But now the player knows he/she rolled badly, and still does not trust it. More realistic is an observant and perceptive person who knows they're doing will gain some degree of confidence after checking it. Had they rolled a Nat 20 they would know it is not trapped. Maybe the DM rolls behind the screen and adds the appropriate bonus? The idea has merit, but such a thing is taboo. Here's another one.
A moderately heavy object is in the way, let's say DC level 10 to lift it. The barbarian rolls a 1 and fails, then the wizard rolls a 20 and says, "Ha! Weakling." A comparable thing happening in real life is unrealistic.
What are some game mechanics with skill checks you think are lacking realism, and how do you fix it with house rules?
Replace all skill checks with passive scores, "taking 10" on check. Add +5 if they have helpful assistance from a teammate or otherwise might normally have advantage. Allow them to "take 20" instead by devoting an appropriately longer stretch of time. Where taking 20 is inappropriate because the situation is one with an element of risk/time sensitive, permit (or even require) a roll instead of taking 10, and forbid taking 20.
Its slightly more complicated for new players (which is probably why passive scores are so under-utilized in 5E vs prior editions), but its actually faster around the table, and leads to more predictable outcomes when the skill check doesn't have vital narrative weight. And where it does, you can still just roll checks as normal.
I think its up for the DM to make a good description and the players engagement to roleplay, more than a unrealistc mechacinc.
In the first exemple: the player make a check for traps in a chest and roll baddly. The DM announce "you get close to the chest and gently passes a needle between the slit in the lid, looking for loops or cramps that would trigger some trap, but finds nothing. You feel confidant that its safe." Than, the player know that the roll were low and its possible to have a trap there, but also it knows that the character are feeling safe and must roleplay as well.
In the secon exemple: First, if its ordinary the DM can always say "you lift the object" without ask for a check. If it isn't its becouse thats a situation that it can go wrong. If the barbarian roll bad, the DM must explain that like "You go confident as its a regular move for you. However, as you lift the object a few inches your dirty fingers slip and the object falls making a cracking noise trhough the chamber." Well, its totally plausible that a strong fellow with slippery fingers would fail that, and than they must be alert that making repetitive checks would make them to loose time and maybe call atention. Than the wizard goes and rolls pretty good. "Well, worried with the big noise the fall made, "char" as smart as he/she is, goes where he/she can use a physics at his/her advantage and easely moves the object clearing the passage." Dont' worry if its a STR check and you justify it by the chars INT, just make a plausible explanation.
Also, I trully recomend you to ban repetitive tests for a single situation. An exemple in another situation: the party is pursuing some fled foes and the ranger fails in a survival check to find its track. Than other player thows "let me roll, I also have a good bonus". You as DM must hold it and say "Well, you follow into the forest looking for footprints and broken branches in the way. After a while, the tracks become more deceptive and you are not sure if you are in the right path. You have passed trhough 1 hour of walking, you (the other player) can make another check."
Than if they want to make check after check untill they pass, they know that it will take time. You can also, rises DC and add random encounter while they keep repiting. Try to encourage them to roleplay more than think on dice results by talking to them and if they were failing to do it use that penalties as random encounters or loss of time.
Now, if they must pass that check or beeing stuck forever, I would say that it have a design problem. Make sure that they can always overcome a situation by different ways. So, if there is a heavy object in the way and the barbarian fails to lift it, the wizard may find a smart way to take it out of the way. And if the object in the way is something that the barbarian can try to lift over and over, well, there's no need to impose a check. Ask for checks only when a possible situational fail would generate another issue.
>But now the player knows he/she rolled badly, and still does not trust it
Good. They can hang onto that feeling of distrust while I roll the damage. Why are we lingering in the moment between when the trap expert said "all clear" and when the party moved forward?
If there's no penalty for failure (ex., There is no trap for them to find), consider adding a cost. Just do it before they roll. For example: "The air down here tastes foul. You can take the time to check, but you'll have to save against poison." Or, instead of doing that, just use their passive numbers and don't let them check for traps at all. Traps in 5e are usually built as quick, thematic save-or-lose-some-HP surprises, not actual encounters or death traps that you need to avoid at all costs.
