blocking the attack in a cool way. One time rolling a nature check to identify a creature an NPC was describing, my character hadn't gotten any sleep because she was on first watch when we got attacked and we'd been walking all night since then, so it was easy to say with a low roll that my character was just too exhausted and burnt out to think straight. My favorite though, is if we're rolling for something that would come completely naturally to us, like artisanal skills or my monk-cleric meditating to commune with her god, the roll is less about success and more about how long it takes; a low roll could mean a few hours of the monk trying to get settled in an uncomfortable environment vs. immediately hitting the sweet spot to focus, or our artificer jeweler helping to make a magic item could reduce or extend the time needed to complete it. And sometimes it is just a moment of clumsiness because hey, it happens to the best of us.
I am firmly in the camp of: A) don't make a roll the GM doesn't call for B) 1s and 20s are not automatically failures or successes, respectively Keeping in mind the optimal rule for Ability Skill checks, I may have you choose a different ability score than the default for a skill check, based on varying circumstances/factors. I am also a fan of occasionally limiting skill checks to those who are at least proficient in a skill; both to reward player choices and to represent the likelihood of access to that knowledge. Definitely agree with you that it's important to decide whether a roll is even called for. One of the things I miss from 3.X (don't remember if it was in 4E), was "Taking 10". Basically, your character would spend 10 minutes on a task, and instead of rolling, the check was treated as if you rolled a 10 on the d20. Great video, as always.
I've definitely been guilty of calling for rolls when they weren't needed in the past and it's something i have become more conscious of recently. Because they do just slow the game down to a crawl and doesn't allow players to actually feel badass and heroic, especially once a campaign starts brushing the higher tiers of play. The comments about not locking plot relevant info behind a single roll have also given me reason to rethink some of the ways I structure my adventures. It hasn't always been the case but there have definitely been times when there's important info the party really needs in order to move forward in the current story arc but then no one rolls high enough and I'm left a bit unsure how to proceed. I want to honour the rolls because I generally really don't like fudging things when the parameters have already been decided but without this clue the adventure is left foundering a bit. This was already happening much less as I've grown in experience but this video was very helpful for putting some stuff into perspective. So thanks!
I really appreciate that all your videos are released with subtitles! I have hearing difficulties and its very disappointing when some of my favorite creators wait a few days to subtitle or dont do it at all
I generally agree with all your points and wouldn’t argue any of them. I do, however, want to make a case for the “dramatic impossible roll”. Generally it’s good to let your players know you’re doing it, though as always, depends on the group. There is something I find compelling as both a player and a GM about the prospect of the characters going up against insurmountable odds with no chance of success for dramatic purposes. Even if that nat 20 is rolled, at best you get a dramatic moment. I think this sort of thing works best with like, a first confrontation with a big villain or something like that. And this shouldn’t be done all the time, but I find sparingly, they can make great moments if everyone is on board
Started writing the comment, thinking it would be short... Well, sorry, but I like the topic, and I had some thoughts. I don't really see a problem with, occassionally, undermining the characters. No matter how eloquent, how confident, how skillfull a speaker you are, sometimes you'll get something caught in your throat, and if done right, it can be hilarious. Another thing I don't necessarily agree on, is the idea that if you cannot succeed or cannot fail, you should never ask for a roll. True, these cases may not call for a roll. Maybe you just give it to them, or you just say: "Not possible for you". But, sometimes, success or failure is more nuanced than simply a DC telling you "Yes you have succeeded" or "No, you failed". Yes, there is a point, where an action would count as a success or a failure. But I find that too black and white. If we take your example of persuading the king to adopt a player and make them his heir. There is no way this should succeed (unless maybe certain exceptional circumstances are present in the story). A nat 1 could signify that the king is highly offended by the very notion that such a feeble underling could even utter such words, even as a joke, and he would have you hanged the next morning for your insolence. A nat 20 could mean that the king and his court starts laughing, and he flips you a gold for giving him the best laugh he's had in a while. Neither of the 2 scenario's are a success, but a roll could still be used to determine certain factors. If we could take this kind of approach to searching for traps you could say: 1-4: You are confident that there no traps. 5-9: You find no traps, but you feel you've missed something. 10-14: You find a pressure plate on the floor, disguised as a tile, which sets off arrows that shoot down the corridor you're in. It took 30 seconds to find 15-20: Same as 10-14, but you spot it almost immediately, and you may get some extra details about the workings of the trap. In this case, there's still a hard cut off with a DC10, which let's you succeed in finding the trap, but the dice roll matters more than just yes/no. It doesn't have to be this strict, you could just set a DC and then do extra based on your instinct at the moment, the further away they are from the DC on either side. But also, if you use this kind of system, you can still implement dice rolls sometimes, where the player won't be (mathematically) able to either succeed or fail. By mathematically, I mean modifier wise. If they have a +11 to a skill, and the DC is 10, they cannot fail. If the DC is 20 and they have -3, they cannot succeed. If they succeed big time, you can make them feel awesome, if they fail big time, you can make it funny. It also lifts the weight just a bit off of the DM's shoulders, so they don't need to constantly be aware of all the modifiers of all the players. I will say though: The story you told about the bard is a good situation of where you might not want to ask for a roll. They should be consistently good at playing in a tavern and getting something out of it. On the other hand, if you ask for a roll, and they roll well, you could describe how even the drunkest, almost passed out patrons, perk up at the sound of their music and start looking energetic, singing along and cheering. And, occassionally, on a nat 1, they might just break a string on their instrument (or might not). Not a mood killer, as they are a good performer, so they can play it off, and work with the crowd, while they change the string. But maybe that happens, and maybe the passed out drunks stay passed out, while the rest of the patrons still have a good time.
I'm still using 10 min = 10 + mod and 20 min = auto success or 20 + modifier rules from back in the day. It just makes sense that someone with proper time and training can just do the thing. I also took it to mean that the regular rolls usually mean you're in a high pressure situation with little to no time, so it makes sense that failure is more likely to happen.
Great video. Personally, I don't like knowledge being used as a skill to begin with. In my own RPG I'm designing, Knowledge is a stat you get from adding together a certain 3 attributes. Every 5 points of Knowledge, you gain a research point. Research points can be spent to learn a new language, or to give yourself a +5 Knowledge bonus to a subject of your choice. So, rather than "roll to see if you know something", the GM just checks your Knowledge stat. This all came from a moment in D&D when I asked a player to make a history check, they rolled a 1 (total of 2) and I had to tell them that despite a low DC, they did NOT know something everyone from their culture would know. It was funny as it happened, but I couldn't stop thinking about how I should have just told them the significance of that sword...
In a trad game such as D&D, CoC, etc? When failure would be dramatically interesting - A great piece of advice I've seen - I think during one of Johnny Chiodini's Dicebreaker video where they talk Luke Westaway through how to DM while Luke was gearing up to DM for his first time - is never ask for a roll you can't afford the players to fail. In a PbtA or game with a similar philosophy? When a move asks for one, and no other time. (Which also strongly discourages describing failure as incompetence from your players via the "Be a fan of the Player Characters" GM principle which the vast majority of PbtA games have)
The big insight I’ve personally taken is failure shouldn’t mean “the task fails” it should mean “something goes wrong”. There’s a twist, you lose resources, you waste valuable time, monsters wander in, an opportunity closes, etc. I don’t call for a roll if nothing can go wrong, they just do it. I don’t call a roll if nothing can go right, they face the consequences.
One of my favorite rolls I ever called for was "Make a saving throw with whatever attribute you want." There was never any chance of failure, but the player's decision still gave them agency in a scene that was almost entirely flavor.
My erly GMs had the same answer which I found a very easy answer, for the question when do you roll? You roll when both answers are in world realistic and narratively interesting?
My DM had us each roll to identify an NPC race. We got a nat 20 and 3 other rolls over 20. Then told us we all don't know and told us the DC was 30. None of us could have rolled that no matter what. I was a little frustrated because why ask us to roll only to 100% fail. So I felt like just telling us, "None of you recognize him" would have been more effective and kept the mystery.
I remember in one of the first episodes of Oxventrue Blades in the Dark the GM was about to call for a roll and stopped himself saying something like 'never ask for a roll when failure would end the adventure' and that stuck with me. I'm not a GM very often but it makes sense.
I remember implementing a house rule in my game to try to make me stop calling for inconsequential rolls: any time I ask a player for a roll, they can choose to fail, and get something like a stacking Inspiration to use later. It did a good job in that nobody ever took me up on it, which means I never called for a roll that any player cared to fail at.
I use graded DCs, so I have no issue having players roll for something impossible: "failing" with a 23 is different from failing with a 6. (I know d20 systems aren't optimal for graded DCs but it's good enough for home games)
I really like the concept of degrees of success/failure. This allows the player to still make a roll even if a critical won't get them the exact result. Going back to "persuading the king" example, if a PC tries to get the king to give up the thrown, or whatever, a critical success could be that he takes the whole thing as a joke and doesn't imprison the party.
My biggest annoyance is when I roll a 1 but have a +14 in a particular skill (expertise and other fun stuff). Treating that 1 as a full on 1 despite meeting a 15dc is annoying as heck. I put a lot of work into a set skill and it feels invalidated when 1 is turned into a total failure regardless despite my skill. But as for calling a check I tend to word things this way “May I roll a perception check to listen to rumors while I sit in the tavern/ travel the town?” Or “May I roll a nature check to understand what unique nature this jelly has?” Note that I said “May I” to start each. I clarify if asked why and I allow the dm to say no with no hard feelings. I honestly rarely argue with a dm. The only time I did was when one tried to two weapon fight with an enemy and was using rules of npcs having similar rules to players. I am very familiar with these rules. The dm tried to have the npc use a long sword and short sword. I said the rule only allows one to two weapon fight if both weapons are light. I then read off the part of phb when dm asked me to. Like I said, I have used this rule before and I understand a dm has a lot to track so I don’t mind sometimes clarifying these rules when I remember them. I still concede to a dm’s ruling so long as they keep it consistent and it doesn’t undermine the class or background features of a player.
When I have a structured adventure I try to have multiple threads that all converge on the same point. I'm comfortable enough as a DM to also change things around behind the scenes, repurpose a motivation, leave a clue somewhere else and I let the dice decide that. The party missed a thread or a clue earlier. I didn't *intend* this random stable hand to be as key as they are, but the players have ignored another thread entirely or missed a clue earlier on. So I implement it there if the dice roll(s) are successful. In a way I almost look at as a Mass Effect dialogue tree, which may sound cheap if you know how that works, but in essence they can fail and succeed forward the only difference is the route but the destination is always the same. It does require a lot of both planning and improv on my part, but it matters to me so I don't mind.
Overall I think that many DM's are caught up on this "debate" because they don't know what "rolling the dice" represents. If success or failure is guaranteed, don't roll for it, if it doesn't make narrative sense, don't roll for it, it's that simple. The DM is the one that calls for rolls if he deems it necessary (agreeing with what you said in "when the players declare their own rolls" section). If a DM deems inappropriate a player's request, he can just deny it - that alone disqualifies any argument made in the spirit of "what if the rogue tries to convince the queen to hand over her kingdom?" - if there's no roll, there's no risk the rogue could actually succeed in this. Likewise, a DM could just deem unecessary to roll for the barbarian to break down a rotten door. Rolls, as you stated, are there to represent chance and circumnstance, things that are out of the control of the players and, when rolled, dictates what happens (even the DM is henceforth obliged to follow what the dice says). That being said, I respect the opinion of those who disagree with me.
I agree with a lot of this, except the bit about failure being coming from outside the players. I think it really depends. For hits and misses at least, I often describe the narration around the score itself - low rolls are a failure of the player, then the next zone is the "natural" defense of the target (i.e. its natural armor or Dexterity to dodge) and then finally any physical armor above that. It's more of a guideline than a hard rule but I find it helps a lot for combat, at least.
I watched a notetaking video recently that got me thinking about lore checks in a way that I'd like to try. The idea being that the player's notes are also the character's in-game notes. Building on that you can hold that if the player is reading through their notes to find something then the character is. In most situations that is not going to cause a problem. You can sit down and sort through your notes about history related to this 'whatever' before you. But sometimes it would be bad to. The battle is going to be over if you stop to read if you've heard anything about creatures like the one you're facing and critical negotiations are going to lose a lot of steam if you ask the king to stop talking for a bit while you read their family tree. In those moments you roll and, depending on the results of your roll, you get a little time to consult your notes which is counted as you just remembering in the moment instead of reading your notes. It gives reasons to research an area in advance and engage in lore as more than flavour. It also encourages you to share notes in downtime moments and covers those "You weren't there. You don't know that" moments. As for all the info they haven't got notes on (Which should be most things. You don't want them going crazy with notes or else they'll never find what they want when they need it and it would just not be fun) anything they would know they know, anything they wouldn't they don't, and anything in between they may get a hazy detail, popular misconception, or rumour on. They have proficiency in nature and they're a sailor, well then, they know all the common sea creatures and plants, have heard of most of the uncommon and rare ones to various degrees, and the anomalies are just as anomalous to them as to non-sea goers. No roll needed. Normal lore checks are so passive. You don't know it or you do and here's a lore dump about it. This way you have the fun of "You passed the history check, but only just, so as the countess stares at you you've got 15 seconds to look through your notes then you must close them and time resumes". *Cue frantic page turning* (Exact times would depend to the complexity of information being sought but you probably don't want to pause for more than a minute even on a high roll) Not for every group, especially not something to spring on a player with dyslexia, but fun for some groups. Also, with the notes counted as in world, a villain may get hold of them to find out all your group knows and who they rely on.
Sorry. Good manners to link sources when you mention them so the notetaking video: ruclips.net/video/m1kHdQBqzVE/видео.html&lc=UgwS2z8JFBEzJbeDkLF4AaABAg.9pyddy0qG369pz6Bw9j3Z4
I try to call for rolls as little as possible because I don’t want to drag things out. Asking for a roll, them rolling, getting a result, me deciding the outcome etc time after time starts to really add up.
Honestly there's a good portion that ignore the pass with detriments section of the rules as well as the possibility that your not rolling to see if you pass but how well you fail at the task. Players love to try to do outlandish things because they feel they can accomplish anything if the stats are high. Some things are impossible but players like trying to do things. There was someone that I used to follow that had a dman good quote for why you'd let someone roll even if they couldn't pass "Because I knew that you’d feel unsatisfied if I didn’t let you roll, and I thought you might find this outcome fun, even if it’s not what you were looking for,”
My DM has started to ask us what causes us to fail whenever we get 1s on ability checks. I remember my group’s last session the dice were fucking with me the whole game. Especially with skills i was proficient in. +5 performance as a bard and I rolled a 1. And when my DM asked me what causes me to fail I just say I’m not confident enough in my performance to make what I was trying to do (act like a monster was coming out of the alley after my character to scare off a mob) be convincing.
One thing I think often gets left out of this discussion is group rolls. Eg the rogue is definitely going to beat the stealth DC personally, but are they going to do so minimally or are they going to bring up the group average significantly, helping the group pass?
One thing to keep in mind is that even if success/failure is impossible you might still want to determine the degree of success or failure. -The bard always performs well enough to get a free room for the night, but roll anyway to see if you catch the attention of that important NPC. -There is no way the blacksmith would ever accept that offer, but roll anyway to see if he is insulted by the offer and refuses to do further business with the party.
Gotta disagree with you hard on that fight in Fantasy High episode 2. It was effectively a flourish, and (arguably) two people died because that bad call cascaded into a much tougher fight than Brennan intended. I think that instance is much better used to illustrate the point that even master level DMs like Brennan Lee Mulligan still make bad calls sometimes.
I fall on the other side, where a natural 20 is the best possible result, instead of an automatic success. But that's because I believe that sometimes impossible rolls are still worth making. Failure can take many forms, so sometimes a roll is worth making to see how bad the failure is (or how well someone succeeds). If I think something is impossible (or trivial), I might say no roll is necessary, but I might have them roll and say "what you are trying won't succeed, but we need to see how badly it goes for you". And that's the important part---setting up the expectation ahead of the roll.
That fight in Fantasy High with the table jumping played out how it did in part because it was, I believe, the first time Brennan had ever GM’d a 5e combat. And it involves a very fateful roll much in line with your video that just wouldn’t be the same if it had been shut down because it was too far fetched. Personally, I love the euphoric rush of the nat 20, and I love how Brennan honors them. (It’s one small critique I have with Matt Mercer. Nothing quite sucks the wind out of the sails of joy like hearing “For a total of…?”) But Brennan also doesn’t just let his players roll willy nilly. There’s a particularly funny stunt in Fantasy High Sophomore Year that involves a PC attempting to use a ribbon dancer to try to slow their fall from a great height. Brennan has them roll at “quadruple disadvantage,” which was rolling 4d20s and taking the lowest. He said “it’s technically possible for you to roll 4 nat 20s.” 😂 The Cypher System’s approach matches pretty closely to how I approached nat 20s in 5e, but I still brought some things over from that system to my 5e games. In Cypher, skills, attacks, basically with any roll, rolling a nat 20 gets you something extra and a nat 1 triggers a “GM intrusion.” It doesn’t necessarily mean an auto success or failure - you can’t roll a 21+ on the d20 so without modifying the difficulty, a nat 20 would still fail that kind of roll - but it does mean that you get a little something for the occasion, a temporary boon of some kind. And GM intrusions on nat 1s *can be* fumbles, but they can just as easily be positive things, or success at a cost opportunities. Both of them just mean “something significant happens that changes the circumstances of the situation.” When that OneD&D dust up happened, I couldn’t help but think that most arguments people had against the rule could be resolved by just taking with your players? Like, if they want to punch the moon, it’s likely that’s an absurd thing that they and you know won’t work, so asking is either a joke or they’re being intentionally disruptive. If they’re being disruptive…that’s a conversation you need to have with the player about tone and expectations. OR it somehow is something that they have a slim chance at doing, and if they succeed, you get a ridiculous and awesome story to share with each other for a long time. Anyway, great video, as always, Mike. Sorry for the essay length comment 😅
I actually agree that rolls should be within context. Our bard wanted to perform for lodgings. I asked if he has that entertainer background feat and he doesn't. That's when I asked him to perform. Although he's a bard, his idea of a bard was to be a "dad" who uses his traits and spells through being a "dad" projecting "dad" energy, so performing in a tavern isn't his specialty. It's something he might do well in if tried, so I let him try. He rolled below a 10 and I saw fit to deny him free lodgings and earn him 2 copper coins from tavern patrons. Our loxodon barbarian wanted to burst through a wall. I wanted to just allow it, but they are level 1, and I want to give him that feeling of getting stronger, outside of just feats and traits. So I had him roll for it. When he gets to level 5 and tries to bust through a wall like that, he will just do it. But when rping a conversation I would only call for rolls if the situation calls for it. If the npc they're talking to us friendly or even indifferent and isn't hiding anything there is no reason for them to roll if I find their words compelling or reasonable. But I also think players enjoy rolling for their high stats, and if one of your players opted for high charisma stats I would offer persuasion or deception checks more often for them, rewarding their choices. Same for our outlander monk who opted into survival and wants to use it. I'll let him roll more survival checks than I would normally for others, bc using your strong skills is fun for most people I know. But if your success is guaranteed then that takes away any stakes of failure, so the success is less rewarding.
I usually ask for rolls only when both success and failure present interesting options. I'm also a fan of either failing forward or else introducing complications... If the players fail a hacking roll, they may still get the relevant information, but they trip a security protocol, or inadvertantly pick up a virus in their system. I find this especially useful for research and investigation type rolls, so I can make sure they get the minimum they need for the story to continue, but also present both concequences for failure and rewards for a particularly good roll. I don't run D&D anymore, so I have no opinion on Nat 1 vs Nat 20 (though in the dim and distant past of AD&D2E, I think I used the auto-fail/auto-success option.)
I'm not a dm or gm, but I like the idea of degrees of success or failure. And having a roll you absolutely can't succeed might work as a you try something you just can't do, but you don't fail as hard as you might have. Or if you can't really fail maybe you just do the thing instead of being super awesome about it, like simply hopping a gap and then firing your arrow instead of flipping over the gap and loosing the arrow while upside down or something. I don't know if that makes sense or if I explain my view well but hopefully the idea comes across.
I once had a rogue character with really high deception & persuasion, but during interactions with NPCs when I would be using those skills the DM always made me roll play it out instead and then decide if it succeeded or not. My character was the one with an 18 charisma score and skill proficiencies not me. I don’t mind some roll play but let me roll the dice. It was so frustrating I ended up leaving the group.
For context: my players like lore details and we are only playing in a group of three (me, the dm plus two players) so there is enough space for everyone to have some extra rolls that would be too much in a bigger group. History checks are the most prevalent rolls in our game after insight checks, so it’s a very specific kind of game that might not be for everyone. If a roll would be an automatic success I like to ask them if they want to, like: „You would know the basics, would you like to roll for some extras. In my experience that gives them the satisfaction of having some sort of an automatic success based on their background or resources they put into their specific values without taking away the fun of rolling the dice and getting to know some fun details. Setting up rolls this way also helps with the problem of random details seeming more important then they are, because the players know that they are extras. From time to time I will hide some important things in those answers, just to make them feel like they made a random discovery that actually helps them. For things that might be automatic losses that wouldn’t break the game if they work I might say something like: that’s not going to work but you can roll to see if you are super lucky (that being a Nat 20). It’s no problem if it doesn’t work, because we aren’t to many people it isn’t a huge time loss and it makes for some really fun moments if something works, especially if we put some creativity into describing what circumstances led to that unlikely success.
My players tend to make specialists so I tend to do tiered success rolls when they ask to do their specialty. If they roll 9 or lower they get the bare minimum info or item they're after. 10 or higher they get better and better stuff at 10,15,20,etc
I am always surprised that this is still something that needs a discussion. Nice to see your take on the subject, I guess this kind of advice will have to be given again and again. My take: 1. There are no skill checks in 5e. 2. Only the DM can call for a roll, the players can only declare actions. 3. Only call for a roll if the outcome is is unclear AND there is a consequence to the action. 4. Yes, no matter how good the argument was, and how eloquently the player stated it, I require a roll because the character might still not get it across the same way. I will hand out a bonus for good arguments. Edited for sounding like an arrogant ****, I hope this is better.
great points. For me auto-success/failure doesn't always apply because I tend to to use sliding scales for a lot of ability check DCs (eg a History check might give more info the higher you roll, not just a pass/fail threshold). Also, some checks can tell you something even if you fail on a super high roll, eg you're a scholarly type specialising in historic evil cults, and ask if you recognise the symbol on the cultists who attacked the party. The DM calls for a religion check and you roll a nat 20. The DM says that you have never seen this symbol before. Is that a success or a failure? You didn't learn the information you were hoping for, but you learned something about the symbol - you can be confident that it's something very esoteric.
I was thinking a lot about the Nat 20 convince the king to adopt me situation and I think I would respond with: "The king is surprised, smiles and then says, 'Yes... You'll do nicely'" And now begins the plot where the Mad king needed a pawn in his 4d chess to overthrow the neighbouring kingdom. "Yes and" and "Yes but" that shit with game of thrones level drama
What about tension and rolls without obvious consequences? What I mean there is dark cave and players decide to sneak in. Now there is nothing to hide from, so they can't fail stealth (it's just an empty cave), but not rolling it, could lead to similar struggle as initiative (rolling means fight to the dead). Also roll like this might not mean anything right now, but it might be a difference, if a local monster will notice, that something was in its layer.