The DM can always make the PC's check if he want to keep player in the dark and avoid metagaming second guessing
That't not a lie, but I don't recomend it. It can be fun-breaking for the players that shouldn't feel passive into the narrative.
I don't see problems for the DM to consider a passive result for irrelevant tasks (as lift a heavy object from the path when it have a narrative importance nor an issue upcoming from a fail). But when an active task performed by the characters whould have a significant result, I would say that it should be up to players to choose what to do and to roll that dice (unless the player agree to don't mind by the roll), but never the DM roll in secret to determine the player choice and result for it. I mean, if you have that check for traps exemple, the player would feel passive into the story creation if the DM roll for he/she.
Well, I'm not saying that it shouldn't be a option, but I think that it must have an agreement between players and DM to not make it a problem for players, and I also think that its better to encourage the players to focus more on roleplay their characters thinking and make choices in that way even when the player knows what could happen(of coure its easyer for players that also DM, but I think they always learn that its more fun after a while).
Giving an exemple: Last session in a table that I'm a PC, the DM said that he would roll my death saving trhows so the rest of the group don't know how I was going. I deny. I didn't allow him to roll my dices and I can keep secret from the rest of the team. I would feel bad if he didn't let me roll my dices and would say that he could play just with their NPCs if I couldn't be in charge of my own char.
Of course I have a bit of experience to not let metagaming overcome my choices in game but if it was I case of a check for traps for example I wouldn't let him roll for me in order to hide a possible bad result, and as DM I don't feel good to make the same with my players even when they are learning and I feel thats more my responsability to encourage them to roleplay that bad result as the char point of view and not from the players knowledge of the mechanics.
First of all, no such thing as skill checks in 5E, it's either an ability check, an attack roll or a saving throw. That said don't treat the rolls as binary "either you succeed or fail" situations but rather as a measure of how bad they succeed or fail or if the attempt has unforeseen consequences.
Take the lifting example. The barbarian rolls a one but they actually do lift the thing but strain a muscle doing so, 1D4 points of damage and disdvantage on Strentgth and Dexterity checks for the next couple of minutes (or until healed). The rogue rolling a 2 checking for traps? Yeah, they actually do notice something that looks like a trap, but when they try to disarm it, bam! Turns out it was a decoy that, depending on the roll to disarm, blows up in their face. The Ranger rolls a 12 instead of a 15 to navigate the dense forest? They still do it but it takes longer than expected so they party has to camp for the night or push through, risking exhaustion.
The basic idea is that a bad roll shouldn't prevent the story from progressing, but that progress might come at a price. Sure, meta-gaming will always be a problem, but that's a player issue. In any way, Matt Colville has done great video about the topic.
I do think some other systems handle failures/complication/success/brilliant success more interestingly than d20 systems do, though not necessarily as gracefully. Replacing single pass/fail DCs with tiered DCs is something that D&D occasionally does, when you see things like a special penalty for failing by 10 or more... but a savvy DM could roll them out in wider application, without changing much about how ability check rolls and skills calculate mechanically or are rolled.
"I check the chest for traps". Rolls a 2. "You don't find any traps" But now the player knows he/she rolled badly, and still does not trust it. More realistic is an observant and perceptive person who knows they're doing will gain some degree of confidence after checking it. Had they rolled a Nat 20 they would know it is not trapped. Maybe the DM rolls behind the screen and adds the appropriate bonus? The idea has merit, but such a thing is taboo.
If you don't want to have the DM keep the result secret (which I absolutely encourage for stealth during exploration), change how you handle the rolls. Checking for traps isn't a discrete process. There's no clear endpoint since there's always a chance they overlooked something. Looking for things is a continuous process where you increase your level of confidence the more time you put in.
In my games, when a player is investigating something specific like a chest, I usually equate 1 roll to 1 minute of activity. They can keep rolling as many times as they want. When they're searching along large surfaces like walls or floor, then that roll and minute is limited to a 5 foot area. If they decide they're going to scrutinize every nook and cranny, I say it takes them 10 minutes or 1 hour or whatever order of magnitude makes sense and jump ahead to the part where they succeed, because if they look hard enough they'll eventually find whatever's there.