I think the problem with the Nat20 rule mostly came from the fact that it undermined how DCs work AND requires the DM to intimately know all of the skills and modifiers of their players. For example, if something is DC: 25, that might be relatively easy for a character with expertise, but difficult if not impossible for a character without the required proficiency. Players may be able to stack things like inspiration or guidance onto a roll to increase the chance of success against a difficult roll. You don't necessarily know in advance whether it is mathematically possible for the character to hit the DC in advance, so you can't always say for certain whether or not they should be allowed to roll. Likewise with critical failures, sufficiently skilled PCs may succeed regardless of the natural 1, but others may not. It just isn't always practical to gauge this. Yes, in certain scenarios, like the "seduce the barmaid" or "king give up their crown" examples, you may be fine just saying that person is not interested and no roll is required, but there's other scenarios, like say a stealth or investigation check, where it is not nearly so clear, or where even the game mechanics expect you to make a roll, and that there may be situations where, despite the PC's best efforts, success is impossible, or failure is impossible, given the DC. Someone below mentioned contested rolls as another good example of a potential problem with this system. Suppose you have a wrestling match between a wizard and a barbarian, using athletics, with the wizard having a -2 mod and the barbarian having a +10. Wizard rolls a 20 and barb rolls an 19. On the dice, the barbarian wins 29 to 18, but the nat 20 means the wizard wins anyway? What sense does that make? I'd honestly rather have it that nat20s don't guarantee a hit on attack rolls than making it an auto-success everywhere else.
The reason why there's no automatic success or failure in my game is because I come from Pathfinder 1, where teaming up changed everything. I enjoy and encourage people to use their time in game, ressources and imagination to pile up to achieve a thing. Guidance, bardic inspiration and your natural bonuses might be just enough to succeed. I also believe that in the case of a very high modifier, rolling a roll that might never fail, is a fantastic way to have that player live their character's fantasy. But I always say no to roll that makes no sens. I have therefore never had a situation where a 1 or a 20, not giving failure or success, was problematic. They were usually indicators of things that are out of there reach yet, or beneath them.
Agree on unnecessary Rolling, but the threshold of what is 'impossible' may depend on the DMs knowledge of the characters as well. Does knowledge that the team's bardic inspiration or guidance *could* make the check doable even if it is impossible without influence that? Does knowing that you're not going to persuade that hotheaded noble mean there aren't risks of them taking drastic retaliation if you roll low? It's a fine line, and YMMV depending on your players.
The performance check isn't necessary for the bard but it does seem weird to not role play that scene. Sure nothing bad can happen but rolling really well might lead to interesting history or rumors being divulged
I was surprised back when that happened during that play test for One D&D, especially since I always thought that pretty much every game already handles it that way. That players cannot simply say they jump to the moon and expect if they roll good enough that this could happen. The systems I grew up with, like the Storyteller systems from White Wolf made that explicitly clear. But I guess that was not so clear in D&D... no wonder I always thought that D&D is badly designed.
I agree with your opinion on rolls. BUT, there are some situations where the DC you set as a DM was so high, that even a Nat 20 would not succeed. In very rare and note wrothy occasions only tho For example, on critical role, when Caleb decoded Avantica's code. The DC was straight intelligence and it was a 25. That was to exaggerate how good that code was. If anyone other than Caleb tired to decode it, they would've failed even with a Natural 20. The same happened in Campaign one When Someone tried to a certain important unlock in a special library towards the latter end of the campaign. All that to say that, the DM might expect the roll to be made from a certain character, and a nat 20 might not be enough by someone else. In rare and note worthy occasions. So rare in fact, that Critical role only had one of those occasions in each campaign until now I have nothing to say for nat 1s. Succeeding on a nat 1 or even on a 4, feels wrong, and so nat 1s should, in my opinion always fail
I do not like the automatic success or failure on a 20 or 1 respectively. Especially on unprompted rolls. However in rolls where a character cannot fail like having a skill bonus of 11 on a DC 10 it is fun to have the character succeed but in a ungraceful way. Like hopping over a gap they stumble a step or two but recover their footing. They still succeeded but the rolled 1 adds flavor to the check. On an impossible task such as demanding the king give you his crown and make you king instead. The roll of a natural 20 isn't a success, it makes it so you and your party aren't tossed in the dungeon.
I definitely agree with not asking for a roll where it isn't needed. I am comfortable telling my players not to roll when something is impossible. I'm also comfortable setting a DC like 30, i.e. near-impossible, but if the right (trained) character makes the roll, they could make it on a nat 20 with their +10... But not everyone in the party could. That said, I don't like the auto success/failure on skill checks mostly for failures. When people invest in their training of a skill and get their modifier up really high, I want to reward that. If the DC was 12 and they ask to roll a 1 but they have a +13, they can succeed that task. They're trained. Sometimes I just say, "you don't need to roll, you succeed," but sometimes they just want to roll dice. 😂 Same with something like a high level rogue or an eloquence bard, where they treat any single digit roll as a 10.
We agree on quite a bit. Skills are competence, the dice are the universe being a dick. Give advantage or disadvantage liberally based on small-b background and other factors. Only roll when someone can think of an interesting consequence of both success and failure. Don't roll for automatic PC features. Etc. I also think it's worth calling out games like Fate, 2d20, and others that have meta-currency that allow players to spend a resource to ensure results go better (or take a hit to ensure they go worse). Such a game feature eases many of these concerns. However, you and I seem to vehemently disagree about the actual auto 1/20 rule, for 2 main reasons: 1. I think the potential abuses you rightly object to are inherent to the idea. Seasoned GMs can learn the nuances of "don't bother rolling when ...", but novice GMs tend to default to "sure, roll". The auto-fail/success loophole is just waiting to happen, and teaching GMs how not to fall to that is a much more difficult and lengthy process. 2. I think GMs should be allowed to set numeric difficulties without memorizing all of their PCs' skill capabilities. It may be that a given ill-suited PC can never achieve such a difficulty, even with a nat 20. That should still be a failure. (and vice versa for success).
If there is no possibility for success or failure, I don't bother asking for a roll. I tell people that rolls aren't always connected to whether or not their character 'failed'. A roll in my game is the subtle influences of the gods, the random influence of chaos, the whims or gut feelings that win out. With social encounters, I rarely ask for a roll from a player. I'm more prone to roll if I feel the NPC is on the fence about a decision/feeling that is connected to whatever they're talking about. If the king doesn't know you and you ask to be adopted, I don't see any reason the king would do that so no roll will happen. If you ask the king for a piece of his army to help defeat a monster, he might say yes to show his people that he will do whatever it takes to protect them or he might say no because the monster is still outside the city walls so he doesn't see the reason to waste resources on that. In that case, I might roll depending on what the PCs do/say. Long before that though, I make it very clear that character creation will have perks and consequences. A noble paladin asking the king who worships the same god will have an easier time than a noble paladin that worships another god, but that second paladin would have an easier time than the antisocial fighter that gets into way too many bar fighters. Depending on how you craft your character, there could be instances where the rogue is more trusted than the paladin. In my world, you have rogues that are thieves but you also have rogues that are locksmiths, scouts, and bodyguards (think ninjas). You also have paladins that fall into the religious zealot category. It's up to the players to figure out their characters and the world will react accordingly. If you want to be a religious rogue that feels that their purpose is to protect innocent and inspire people to do good, then that's perfectly doable in my game. Of course, that means that if their actions convey them more as a fanatic murder hobo, well then that's what they really are and the world will act accordingly because the gods do exist and they can tell one of their champions who also has an adventuring party that your fanatic murder hobo is perverting the scripture. There is no roll for that because at that point, several choices were made and the path was followed to that end. Choices matter and seal your fate, not the dice. (With that said, I also encourage players to narrate how their character goofs if they do, not just if they succeed. Had a tabaxi player roll a nat 1 trying to jump over a sewage river in the sewers and the player described the feline doing a tiny hop instead of the epic jump like so many hilarious cat jumping fail videos. I gave inspiration for that.)
I would argue in your example of the entertainer feat that a failed roll would make the influence of the bard to other innkeepers harder to get room and board. If you’re trying to build a living world outside of the PCs, rumours spreading of a talentless bard would make the bard try harder to get their reputation up. Even if the roll is just to convince the next innkeep to let them perform because of less than stellar prior performances.
I love the idea of "the dice represent the reaction of the world", as it means the players don't have to feel bad about their failures. But if we extend that logically, what about their successes? If they roll a crit in combat, that would mean the enemy trips and stumbles onto the player's weapon! Now I'm sure every GM here would have 20s be entirely down to the PCs competency, but I thought it was an interesting comparison.
In fairness, “Appreciate how your competency led to your successes, don’t take your failures as a reflection of you personally” is also just my advice for getting through life without spiraling into anxiety whenever something doesn’t go your way ¯\_(ツ)_/¯
My biggest issue against critical successes and critical failures is the dreaded "super critical failure". I'm talking about being asked to roll sleight of hand to steal keys from a guard, and rolling a 1, and not just failing to pickpocket him. Instead, the GM chortles, and describes me trip forward, pushing the guard over a railing to his death, and thereby summoning two thousand more guards. Who also now have laser guns. But there is no way to write core rules that dictate that critical failures and successes must be "reasonable". So it is better to avoid that entirely by removing that for ability checks.
I didn't know there was a debate. Nat 20's & 1's were always automatic for me. That's how I learned in 2nd edition and just thought it was an unwritten rule since I started playing (1989).
The only time I would ever consider having a player roll for an impossible task is to determine how badly they failed. Even in that scenario, I feel like there are plenty of other factors that could tell me how bad they failed without having them make that roll and risk the circumstance where a player is pissed that they rolled a natural 20 and it was a compete waste. For example, lets say the Rogue insists they wants to climb up a completely sheer surface 100ft high with no pitons or any other climbing gear despite being told the feat seems impossible. If the Rogue has expertise in Acrobatics or Athletics, they try to scramble up the wall, but only get so far with their momentum before gravity pulls them back down and they land squarely on their feet. If they have proficiency in either of those skills, same thing, but they can't quite land squarely and have to tuck into a roll to avoid hurting themselves. If they have no proficiency in either skill, they fall prone and take a token 1HP of damage. I DEFINITELY never want to make a player roll for trivial tasks, because I love rewarding Skill Proficiency and Expertise. I personally find it incredibly satisfying to completely bypass a skill check because of my choices during character creation and while leveling up, and I like to reward my players in that way. Let's give the Rogue some redemption. They want to pick a lock. and I determine that the owners of this place are either too cheap or too incompetent to fit good locks to the doors. The lock LOOKS like a standard, quality-made lock to the casual observer, but anyone with at least Proficiency in Thieves' Tools that pokes around inside will inevitably find a fatal security flaw. If the Rogue only has proficiency in Thieves' Tools, it takes a moment, but before long they found the flaw and cracked open the door. If the Rogue has expertise, then it takes mere seconds for the Rogue to effortlessly pop the lock open.
Natural 1 is always a crit fail and Natural 20 is always a crit success was how I was taught to play. I see why it can be a problem sometimes (more with the successes than failures), but it’s still my preference for a way to play.
I'm against that "nat20 always succeed" because it would for sure create a drama at he table (in a bad way). That's a rule I would change and make clear during session zero that nat20 is "best possible outcome" and not "auto success". Using the example of a player demanding to be adopted by the king: I would definetly ask for a roll here. Not to see if the player succeeds but to check if the king will have the group arrested on the spot or just find it funny and send the group away.
Personally I really like degrees of success and failure. Like a low roll won’t mean you don’t pick the lock, but it takes you a while or is loud and someone may come to investigate. Sometimes of course a fail is a fail but if it it something a player has built into I’m more liable to go degree of success then out right failure.
Strong disagreement. Sometimes, there is a reason to roll even if you can't succeed. Why? Because instead of rolling to succeed, you are rolling to determine the level of failure. Failing with a 17 and failing with a 2 are VERY different results. There are abilities that say, "If you fail by 5 or more" regarding a saving throw, for example maybe you cant succeed, but maybe you can come within 5 to avoid the additional effect
For me I prefer how close the player can get to the DC can give a partial success, and if they equal or pass the DC it fully succeeds. For instance dc is 30 to climb the extremely slippery wall. Pc only has a 9 so can never fully succeed, but what I would do is have it partially succeed if they get a 20 or higher. Then I would narrate they got partway up, and have them roll again.
I am strongly against the automatic 1 and 20 success and fail. Among many problems, it devalues elements such as Bardic Inspiration and guidance that might allow a character to hit a DC 25 even if 20+their bonus does not - or could rescue a natural 1 to a passing grade through a great inspiration roll. BUT. Even though I don't hold by a nat 20 being an auto success, it STILL should almost always be a success since a DC higher than 20 should be very rare. Too often I see DMs using absurdly high DCs (perhaps it's an arms race against optimised players?). I believe a player should have no proficiency and a 10 in an ability score and still have a roughly even chance of success on a lot of skill checks using that. Let people be successful even when not in their speciality.
As much as I do not care as the player or what other people do (it truly is not that big of a deal), I use rolls at my table with the idea of how hard they fail or how big they succeed even if the roll would be a guaranteed success. Personally I am not too big of a fan of people doing stuff they know would be out of this world, like fucking a dragon or murdering the queen in front of the king and staying all buddy buddy with them, so the succeed part would be out of the question for both of them. The roll at that point would just be how hard the retaliation would be from the other side. Rolling high at that point would just mean that the king might give them a minute to explain instead of instantly throwing them into the dungeons, as them instantly being set free due to a nat 20 feels weird to me. On the other side, having a nat 1 always being a fail is pretty jarring too. A barbarian with 20 STR should EASILY be able to kick down a door, the same for your average Rouge being able to open the average door. Sure, most of the time you can (and in my opinion should) say they can just do the thing without a roll, but I have seen some really toxic moments where someone has some absurd + to the skill they are using and roll a 1 and end up hurting themselves or breaking the thing forever. Not fun either, unless the players specifically ask for that kind of thing. All by all, it truly does not matter in the slightest. What matters is that your table communicates about their wants and needs to succeed in having fun at the table!
My general rule is only role of both success and failure is both possible however I do have an exception any knowledge checks and this includes investigation and perception by the way I’ll ask my players to role for those even if I have no information to give my reason being that way if they role low on the ones that matter they won’t be 100% they missed anything
I can't BELIVE you destroyed my long-held opinion about nat 20s and 1s being automatic successes and fails with a single sentence. I was genuinely surprised when you said you treat them that way. But I had to literally pause and sit back and reflect when you said; "If a player doesn't pass on a 20 or fail on a 1, the player shouldn't roll at all." I am not confident that I am 100% sold because I don't always want players to KNOW when something is possible or not. and if your standing rule is "20 passed, 1 fails" you can't have sneaky moments like that without potentially going back on the rules. But I may start implementing this in a more unofficial way. Not coming out and saying it's a rule, but thinking to myself "Will a 20 succeed/1 fail? if not, what is the value of rolling here?" Or maybe I will make it a rule. Will have to reflect further. I've also commented on the dice being an outside factor rather than character action to a fellow DM before who is bothered that the very swingy value of a d20 can make heroes look like fools. I had offered it as an alternative way of timing, and I liked it and thought "Maybe I should start thinking of it that way too". Then promptly forgot about it. So thank you for bringing it back to my mind because I really do want to start thinking about it that way lol. On the note of your other statement about not locking the story behind success/failure; I am struggling with this with the modual I am running. The system I use determines the degree of success not by the total roll but by the value of the special die specifically. So I'm having problems where the story is locking a lot of stuff behind getting a 4+ on the Dragon Dice(off colour d6 of a 3d6 roll), and my tracker (for example) super succeeded because of her skill and powers..but the actual die was just a 3. I'm thinking of changing how that rule works because honestly it's the one thing I don't like and feels wrong and this video helped me feel justified in that.
Part of this is why I don't like the Intelligence Ability Score in name. Not all characters want to have "high intelligence" which implies to us playing that they don't know as many things or specific knowledge.
This whole debate assumes DCs are just success or fail. There are always degrees of success and failure. There is spectacular success, success as much as you'd expect, minimum success with downsides, failure but recovery and no cost, failure, critical failure with additional consequences. Sometimes, even if they're super proficient in a skill but roll a 1, they still fail but they catch themselves and suffer no substantial consequences - but most importantly they still don't succeed. This is "a 1 is always a failure" without making 1/20 rolls a campaign-derailing consequence. Failures can be safe, or baaaad. That's also why we have different DCs and where circumstances or player actions can change the DC. they also change all those scales of minor success or fail
![Quando Rondo - Life Goes On [Official Music Video]](http://i.ytimg.com/vi/eIf3fhMXAow/mqdefault.jpg)
Where do you land on the “natural 20s are always successes/natural 1s are always failures” debate?
Thanks so much to OnlyCrits for sponsoring this video! Visit www.onlycrits.com/supergeekmike and use the promo code SUPERGEEK at checkout!
www.onlycrits.com/supergeekmike
If success or failure is impossible, why are you making them roll (unless you're solely rolling to determine the degree of success/failue) - At the same time... Success and failure can mean different things in different contexts. In your 'I ask the king to adopt me' example, that's always going to _fail_ but I might ask for a roll to see how badly this goes for you, where a success against whatever DC I set in my head leading to the King finding it amusing rather than leading to a prison break episode next session (Though I'd probably have something in between those outcomes.)
@@Stephen-Fox Good rule of thumb, and it's a nuance that 5e doesn't have baked into it.
I don't like it because it does imply that there's always a chance of success, and I think it leads to players declaring rolls instead of actions. But I'm a weirdo that doesn't even like it as a rule for attack rolls and saving throws. So what do I know?
I think the lowest roll always fails the highest roll always succeeds would be better if dnd wasn't 1d20 based. It is just too probable to roll crits with a 1d20, making them way less special in my eyes.
@@user-fe2ne7qx9z How about a 38 sided dice?
Sometimes, a natural 20 isn't success but simply the best possible result. In a case where the stated goal is impossible, such as the classic Persuasion to get the king to give up his throne, a natural 20 or other high roll may have him laugh it off or like your ambition whereas a low roll may have him interpret it as a threat or otherwise take offense.
This has been my stance as well. There's Also the degree of failure that can be important as if a task is impossible but a player wants to do it then how well do they fail becomes important especially in dangerous attempts.
Conversely, I like using DC 1 checks for tasks where there is no chance of failure. A 1 means they succeed, but almost fumble and a 20 means they totally nail it with some extra flair on top.
I will sometimes tell my players "roll to see how cool you look" when they're doing something that shouldn't be able to fail at. Like if my rogue wants to do some parkour and I've already established that these rooftops are pretty closely spaced, I'll still let him roll, because he WANTS to roll. He took that acrobatics proficiency for a reason and he wants to use it, even just to show off. That way I make it clear to the player that I'm respecting his character build, I'm not going to let him fail at something his character should be an expert at, but at the same time the roll isn't meaningless. If he rolls high I describe him looking like a gymnast with flowery language, and if he gets a 1 I can describe how he mistimes his jump and hits the next roof harder than he meant to, causing him to stumble before he catches himself. He still 100% makes it to the next roof, but it looks way less cool.
Another version I'll do sometimes is "roll to see how long it takes." This can actually work for something the players shouldn't be able to fail, and something they shouldn't be able to succeed. If they're doing research in a library, a high roll means they find the answer in ten minutes, or within ten minutes they become certain this library doesn't have the answer at all. A low roll means they spend hours there before finally finding the right book, or they waste hours searching before realizing they can't find the answer here. That way I can give them the answer I want to give either way and thus progress the story, but I'm still rewarding someone who put a proficiency in investigation by letting them use that skill. Because we all know how frustrating it is to have a skill and never get to use it.
What's funny is that the "roll to see how long it takes" is the opposite of taking 10 or taking 20. If you roll a 20, it goes quickly, but if you take 20, it takes a long time.
It seems so simple, but I actually never thought about using the dice roll as outside forces as a base. I had a player with a +15 Athletics try to climb a destroyed house, and he rolled a 1, I just narrated him not being able to find firm ground, and he felt really bad. Looking back I think that if I just narrated him trying to climb, but the wreckage just kept falling and he manage to dodge out of the way or break it before it hit him would feel like a descent outcome. Even in failure his character would still manage to do something cool.
Yeah, I'm definitely going to try and do this from now on.
Love this
This also works well as a way to justify not allowing skill dogpiling. I first hit on the concept in my own games when I had to articulate why a player couldn't make multiple attempts at picking a lock: "Because you failed that first roll, this is an uncommonly hard lock."
There have definitely been times I've asked for a roll not because an outcome was uncertain, but because not asking for it gave the players information that I thought would hurt the scene. The most obvious would be a character sneaking around and scouting out unknown areas that end up being uninhabited; I know there's no one around to hear them, but I still say, "Roll stealth" because if I don't, I'm functionally saying, "The coast is clear" and removing most of the tension. (And, by extension, it means every time I do ask for stealth, they know there's something around that could detect them, even if they haven't detected it yet.)
Exactly. A lot of people making this discussion tend to miss the harmful effect of this rule on the essential restricted flow of information. You gave a great common example for this, and I'll add to this revealing difficulty levels and ACs, when you say to your players: "well, may I have the paladin and the wizard roll stealth, but not the monk and the ranger, you'll be OK.". Same goes for looking for traps or taking a watch at night.
A similar problematic case, much rarer but very crucial, is when an important thing is hidden or disguised. If the BBEG is trying to escape while leaving a decoy behind for the party to mess with, you don't want to reveal it right away with "no need to roll, he is dead."; Similarly, if he is spying on the king looking like a simple maid - you won't want the barbarian to accidentally stumble on it while trying to push through her.
Unless you do want to; And this is the point, the DM should have the absolute power regarding what information to reveal, when and in what rate and timing, and not be forced to it by the rules. Only by the actions of the players.
It's worth considering that, even if the DM knows that what the player is trying to accomplish is impossible (or impossible for them to fail at), it might still be worth calling for a roll if there's a range of possible outcomes beyond simple success or failure.
Take your entertainer bard, for example: she doesn't need to roll to get free room and board, but if she wants to play for tips on top of that, then the performance roll would determine how much she gets that night. After all, that's going to be affected by factors outside the character's control, like how rich the crowd is, and what kind of mood they were in before she started playing.
Or, to use the "players demand the king hand over all his riches" example: he's never going to agree to something that ridiculous, but a persuasion roll could determine whether the king is in a good enough mood that day to be amused/impressed by the party's audacity, or if he has them thrown out on their ears for their insolence.
I'm fairly certain something like that has happened at least once on Critical Role. In a later C1 episode, two PCs compete against each other to try and seduce the same NPC, and after some roleplay Matt has them both roll persuasion checks. Now although Matt doesn't say this explicitly, I'm fairly certain that there was no roll that would have resulted in the NPC saying yes, if only because I don't think that's behavior Matt wants to encourage in his players. But the PC who rolled extremely high got a "I'm flattered but no" in response, and the NPC remained friendly towards him afterwards, going to so far as to later introduce him to a friend of hers who was more open to the idea of a one night stand. Meanwhile, the PC who rolled low was essentially told to eff off, and got the cold shoulder for the rest of the night.
I’ve also cut down on the amount of rolls I call for. A lot of the time I use the degrees of success or failure and use their roll to determine how long something takes. Roll a 1:”It takes you a few hours, but you find that cave you were looking for.”
I'm currently playing a rogue with an entertainer background, and what my DM and I decided on is that the performance roll is not for how well she performs, but how well the audiences receives the message she's trying to convey with her performance. This has worked out really well and has culminated in her using her music to start a gladiator revolution.
I was thinking a similar thing. For the bard example, let them roll. That's how much gold they receive.
Nat1: "The audience is appreciative, but not particularly wealthy. You get a handful of copper and silver, that comes out to about 1 gold."
Nat20: Roll again. Multiply the results together. So you roll a 20 and a 12. "You see a number of very finely dressed patrons in the audience, who are tapping their feet appreciatively. At the end of the performance, they come forward and drop several coins into your hat. You find a total of 240 gold after everyone drifts away."
I am definitely going to rethink when to call for rolls going forward; I hate locking essentially story stuff behind a roll, and I want to reward my players for their choices. I *really* liked your player's argument about Performance checks; maybe if the crowd was hostile or very different culture should that sort of thing require a roll.
As long as you don’t write yourself into a dead end it is perfectly fine to put things behind roles. and put out consequences for failing them. Your heroes have to go through a certain amount of personal risk and the dice simulate that. If you want to reward good choices you should do it in the dice roll. Advantage and disadvantage are one thing. But you’re the DM. If you only think they have a slight advantage you can just give them a plus to aura +3 it doesn’t have to be +5. Which is effectively what rolling the d20 two times gives you is essentially statistically a +5
I've always seen natural 20s as not necessarily "success," but just the best possible outcome. Likewise, natural 1s would be the worst. This comes into play mostly when it comes to contested rolls or interactions where there can be varying degrees of success. Then there is combat in which 20s always hit, and 1s always miss.