Here's another one. A moderately heavy object is in the way, let's say DC level 10 to lift it. The barbarian rolls a 1 and fails, then the wizard rolls a 20 and says, "Ha! Weakling." A comparable thing happening in real life is unrealistic.
I don't use rolls for feats of strength. When I do a bench press there's no random chance I'll spontaneously fail to lift it. Unless you're doing a complex movement like an olympic weight lift, luck and skill barely factor into your capacity to move a weight. When lifting really heavy things, I set a minimum strength score and use rolls to account for endurance and fatigue.
Some other things I usually take into consideration:
Proficiency Bonus on its own doesn't do a good job of distinguishing between untrained people and experts. I often use a higher DC for untrained characters than proficient characters, and sometimes rule that an untrained character has no chance of success at all.
Straight ability checks and checks that include proficiency bonus are effectively working on different DC scales. The game's suggested max DC 30 for nearly impossible tasks is sensible for skills and tools, but 20 or 25 are better upper limits for simple ability checks with no PB.
Don't allow skills unless the task is really something you can get orders of magnitude better at with practice. If you allow Athletics for lifting weights, the Rogue with Expertise is going to outlift the actual strong characters.
Using ability checks for combat maneuvers was a really bad idea. Stick to attack rolls and saving throws.
The DMG's guidelines for social interactions are quite sensible, I suggest you use them. The short version is that the NPC's attitude towards the players sets a default reaction to persuasion attempts (e.g. an indifferent character won't take risks on your behalf by default) and beating DC 10 or DC 20 improves their response by 1 or 2 levels respectively.
The DM can always make the PC's check if he want to keep player in the dark and avoid metagaming second guessing
That't not a lie, but I don't recomend it. It can be fun-breaking for the players that shouldn't feel passive into the narrative.
I don't see problems for the DM to consider a passive result for irrelevant tasks (as lift a heavy object from the path when it have a narrative importance nor an issue upcoming from a fail). But when an active task performed by the characters whould have a significant result, I would say that it should be up to players to choose what to do and to roll that dice (unless the player agree to don't mind by the roll), but never the DM roll in secret to determine the player choice and result for it. I mean, if you have that check for traps exemple, the player would feel passive into the story creation if the DM roll for he/she.
Well, I'm not saying that it shouldn't be a option, but I think that it must have an agreement between players and DM to not make it a problem for players, and I also think that its better to encourage the players to focus more on roleplay their characters thinking and make choices in that way even when the player knows what could happen(of coure its easyer for players that also DM, but I think they always learn that its more fun after a while).
A player would be overacting if 1% of his check was fun breaking because they're sometimes made by the DM to avoid the player knowing the die result and having to rely on the DM narration to find out.
The PC check i could make as DM would be knowledge check which the result may not be apparent or obvious it is success or failure. Or Passive check could be used for the same reason, which also doesn't have playing rolling any check.
Another possible solution could be for the DM to have the player make the check in secret, so that no other player know the results and only have the DM narrative.
Giving an exemple: Last session in a table that I'm a PC, the DM said that he would roll my death saving trhows so the rest of the group don't know how I was going. I deny. I didn't allow him to roll my dices and I can keep secret from the rest of the team. I would feel bad if he didn't let me roll my dices and would say that he could play just with their NPCs if I couldn't be in charge of my own char.
Death saving throw has no bearing wether the player know the result of his PC's save or not, there's nothing he can do about it so i don't see any value in making it as a DM. In my campaign, i have my player roll their PC's death saving throw in secret though, so that other players don't know the result. It creates more tension as a result.
Tension over death saves might not be what a player wants, though. It's additional stress, right? And the regular rules say they roll it themselves. A house rule that adds more stress for a player is maybe not something to assume is fine without asking. We play the game to have fun, after all. (For context, my table is using the same rule as yours for our Avernus campaign. I'm just playing devil's advocate here. It feels odd to call this hypothetical person the devil, but that's the phrase.)