Yeah agree. In the 'roll a nat 20 on the persuasion check to ask the king to give up his throne' example; the Nat 20 doesn' t mean the king agrees and gives the players what they want, the Nat 20 just means that the king isn't totally furious and executes them on the spot !
One thing I've thought about a lot on the issue of whether a DM should or shouldn't have a player roll for something that a Natural 20 can't accomplish is the possibility of having them roll to demonstrate that a thing is possible to accomplish even if a character can't right now. If the party artificer or wizard wants to understand the magic bound to an artefact, for instance, but they don't the requisite skill in Arcana yet, just shutting them down might tell them that this is an unsolvable problem when it's actually a problem that you want them to solve in the future. Asking them to roll and then telling them that there are elements of magical theory implicit to the item's design that are beyond them implies that those elements can be solved, if not at their current level of skill. Thus, they have the motivation to find out more about the artefact or the magic and come back to solve it several levels later when they have a +16 bonus (complete with the dip into rogue for Expertise, Flash of Genius, Enhance Ability for advantage and Bardic Inspiration just in case).
love this, just had this discussion last week with a friend about a sailor climbing a mast while not in combat and they had lots of time. they cant fail that, and if they did it would be no fun so theres no roll needed. unless you want to have the roll be degrees of success like a 1 is it takes them 2 tries and someone helping them. is that fun? go for it is it not? then dont. they shouldn't be able to fail that action they want to do if theres no risks or story involved.
I call for extra rolls pretty often cuz my table is in the "rolling dice is more fun than not" camp, but I take my cues from Dungeon World and usually include degrees of success, and sometimes cost trade-offs - eg if they roll a 14 on a DC15 Arcana I'll give them the same info about a magic object, but remove a key detail... Or show hints of a key detail they might be able to uncover if they would like to clutch it in two hands and line up with some finger-holes (beating a 15 would've explained the finger holes. Not beating it means they don't know and will have to take a leap of faith).
I'm also fond of making low rolls in tense situations "success at a meaningful time cost," and the good old "you COULD easily do that, normally, but this place is not normal. Roll to make sure you DON'T roll a 1." Basically leaving a 5% chance for shenanigans and muckery.
And yeah I don't like when players say "can I roll X on it?" I will ALWAYS turn that around and ask "what information are you trying to discover/what outcome are you trying to set up for?"
I think your tangent on failure being the stopping point of the adventure gets to the heart of this problem. Once rolling has a more nuanced outlook, this issue dissolves pretty quickly. Taking your example of the Entertainer background, I would still have the player roll for their performance just to see what the audience reaction is. It wouldn't take anything away (the character has invested in that Background and we're not going to punish that) but I would provide upsides if the character rolls well (say, that room being one level above their current lifestyle as compensation for drawing a large crowd). Just because there's no room for failure doesn't mean there isn't room for a roll; just because there's set conditions for a pass and fail mechanically doesn't mean those conditions have to be good and bad!
The other things I'd say is that success or failure should be a mixture of both character and outside factors. Having technical proficiency is only one measurement of success. A performer can play the most exquisitely composed piece, but if it's to a bunch of commoners who can't appreciate those details then it may fall flat. Both knowing the audience you're performing for and choosing what would be best to perform for that audience and choices that the character is at fault for _and_ are effected by the outside world. A great way to find that balance, I've found, is to let the player tell you why the check didn't go over as planned. If you let them decide the narration for the outcome they'll often be much more receptive to letting their character fall flat or find a middle ground between character and outside factors.
I just realized that I've technically always done the "nat 20 always success, nat 1 always fail" because if the DC is so high they cannot succeed, I'll let them know that what they're tying to do seems impossible. And if the DC is so low they cannot fail, I just describe them succeeding. That way my players know that when they're rolling that it will actually mean something about what happens.
Only if a player still wants to try something that I said is impossible, which in that case it will be a "how much to they get hurt" check or a check to see if they can make the challenge a little easier for the next person depending on context.
Yeah, it seems surprising to me that there are tables out there were that isnt the case were DMs allow their players to rol even if success or failure is imposible
my thing is, sometimes the extra bonuses get you to the DC 30 (labeled Impossible in the DM's Guide) before you hit 20 and sometimes only if you hit 20, but how would you know if they didn't roll vs...yeah, I'm not gonna make you roll because there is no chance in hell you'll persuade the guy who doesn't speak Elvish whatever you're persuading. Idc how high you roll, it's not happening
There are a few circumstances when the auto-succeed/auto-fail rule causes problems: opposed rolls, graduated rolls, and variable bonuses. Opposed rolls I think is pretty clear why it could be a problem. Graduated rolls, or rolls with different grades of success, invalidate the idea of a success/failure binary. And with variable bonus, like those from Guidance and Bardic Inspiration, there's a chance for results significantly higher than a DC which makes a task normally impossible for a character, but it's only a chance. If someone can't succeed without Bardic Inspiration, do you refuse the roll? If they get a 20 but don't get the DC, do you give it to them? I didn't like the rule. I do encourage the use of the passive score or auto-average result in unstressful situations, though. Like the entertainer bard, if a task is something you have enough experience with you can reasonably allow the action to take place with the assumption of a 10 on the roll if there's nothing happening that would disrupt them.
I really like Fabula Ultima's approach where you can invoke your "Traits and Bonds" (who you are and the strong relationships you have) in order to spend a limited resource to ease related check (this guy killed my parents so I can punch him in the face even better) and to get easier access to equipment (a herbalist is more likely to have a specific flower on them than a princess who just ran away from home). I have no idea how to put something similar into DnD
On the "Entertainer performance check" - I like to instead make the check relate to "extras" they might be able to get.
After considering they usually have really good +'s to performance, on something like a 16+, or 20+, the owner of the tavern themselves is so entertained maybe they shout the whole tavern a round of drinks and make it an event (that the rest of the patrons would then be extra friendly towards the performer), or even giving their best room(s) to the party.
It doesn't necessarily have to have a gameplay impact, sometimes those extras are enough for people to have a more memorable moment. But it otherwise can still have more impact: better rapport with the owner which may help down the line, or inspiration or some other bonus for spending the night in relative luxury to what they were expecting.
Also doesn't have to be every time if you need to move the game along slightly faster!
I believe in degrees of failure and degrees of success.
If someone wants to persuade someone else of something that will always fail, they can still roll to see if the person takes offense or politely declines.
If you try to do something that you cannot fail, you might do it in such an amazing way that the group gains some additional benefit. You might be so good at tracking that you would never be able to lose the tracks you are following, but if you roll well enough you might gain additional insight about the creature you are tracking.
My DM does a little bit of a mix of both re: whether success or failure is related to the character or external effects, depending on the circumstances and even the individual character. Like, our warforged DMPC is extremely awkward and physically and mentally out of place because of her backstory, so rolling low on stealth checks is almost always a mistake on her part, whereas in combat the DM tends towards low rolls and especially nat1s being the opponent blocking the attack in a cool way. One time rolling a nature check to identify a creature an NPC was describing, my character hadn't gotten any sleep because she was on first watch when we got attacked and we'd been walking all night since then, so it was easy to say with a low roll that my character was just too exhausted and burnt out to think straight. My favorite though, is if we're rolling for something that would come completely naturally to us, like artisanal skills or my monk-cleric meditating to commune with her god, the roll is less about success and more about how long it takes; a low roll could mean a few hours of the monk trying to get settled in an uncomfortable environment vs. immediately hitting the sweet spot to focus, or our artificer jeweler helping to make a magic item could reduce or extend the time needed to complete it. And sometimes it is just a moment of clumsiness because hey, it happens to the best of us.
I am firmly in the camp of:
A) don't make a roll the GM doesn't call for
B) 1s and 20s are not automatically failures or successes, respectively
Keeping in mind the optimal rule for Ability Skill checks, I may have you choose a different ability score than the default for a skill check, based on varying circumstances/factors.
I am also a fan of occasionally limiting skill checks to those who are at least proficient in a skill; both to reward player choices and to represent the likelihood of access to that knowledge.
Definitely agree with you that it's important to decide whether a roll is even called for. One of the things I miss from 3.X (don't remember if it was in 4E), was "Taking 10". Basically, your character would spend 10 minutes on a task, and instead of rolling, the check was treated as if you rolled a 10 on the d20.
Great video, as always.
How did I forget about 'taking 10' and 'taking 20'? I wasn't always a big fan of the latter, but I definitely miss the former.
I've definitely been guilty of calling for rolls when they weren't needed in the past and it's something i have become more conscious of recently. Because they do just slow the game down to a crawl and doesn't allow players to actually feel badass and heroic, especially once a campaign starts brushing the higher tiers of play. The comments about not locking plot relevant info behind a single roll have also given me reason to rethink some of the ways I structure my adventures. It hasn't always been the case but there have definitely been times when there's important info the party really needs in order to move forward in the current story arc but then no one rolls high enough and I'm left a bit unsure how to proceed. I want to honour the rolls because I generally really don't like fudging things when the parameters have already been decided but without this clue the adventure is left foundering a bit. This was already happening much less as I've grown in experience but this video was very helpful for putting some stuff into perspective. So thanks!
I really appreciate that all your videos are released with subtitles! I have hearing difficulties and its very disappointing when some of my favorite creators wait a few days to subtitle or dont do it at all
Thank you! ☺️
I generally agree with all your points and wouldn’t argue any of them. I do, however, want to make a case for the “dramatic impossible roll”. Generally it’s good to let your players know you’re doing it, though as always, depends on the group. There is something I find compelling as both a player and a GM about the prospect of the characters going up against insurmountable odds with no chance of success for dramatic purposes. Even if that nat 20 is rolled, at best you get a dramatic moment. I think this sort of thing works best with like, a first confrontation with a big villain or something like that. And this shouldn’t be done all the time, but I find sparingly, they can make great moments if everyone is on board
Started writing the comment, thinking it would be short... Well, sorry, but I like the topic, and I had some thoughts.
I don't really see a problem with, occassionally, undermining the characters. No matter how eloquent, how confident, how skillfull a speaker you are, sometimes you'll get something caught in your throat, and if done right, it can be hilarious.
Another thing I don't necessarily agree on, is the idea that if you cannot succeed or cannot fail, you should never ask for a roll. True, these cases may not call for a roll. Maybe you just give it to them, or you just say: "Not possible for you".
But, sometimes, success or failure is more nuanced than simply a DC telling you "Yes you have succeeded" or "No, you failed". Yes, there is a point, where an action would count as a success or a failure. But I find that too black and white.
If we take your example of persuading the king to adopt a player and make them his heir. There is no way this should succeed (unless maybe certain exceptional circumstances are present in the story).
A nat 1 could signify that the king is highly offended by the very notion that such a feeble underling could even utter such words, even as a joke, and he would have you hanged the next morning for your insolence.
A nat 20 could mean that the king and his court starts laughing, and he flips you a gold for giving him the best laugh he's had in a while.
Neither of the 2 scenario's are a success, but a roll could still be used to determine certain factors.
If we could take this kind of approach to searching for traps you could say:
1-4: You are confident that there no traps.
5-9: You find no traps, but you feel you've missed something.
10-14: You find a pressure plate on the floor, disguised as a tile, which sets off arrows that shoot down the corridor you're in. It took 30 seconds to find
15-20: Same as 10-14, but you spot it almost immediately, and you may get some extra details about the workings of the trap.
In this case, there's still a hard cut off with a DC10, which let's you succeed in finding the trap, but the dice roll matters more than just yes/no.
It doesn't have to be this strict, you could just set a DC and then do extra based on your instinct at the moment, the further away they are from the DC on either side.
But also, if you use this kind of system, you can still implement dice rolls sometimes, where the player won't be (mathematically) able to either succeed or fail. By mathematically, I mean modifier wise. If they have a +11 to a skill, and the DC is 10, they cannot fail. If the DC is 20 and they have -3, they cannot succeed.
If they succeed big time, you can make them feel awesome, if they fail big time, you can make it funny. It also lifts the weight just a bit off of the DM's shoulders, so they don't need to constantly be aware of all the modifiers of all the players.
I will say though: The story you told about the bard is a good situation of where you might not want to ask for a roll. They should be consistently good at playing in a tavern and getting something out of it. On the other hand, if you ask for a roll, and they roll well, you could describe how even the drunkest, almost passed out patrons, perk up at the sound of their music and start looking energetic, singing along and cheering. And, occassionally, on a nat 1, they might just break a string on their instrument (or might not). Not a mood killer, as they are a good performer, so they can play it off, and work with the crowd, while they change the string. But maybe that happens, and maybe the passed out drunks stay passed out, while the rest of the patrons still have a good time.
I'm still using 10 min = 10 + mod and 20 min = auto success or 20 + modifier rules from back in the day. It just makes sense that someone with proper time and training can just do the thing. I also took it to mean that the regular rolls usually mean you're in a high pressure situation with little to no time, so it makes sense that failure is more likely to happen.
That makes sense!
Great video. Personally, I don't like knowledge being used as a skill to begin with.
In my own RPG I'm designing, Knowledge is a stat you get from adding together a certain 3 attributes. Every 5 points of Knowledge, you gain a research point. Research points can be spent to learn a new language, or to give yourself a +5 Knowledge bonus to a subject of your choice. So, rather than "roll to see if you know something", the GM just checks your Knowledge stat.
This all came from a moment in D&D when I asked a player to make a history check, they rolled a 1 (total of 2) and I had to tell them that despite a low DC, they did NOT know something everyone from their culture would know.
It was funny as it happened, but I couldn't stop thinking about how I should have just told them the significance of that sword...
In a trad game such as D&D, CoC, etc? When failure would be dramatically interesting - A great piece of advice I've seen - I think during one of Johnny Chiodini's Dicebreaker video where they talk Luke Westaway through how to DM while Luke was gearing up to DM for his first time - is never ask for a roll you can't afford the players to fail.
In a PbtA or game with a similar philosophy? When a move asks for one, and no other time. (Which also strongly discourages describing failure as incompetence from your players via the "Be a fan of the Player Characters" GM principle which the vast majority of PbtA games have)
The big insight I’ve personally taken is failure shouldn’t mean “the task fails” it should mean “something goes wrong”. There’s a twist, you lose resources, you waste valuable time, monsters wander in, an opportunity closes, etc. I don’t call for a roll if nothing can go wrong, they just do it. I don’t call a roll if nothing can go right, they face the consequences.
One of my favorite rolls I ever called for was "Make a saving throw with whatever attribute you want." There was never any chance of failure, but the player's decision still gave them agency in a scene that was almost entirely flavor.
My erly GMs had the same answer which I found a very easy answer, for the question when do you roll?
You roll when both answers are in world realistic and narratively interesting?
My DM had us each roll to identify an NPC race. We got a nat 20 and 3 other rolls over 20. Then told us we all don't know and told us the DC was 30. None of us could have rolled that no matter what. I was a little frustrated because why ask us to roll only to 100% fail. So I felt like just telling us, "None of you recognize him" would have been more effective and kept the mystery.
I remember in one of the first episodes of Oxventrue Blades in the Dark the GM was about to call for a roll and stopped himself saying something like 'never ask for a roll when failure would end the adventure' and that stuck with me. I'm not a GM very often but it makes sense.
I remember implementing a house rule in my game to try to make me stop calling for inconsequential rolls: any time I ask a player for a roll, they can choose to fail, and get something like a stacking Inspiration to use later. It did a good job in that nobody ever took me up on it, which means I never called for a roll that any player cared to fail at.
I use graded DCs, so I have no issue having players roll for something impossible: "failing" with a 23 is different from failing with a 6. (I know d20 systems aren't optimal for graded DCs but it's good enough for home games)
I really like the concept of degrees of success/failure. This allows the player to still make a roll even if a critical won't get them the exact result. Going back to "persuading the king" example, if a PC tries to get the king to give up the thrown, or whatever, a critical success could be that he takes the whole thing as a joke and doesn't imprison the party.
My biggest annoyance is when I roll a 1 but have a +14 in a particular skill (expertise and other fun stuff). Treating that 1 as a full on 1 despite meeting a 15dc is annoying as heck. I put a lot of work into a set skill and it feels invalidated when 1 is turned into a total failure regardless despite my skill.
But as for calling a check I tend to word things this way
“May I roll a perception check to listen to rumors while I sit in the tavern/ travel the town?”
Or
“May I roll a nature check to understand what unique nature this jelly has?”
Note that I said “May I” to start each. I clarify if asked why and I allow the dm to say no with no hard feelings. I honestly rarely argue with a dm. The only time I did was when one tried to two weapon fight with an enemy and was using rules of npcs having similar rules to players. I am very familiar with these rules. The dm tried to have the npc use a long sword and short sword. I said the rule only allows one to two weapon fight if both weapons are light. I then read off the part of phb when dm asked me to. Like I said, I have used this rule before and I understand a dm has a lot to track so I don’t mind sometimes clarifying these rules when I remember them. I still concede to a dm’s ruling so long as they keep it consistent and it doesn’t undermine the class or background features of a player.
When I have a structured adventure I try to have multiple threads that all converge on the same point. I'm comfortable enough as a DM to also change things around behind the scenes, repurpose a motivation, leave a clue somewhere else and I let the dice decide that. The party missed a thread or a clue earlier. I didn't *intend* this random stable hand to be as key as they are, but the players have ignored another thread entirely or missed a clue earlier on. So I implement it there if the dice roll(s) are successful.
In a way I almost look at as a Mass Effect dialogue tree, which may sound cheap if you know how that works, but in essence they can fail and succeed forward the only difference is the route but the destination is always the same. It does require a lot of both planning and improv on my part, but it matters to me so I don't mind.
Overall I think that many DM's are caught up on this "debate" because they don't know what "rolling the dice" represents. If success or failure is guaranteed, don't roll for it, if it doesn't make narrative sense, don't roll for it, it's that simple. The DM is the one that calls for rolls if he deems it necessary (agreeing with what you said in "when the players declare their own rolls" section).
If a DM deems inappropriate a player's request, he can just deny it - that alone disqualifies any argument made in the spirit of "what if the rogue tries to convince the queen to hand over her kingdom?" - if there's no roll, there's no risk the rogue could actually succeed in this. Likewise, a DM could just deem unecessary to roll for the barbarian to break down a rotten door.
Rolls, as you stated, are there to represent chance and circumnstance, things that are out of the control of the players and, when rolled, dictates what happens (even the DM is henceforth obliged to follow what the dice says).
That being said, I respect the opinion of those who disagree with me.
I agree with a lot of this, except the bit about failure being coming from outside the players. I think it really depends. For hits and misses at least, I often describe the narration around the score itself - low rolls are a failure of the player, then the next zone is the "natural" defense of the target (i.e. its natural armor or Dexterity to dodge) and then finally any physical armor above that. It's more of a guideline than a hard rule but I find it helps a lot for combat, at least.
I watched a notetaking video recently that got me thinking about lore checks in a way that I'd like to try. The idea being that the player's notes are also the character's in-game notes. Building on that you can hold that if the player is reading through their notes to find something then the character is. In most situations that is not going to cause a problem. You can sit down and sort through your notes about history related to this 'whatever' before you. But sometimes it would be bad to. The battle is going to be over if you stop to read if you've heard anything about creatures like the one you're facing and critical negotiations are going to lose a lot of steam if you ask the king to stop talking for a bit while you read their family tree. In those moments you roll and, depending on the results of your roll, you get a little time to consult your notes which is counted as you just remembering in the moment instead of reading your notes. It gives reasons to research an area in advance and engage in lore as more than flavour. It also encourages you to share notes in downtime moments and covers those "You weren't there. You don't know that" moments.
As for all the info they haven't got notes on (Which should be most things. You don't want them going crazy with notes or else they'll never find what they want when they need it and it would just not be fun) anything they would know they know, anything they wouldn't they don't, and anything in between they may get a hazy detail, popular misconception, or rumour on. They have proficiency in nature and they're a sailor, well then, they know all the common sea creatures and plants, have heard of most of the uncommon and rare ones to various degrees, and the anomalies are just as anomalous to them as to non-sea goers. No roll needed.
Normal lore checks are so passive. You don't know it or you do and here's a lore dump about it. This way you have the fun of "You passed the history check, but only just, so as the countess stares at you you've got 15 seconds to look through your notes then you must close them and time resumes". *Cue frantic page turning* (Exact times would depend to the complexity of information being sought but you probably don't want to pause for more than a minute even on a high roll)
Not for every group, especially not something to spring on a player with dyslexia, but fun for some groups. Also, with the notes counted as in world, a villain may get hold of them to find out all your group knows and who they rely on.
Sorry. Good manners to link sources when you mention them so the notetaking video: ruclips.net/video/m1kHdQBqzVE/видео.html&lc=UgwS2z8JFBEzJbeDkLF4AaABAg.9pyddy0qG369pz6Bw9j3Z4
I try to call for rolls as little as possible because I don’t want to drag things out. Asking for a roll, them rolling, getting a result, me deciding the outcome etc time after time starts to really add up.
Honestly there's a good portion that ignore the pass with detriments section of the rules as well as the possibility that your not rolling to see if you pass but how well you fail at the task.
Players love to try to do outlandish things because they feel they can accomplish anything if the stats are high.
Some things are impossible but players like trying to do things. There was someone that I used to follow that had a dman good quote for why you'd let someone roll even if they couldn't pass "Because I knew that you’d feel unsatisfied if I didn’t let you roll, and I thought you might find this outcome fun, even if it’s not what you were looking for,”
My DM has started to ask us what causes us to fail whenever we get 1s on ability checks.
I remember my group’s last session the dice were fucking with me the whole game. Especially with skills i was proficient in. +5 performance as a bard and I rolled a 1. And when my DM asked me what causes me to fail I just say I’m not confident enough in my performance to make what I was trying to do (act like a monster was coming out of the alley after my character to scare off a mob) be convincing.
One thing I think often gets left out of this discussion is group rolls. Eg the rogue is definitely going to beat the stealth DC personally, but are they going to do so minimally or are they going to bring up the group average significantly, helping the group pass?
One thing to keep in mind is that even if success/failure is impossible you might still want to determine the degree of success or failure.
-The bard always performs well enough to get a free room for the night, but roll anyway to see if you catch the attention of that important NPC.
-There is no way the blacksmith would ever accept that offer, but roll anyway to see if he is insulted by the offer and refuses to do further business with the party.
Gotta disagree with you hard on that fight in Fantasy High episode 2. It was effectively a flourish, and (arguably) two people died because that bad call cascaded into a much tougher fight than Brennan intended. I think that instance is much better used to illustrate the point that even master level DMs like Brennan Lee Mulligan still make bad calls sometimes.
I fall on the other side, where a natural 20 is the best possible result, instead of an automatic success. But that's because I believe that sometimes impossible rolls are still worth making. Failure can take many forms, so sometimes a roll is worth making to see how bad the failure is (or how well someone succeeds). If I think something is impossible (or trivial), I might say no roll is necessary, but I might have them roll and say "what you are trying won't succeed, but we need to see how badly it goes for you". And that's the important part---setting up the expectation ahead of the roll.
That fight in Fantasy High with the table jumping played out how it did in part because it was, I believe, the first time Brennan had ever GM’d a 5e combat. And it involves a very fateful roll much in line with your video that just wouldn’t be the same if it had been shut down because it was too far fetched.
Personally, I love the euphoric rush of the nat 20, and I love how Brennan honors them. (It’s one small critique I have with Matt Mercer. Nothing quite sucks the wind out of the sails of joy like hearing “For a total of…?”) But Brennan also doesn’t just let his players roll willy nilly. There’s a particularly funny stunt in Fantasy High Sophomore Year that involves a PC attempting to use a ribbon dancer to try to slow their fall from a great height. Brennan has them roll at “quadruple disadvantage,” which was rolling 4d20s and taking the lowest. He said “it’s technically possible for you to roll 4 nat 20s.” 😂
The Cypher System’s approach matches pretty closely to how I approached nat 20s in 5e, but I still brought some things over from that system to my 5e games. In Cypher, skills, attacks, basically with any roll, rolling a nat 20 gets you something extra and a nat 1 triggers a “GM intrusion.” It doesn’t necessarily mean an auto success or failure - you can’t roll a 21+ on the d20 so without modifying the difficulty, a nat 20 would still fail that kind of roll - but it does mean that you get a little something for the occasion, a temporary boon of some kind. And GM intrusions on nat 1s *can be* fumbles, but they can just as easily be positive things, or success at a cost opportunities. Both of them just mean “something significant happens that changes the circumstances of the situation.”
When that OneD&D dust up happened, I couldn’t help but think that most arguments people had against the rule could be resolved by just taking with your players? Like, if they want to punch the moon, it’s likely that’s an absurd thing that they and you know won’t work, so asking is either a joke or they’re being intentionally disruptive. If they’re being disruptive…that’s a conversation you need to have with the player about tone and expectations. OR it somehow is something that they have a slim chance at doing, and if they succeed, you get a ridiculous and awesome story to share with each other for a long time.
Anyway, great video, as always, Mike. Sorry for the essay length comment 😅
I actually agree that rolls should be within context. Our bard wanted to perform for lodgings. I asked if he has that entertainer background feat and he doesn't. That's when I asked him to perform. Although he's a bard, his idea of a bard was to be a "dad" who uses his traits and spells through being a "dad" projecting "dad" energy, so performing in a tavern isn't his specialty. It's something he might do well in if tried, so I let him try. He rolled below a 10 and I saw fit to deny him free lodgings and earn him 2 copper coins from tavern patrons.
Our loxodon barbarian wanted to burst through a wall. I wanted to just allow it, but they are level 1, and I want to give him that feeling of getting stronger, outside of just feats and traits. So I had him roll for it. When he gets to level 5 and tries to bust through a wall like that, he will just do it.
But when rping a conversation I would only call for rolls if the situation calls for it. If the npc they're talking to us friendly or even indifferent and isn't hiding anything there is no reason for them to roll if I find their words compelling or reasonable.
But I also think players enjoy rolling for their high stats, and if one of your players opted for high charisma stats I would offer persuasion or deception checks more often for them, rewarding their choices. Same for our outlander monk who opted into survival and wants to use it. I'll let him roll more survival checks than I would normally for others, bc using your strong skills is fun for most people I know. But if your success is guaranteed then that takes away any stakes of failure, so the success is less rewarding.
I usually ask for rolls only when both success and failure present interesting options. I'm also a fan of either failing forward or else introducing complications... If the players fail a hacking roll, they may still get the relevant information, but they trip a security protocol, or inadvertantly pick up a virus in their system. I find this especially useful for research and investigation type rolls, so I can make sure they get the minimum they need for the story to continue, but also present both concequences for failure and rewards for a particularly good roll.
I don't run D&D anymore, so I have no opinion on Nat 1 vs Nat 20 (though in the dim and distant past of AD&D2E, I think I used the auto-fail/auto-success option.)
Making "failing forward" your default can be challenging, especially if your default for a few decades has been "fail - end of line".
@@mkang8782 Think of it like Improv. Whatever will keep the scene moving... do that.
I'm not a dm or gm, but I like the idea of degrees of success or failure. And having a roll you absolutely can't succeed might work as a you try something you just can't do, but you don't fail as hard as you might have. Or if you can't really fail maybe you just do the thing instead of being super awesome about it, like simply hopping a gap and then firing your arrow instead of flipping over the gap and loosing the arrow while upside down or something. I don't know if that makes sense or if I explain my view well but hopefully the idea comes across.
I once had a rogue character with really high deception & persuasion, but during interactions with NPCs when I would be using those skills the DM always made me roll play it out instead and then decide if it succeeded or not.
My character was the one with an 18 charisma score and skill proficiencies not me. I don’t mind some roll play but let me roll the dice.
It was so frustrating I ended up leaving the group.
For context: my players like lore details and we are only playing in a group of three (me, the dm plus two players) so there is enough space for everyone to have some extra rolls that would be too much in a bigger group. History checks are the most prevalent rolls in our game after insight checks, so it’s a very specific kind of game that might not be for everyone.
If a roll would be an automatic success I like to ask them if they want to, like: „You would know the basics, would you like to roll for some extras. In my experience that gives them the satisfaction of having some sort of an automatic success based on their background or resources they put into their specific values without taking away the fun of rolling the dice and getting to know some fun details. Setting up rolls this way also helps with the problem of random details seeming more important then they are, because the players know that they are extras. From time to time I will hide some important things in those answers, just to make them feel like they made a random discovery that actually helps them.
For things that might be automatic losses that wouldn’t break the game if they work I might say something like: that’s not going to work but you can roll to see if you are super lucky (that being a Nat 20). It’s no problem if it doesn’t work, because we aren’t to many people it isn’t a huge time loss and it makes for some really fun moments if something works, especially if we put some creativity into describing what circumstances led to that unlikely success.
My players tend to make specialists so I tend to do tiered success rolls when they ask to do their specialty. If they roll 9 or lower they get the bare minimum info or item they're after. 10 or higher they get better and better stuff at 10,15,20,etc
I am always surprised that this is still something that needs a discussion. Nice to see your take on the subject, I guess this kind of advice will have to be given again and again.
My take:
1. There are no skill checks in 5e.
2. Only the DM can call for a roll, the players can only declare actions.
3. Only call for a roll if the outcome is is unclear AND there is a consequence to the action.
4. Yes, no matter how good the argument was, and how eloquently the player stated it, I require a roll because the character might still not get it across the same way. I will hand out a bonus for good arguments.
Edited for sounding like an arrogant ****, I hope this is better.
great points. For me auto-success/failure doesn't always apply because I tend to to use sliding scales for a lot of ability check DCs (eg a History check might give more info the higher you roll, not just a pass/fail threshold). Also, some checks can tell you something even if you fail on a super high roll, eg you're a scholarly type specialising in historic evil cults, and ask if you recognise the symbol on the cultists who attacked the party. The DM calls for a religion check and you roll a nat 20. The DM says that you have never seen this symbol before. Is that a success or a failure? You didn't learn the information you were hoping for, but you learned something about the symbol - you can be confident that it's something very esoteric.
I was thinking a lot about the Nat 20 convince the king to adopt me situation and I think I would respond with:
"The king is surprised, smiles and then says, 'Yes... You'll do nicely'"
And now begins the plot where the Mad king needed a pawn in his 4d chess to overthrow the neighbouring kingdom.
"Yes and" and "Yes but" that shit with game of thrones level drama
What about tension and rolls without obvious consequences?
What I mean there is dark cave and players decide to sneak in. Now there is nothing to hide from, so they can't fail stealth (it's just an empty cave), but not rolling it, could lead to similar struggle as initiative (rolling means fight to the dead). Also roll like this might not mean anything right now, but it might be a difference, if a local monster will notice, that something was in its layer.
I think the problem with the Nat20 rule mostly came from the fact that it undermined how DCs work AND requires the DM to intimately know all of the skills and modifiers of their players. For example, if something is DC: 25, that might be relatively easy for a character with expertise, but difficult if not impossible for a character without the required proficiency. Players may be able to stack things like inspiration or guidance onto a roll to increase the chance of success against a difficult roll. You don't necessarily know in advance whether it is mathematically possible for the character to hit the DC in advance, so you can't always say for certain whether or not they should be allowed to roll. Likewise with critical failures, sufficiently skilled PCs may succeed regardless of the natural 1, but others may not. It just isn't always practical to gauge this. Yes, in certain scenarios, like the "seduce the barmaid" or "king give up their crown" examples, you may be fine just saying that person is not interested and no roll is required, but there's other scenarios, like say a stealth or investigation check, where it is not nearly so clear, or where even the game mechanics expect you to make a roll, and that there may be situations where, despite the PC's best efforts, success is impossible, or failure is impossible, given the DC.
Someone below mentioned contested rolls as another good example of a potential problem with this system. Suppose you have a wrestling match between a wizard and a barbarian, using athletics, with the wizard having a -2 mod and the barbarian having a +10. Wizard rolls a 20 and barb rolls an 19. On the dice, the barbarian wins 29 to 18, but the nat 20 means the wizard wins anyway? What sense does that make?
I'd honestly rather have it that nat20s don't guarantee a hit on attack rolls than making it an auto-success everywhere else.
The reason why there's no automatic success or failure in my game is because I come from Pathfinder 1, where teaming up changed everything. I enjoy and encourage people to use their time in game, ressources and imagination to pile up to achieve a thing. Guidance, bardic inspiration and your natural bonuses might be just enough to succeed.
I also believe that in the case of a very high modifier, rolling a roll that might never fail, is a fantastic way to have that player live their character's fantasy.
But I always say no to roll that makes no sens. I have therefore never had a situation where a 1 or a 20, not giving failure or success, was problematic. They were usually indicators of things that are out of there reach yet, or beneath them.
Agree on unnecessary Rolling, but the threshold of what is 'impossible' may depend on the DMs knowledge of the characters as well. Does knowledge that the team's bardic inspiration or guidance *could* make the check doable even if it is impossible without influence that? Does knowing that you're not going to persuade that hotheaded noble mean there aren't risks of them taking drastic retaliation if you roll low?
It's a fine line, and YMMV depending on your players.
The performance check isn't necessary for the bard but it does seem weird to not role play that scene. Sure nothing bad can happen but rolling really well might lead to interesting history or rumors being divulged
Thank you for adding “clickity-clack math rocks” to my vocabulary
I was surprised back when that happened during that play test for One D&D, especially since I always thought that pretty much every game already handles it that way. That players cannot simply say they jump to the moon and expect if they roll good enough that this could happen. The systems I grew up with, like the Storyteller systems from White Wolf made that explicitly clear. But I guess that was not so clear in D&D... no wonder I always thought that D&D is badly designed.
I agree with your opinion on rolls.
BUT, there are some situations where the DC you set as a DM was so high, that even a Nat 20 would not succeed. In very rare and note wrothy occasions only tho
For example, on critical role, when Caleb decoded Avantica's code. The DC was straight intelligence and it was a 25. That was to exaggerate how good that code was. If anyone other than Caleb tired to decode it, they would've failed even with a Natural 20.
The same happened in Campaign one When Someone tried to a certain important unlock in a special library towards the latter end of the campaign.
All that to say that, the DM might expect the roll to be made from a certain character, and a nat 20 might not be enough by someone else. In rare and note worthy occasions. So rare in fact, that Critical role only had one of those occasions in each campaign until now
I have nothing to say for nat 1s. Succeeding on a nat 1 or even on a 4, feels wrong, and so nat 1s should, in my opinion always fail
I do not like the automatic success or failure on a 20 or 1 respectively. Especially on unprompted rolls.
However in rolls where a character cannot fail like having a skill bonus of 11 on a DC 10 it is fun to have the character succeed but in a ungraceful way.
Like hopping over a gap they stumble a step or two but recover their footing. They still succeeded but the rolled 1 adds flavor to the check.
On an impossible task such as demanding the king give you his crown and make you king instead. The roll of a natural 20 isn't a success, it makes it so you and your party aren't tossed in the dungeon.
What annoys me the most is when they ask for a roll and EXPECT a success and will bend over backwards to make it that, or vice-versa
Can you make a video about all your DND shirts?
I definitely agree with not asking for a roll where it isn't needed. I am comfortable telling my players not to roll when something is impossible. I'm also comfortable setting a DC like 30, i.e. near-impossible, but if the right (trained) character makes the roll, they could make it on a nat 20 with their +10... But not everyone in the party could.
That said, I don't like the auto success/failure on skill checks mostly for failures. When people invest in their training of a skill and get their modifier up really high, I want to reward that. If the DC was 12 and they ask to roll a 1 but they have a +13, they can succeed that task. They're trained. Sometimes I just say, "you don't need to roll, you succeed," but sometimes they just want to roll dice. 😂 Same with something like a high level rogue or an eloquence bard, where they treat any single digit roll as a 10.
Helpful, thoughtful content
We agree on quite a bit. Skills are competence, the dice are the universe being a dick. Give advantage or disadvantage liberally based on small-b background and other factors. Only roll when someone can think of an interesting consequence of both success and failure. Don't roll for automatic PC features. Etc.
I also think it's worth calling out games like Fate, 2d20, and others that have meta-currency that allow players to spend a resource to ensure results go better (or take a hit to ensure they go worse). Such a game feature eases many of these concerns.
However, you and I seem to vehemently disagree about the actual auto 1/20 rule, for 2 main reasons:
1. I think the potential abuses you rightly object to are inherent to the idea. Seasoned GMs can learn the nuances of "don't bother rolling when ...", but novice GMs tend to default to "sure, roll". The auto-fail/success loophole is just waiting to happen, and teaching GMs how not to fall to that is a much more difficult and lengthy process.
2. I think GMs should be allowed to set numeric difficulties without memorizing all of their PCs' skill capabilities. It may be that a given ill-suited PC can never achieve such a difficulty, even with a nat 20. That should still be a failure. (and vice versa for success).
If there is no possibility for success or failure, I don't bother asking for a roll. I tell people that rolls aren't always connected to whether or not their character 'failed'. A roll in my game is the subtle influences of the gods, the random influence of chaos, the whims or gut feelings that win out. With social encounters, I rarely ask for a roll from a player. I'm more prone to roll if I feel the NPC is on the fence about a decision/feeling that is connected to whatever they're talking about. If the king doesn't know you and you ask to be adopted, I don't see any reason the king would do that so no roll will happen. If you ask the king for a piece of his army to help defeat a monster, he might say yes to show his people that he will do whatever it takes to protect them or he might say no because the monster is still outside the city walls so he doesn't see the reason to waste resources on that. In that case, I might roll depending on what the PCs do/say. Long before that though, I make it very clear that character creation will have perks and consequences. A noble paladin asking the king who worships the same god will have an easier time than a noble paladin that worships another god, but that second paladin would have an easier time than the antisocial fighter that gets into way too many bar fighters. Depending on how you craft your character, there could be instances where the rogue is more trusted than the paladin. In my world, you have rogues that are thieves but you also have rogues that are locksmiths, scouts, and bodyguards (think ninjas). You also have paladins that fall into the religious zealot category. It's up to the players to figure out their characters and the world will react accordingly. If you want to be a religious rogue that feels that their purpose is to protect innocent and inspire people to do good, then that's perfectly doable in my game. Of course, that means that if their actions convey them more as a fanatic murder hobo, well then that's what they really are and the world will act accordingly because the gods do exist and they can tell one of their champions who also has an adventuring party that your fanatic murder hobo is perverting the scripture. There is no roll for that because at that point, several choices were made and the path was followed to that end. Choices matter and seal your fate, not the dice. (With that said, I also encourage players to narrate how their character goofs if they do, not just if they succeed. Had a tabaxi player roll a nat 1 trying to jump over a sewage river in the sewers and the player described the feline doing a tiny hop instead of the epic jump like so many hilarious cat jumping fail videos. I gave inspiration for that.)
I would argue in your example of the entertainer feat that a failed roll would make the influence of the bard to other innkeepers harder to get room and board. If you’re trying to build a living world outside of the PCs, rumours spreading of a talentless bard would make the bard try harder to get their reputation up. Even if the roll is just to convince the next innkeep to let them perform because of less than stellar prior performances.
I love the idea of "the dice represent the reaction of the world", as it means the players don't have to feel bad about their failures. But if we extend that logically, what about their successes? If they roll a crit in combat, that would mean the enemy trips and stumbles onto the player's weapon! Now I'm sure every GM here would have 20s be entirely down to the PCs competency, but I thought it was an interesting comparison.
In fairness, “Appreciate how your competency led to your successes, don’t take your failures as a reflection of you personally” is also just my advice for getting through life without spiraling into anxiety whenever something doesn’t go your way ¯\_(ツ)_/¯
My biggest issue against critical successes and critical failures is the dreaded "super critical failure". I'm talking about being asked to roll sleight of hand to steal keys from a guard, and rolling a 1, and not just failing to pickpocket him. Instead, the GM chortles, and describes me trip forward, pushing the guard over a railing to his death, and thereby summoning two thousand more guards. Who also now have laser guns.
But there is no way to write core rules that dictate that critical failures and successes must be "reasonable". So it is better to avoid that entirely by removing that for ability checks.
I didn't know there was a debate. Nat 20's & 1's were always automatic for me. That's how I learned in 2nd edition and just thought it was an unwritten rule since I started playing (1989).
The only time I would ever consider having a player roll for an impossible task is to determine how badly they failed. Even in that scenario, I feel like there are plenty of other factors that could tell me how bad they failed without having them make that roll and risk the circumstance where a player is pissed that they rolled a natural 20 and it was a compete waste.
For example, lets say the Rogue insists they wants to climb up a completely sheer surface 100ft high with no pitons or any other climbing gear despite being told the feat seems impossible. If the Rogue has expertise in Acrobatics or Athletics, they try to scramble up the wall, but only get so far with their momentum before gravity pulls them back down and they land squarely on their feet. If they have proficiency in either of those skills, same thing, but they can't quite land squarely and have to tuck into a roll to avoid hurting themselves. If they have no proficiency in either skill, they fall prone and take a token 1HP of damage.
I DEFINITELY never want to make a player roll for trivial tasks, because I love rewarding Skill Proficiency and Expertise. I personally find it incredibly satisfying to completely bypass a skill check because of my choices during character creation and while leveling up, and I like to reward my players in that way.
Let's give the Rogue some redemption. They want to pick a lock. and I determine that the owners of this place are either too cheap or too incompetent to fit good locks to the doors. The lock LOOKS like a standard, quality-made lock to the casual observer, but anyone with at least Proficiency in Thieves' Tools that pokes around inside will inevitably find a fatal security flaw. If the Rogue only has proficiency in Thieves' Tools, it takes a moment, but before long they found the flaw and cracked open the door. If the Rogue has expertise, then it takes mere seconds for the Rogue to effortlessly pop the lock open.
Natural 1 is always a crit fail and Natural 20 is always a crit success was how I was taught to play. I see why it can be a problem sometimes (more with the successes than failures), but it’s still my preference for a way to play.
Love your takes dude
I'm against that "nat20 always succeed" because it would for sure create a drama at he table (in a bad way). That's a rule I would change and make clear during session zero that nat20 is "best possible outcome" and not "auto success".
Using the example of a player demanding to be adopted by the king: I would definetly ask for a roll here. Not to see if the player succeeds but to check if the king will have the group arrested on the spot or just find it funny and send the group away.
I admittedly usually call for a roll when I need 10 seconds to improvise something.
Personally I really like degrees of success and failure. Like a low roll won’t mean you don’t pick the lock, but it takes you a while or is loud and someone may come to investigate. Sometimes of course a fail is a fail but if it it something a player has built into I’m more liable to go degree of success then out right failure.
Strong disagreement.
Sometimes, there is a reason to roll even if you can't succeed.
Why?
Because instead of rolling to succeed, you are rolling to determine the level of failure.
Failing with a 17 and failing with a 2 are VERY different results.
There are abilities that say, "If you fail by 5 or more" regarding a saving throw, for example
maybe you cant succeed, but maybe you can come within 5 to avoid the additional effect
For me I prefer how close the player can get to the DC can give a partial success, and if they equal or pass the DC it fully succeeds.
For instance dc is 30 to climb the extremely slippery wall. Pc only has a 9 so can never fully succeed, but what I would do is have it partially succeed if they get a 20 or higher. Then I would narrate they got partway up, and have them roll again.
Thanks for the video and advice 😃
I am strongly against the automatic 1 and 20 success and fail.
Among many problems, it devalues elements such as Bardic Inspiration and guidance that might allow a character to hit a DC 25 even if 20+their bonus does not - or could rescue a natural 1 to a passing grade through a great inspiration roll.
BUT. Even though I don't hold by a nat 20 being an auto success, it STILL should almost always be a success since a DC higher than 20 should be very rare. Too often I see DMs using absurdly high DCs (perhaps it's an arms race against optimised players?).
I believe a player should have no proficiency and a 10 in an ability score and still have a roughly even chance of success on a lot of skill checks using that. Let people be successful even when not in their speciality.
As much as I do not care as the player or what other people do (it truly is not that big of a deal), I use rolls at my table with the idea of how hard they fail or how big they succeed even if the roll would be a guaranteed success.
Personally I am not too big of a fan of people doing stuff they know would be out of this world, like fucking a dragon or murdering the queen in front of the king and staying all buddy buddy with them, so the succeed part would be out of the question for both of them. The roll at that point would just be how hard the retaliation would be from the other side. Rolling high at that point would just mean that the king might give them a minute to explain instead of instantly throwing them into the dungeons, as them instantly being set free due to a nat 20 feels weird to me.
On the other side, having a nat 1 always being a fail is pretty jarring too. A barbarian with 20 STR should EASILY be able to kick down a door, the same for your average Rouge being able to open the average door. Sure, most of the time you can (and in my opinion should) say they can just do the thing without a roll, but I have seen some really toxic moments where someone has some absurd + to the skill they are using and roll a 1 and end up hurting themselves or breaking the thing forever. Not fun either, unless the players specifically ask for that kind of thing.
All by all, it truly does not matter in the slightest. What matters is that your table communicates about their wants and needs to succeed in having fun at the table!
My general rule is only role of both success and failure is both possible however I do have an exception any knowledge checks and this includes investigation and perception by the way I’ll ask my players to role for those even if I have no information to give my reason being that way if they role low on the ones that matter they won’t be 100% they missed anything
I can't BELIVE you destroyed my long-held opinion about nat 20s and 1s being automatic successes and fails with a single sentence.
I was genuinely surprised when you said you treat them that way. But I had to literally pause and sit back and reflect when you said; "If a player doesn't pass on a 20 or fail on a 1, the player shouldn't roll at all."
I am not confident that I am 100% sold because I don't always want players to KNOW when something is possible or not. and if your standing rule is "20 passed, 1 fails" you can't have sneaky moments like that without potentially going back on the rules. But I may start implementing this in a more unofficial way. Not coming out and saying it's a rule, but thinking to myself "Will a 20 succeed/1 fail? if not, what is the value of rolling here?" Or maybe I will make it a rule. Will have to reflect further.
I've also commented on the dice being an outside factor rather than character action to a fellow DM before who is bothered that the very swingy value of a d20 can make heroes look like fools. I had offered it as an alternative way of timing, and I liked it and thought "Maybe I should start thinking of it that way too". Then promptly forgot about it. So thank you for bringing it back to my mind because I really do want to start thinking about it that way lol.
On the note of your other statement about not locking the story behind success/failure; I am struggling with this with the modual I am running. The system I use determines the degree of success not by the total roll but by the value of the special die specifically. So I'm having problems where the story is locking a lot of stuff behind getting a 4+ on the Dragon Dice(off colour d6 of a 3d6 roll), and my tracker (for example) super succeeded because of her skill and powers..but the actual die was just a 3. I'm thinking of changing how that rule works because honestly it's the one thing I don't like and feels wrong and this video helped me feel justified in that.
Roll when there is a task with a DC that can be failed based on a characters skill modifier +1.
Part of this is why I don't like the Intelligence Ability Score in name. Not all characters want to have "high intelligence" which implies to us playing that they don't know as many things or specific knowledge.
This whole debate assumes DCs are just success or fail. There are always degrees of success and failure. There is spectacular success, success as much as you'd expect, minimum success with downsides, failure but recovery and no cost, failure, critical failure with additional consequences.
Sometimes, even if they're super proficient in a skill but roll a 1, they still fail but they catch themselves and suffer no substantial consequences - but most importantly they still don't succeed. This is "a 1 is always a failure" without making 1/20 rolls a campaign-derailing consequence. Failures can be safe, or baaaad. That's also why we have different DCs and where circumstances or player actions can change the DC. they also change all those scales of minor success or fail
Nothing clever to say just here for the algorithm
Honestly RAW is a boring as hell way to play anyway. ^^;