To head back to OP's question, I think a good guiding principle is something like this: Never let yourself say "roll X" when what you mean is "let me think for a moment." It's a bad habit.
A player would be overacting if 1% of his check was fun breaking because they're sometimes made by the DM to avoid the player knowing the die result and having to rely on the DM narration to find out.
The PC check i could make as DM would be knowledge check which the result may not be apparent or obvious it is success or failure. Or Passive check could be used for the same reason, which also doesn't have playing rolling any check.
Another possible solution could be for the DM to have the player make the check in secret, so that no other player know the results and only have the DM narrative.
Death saving throw has no bearing wether the player know the result of his PC's save or not, there's nothing he can do about it so i don't see any value in making it as a DM. In my campaign, i have my player roll their PC's death saving throw in secret though, so that other players don't know the result. It creates more tension as a result.
Yep. I see that you can think it's maybe overeacting, but when I said that I was thinking of how much people like to roll dices and be an active part of the narrative. Even if the possible results doesn't change by who roll the dices, people like to make that roll and specially newer players want that feeling. What I'm saying is that I see a better point in the DM to encourage players to roleplay more, than call some character rolls for him/her self.
That's why i prefer to have passive checks when situations can came with potential metagaming gessing (as much as hiden lies from an NPC for example) and let the players roll only when they call it. Also, it can be only 1% of the rolls called by the DM but it relies much on the campaing it self. I have sessions that most of the rolls can have a bit of metagaming revealed on the roll and the way I deal with it is chose the best words to make clear the players guess isn't the char guess. I think my players were improve well their roleplay in a way they wouldn't if I just call for me to roll that checks, you know (also, they've learnd to think as the char even in other situations).
Well, I already run out the topic of the forum lol. But I think you get my point right? Not that its a bad option to the DM roll for players, but I think that we have better options.
And yeah, I agree that I made a bad exemple, it was a fresh situation that came in mind. But I think I would deny even if it was a "check for traps" or "is he lying?" situation becouse I value the trust between players/DM and to improve the roleplay in the table an I see that it call the roll for him would just avoid the problem of metagaming thinking but not to solve it.
The main problem is when players are metagaming based on the result of their check, which prompt further actions. How many times do we often see party keep looking because a player rolled low on Perception or Investigation check. Proof is that when it's a high roll and they find nothing, nobody press further. That is the result of metagaming right there.
Where if DM handles knowledge roll internally, whenever he says you see nothing, it cut down metagaming
If you don't want to have the DM keep the result secret (which I absolutely encourage for stealth during exploration), change how you handle the rolls.
I encourage too. I since then also changed how i handle Stealth checks, i now only have the player make a Stealth check when a possibility of detection occur. It avoid unecessary rolls in the event that no one comes across the hidden PC, and it avoid any possible metagaming based on the Stealth check beforehand.
I encourage too. I since then also changed how i handle Stealth checks, i now only have the player make a Stealth check when a possibility of detection occur. It avoid unecessary rolls in the event that no one comes across the hidden PC, and it avoid any possible metagaming based on the Stealth check beforehand.
Yep. I ask for stealth checks only when thay say that want to go unnoticed (so they roll even if there is no one around) and keep that roll noted for until it needed or they change tatics. Otherwise, I assume they arn't being stealthy and any NPC with perception 10 or more notice them, or I roll for it.
Also I usually ask how they want to be stealthy, depending on their description I evaluate the better way to handle that situation without rolls. Unless it matters for the narrative to use that luck factor.
Oh i never assume PCs are stealthy unless they specifically say they try to and don't make them do any Stealth checks until it becomes necessary basically.
I ask for stealth checks only when thay say that want to go unnoticed (so they roll even if there is no one around) and keep that roll noted for until it needed or they change tatics.
Also, about stealth checks, I describe their success based on the result. If the roll ar too low they aren't stealthy and also make clear noise or something that they know they fail, if its not too low or in the middle, they just assume that they are being stealth (in metagame they know that it relies on perception from other so), if its too high characters feel confident (as much as the players would).
In other checks, that relies on fixed DC, thats not so easy to do that, so when they roll badly I just say "you (character) know you fail, but you already loose some time in it and must rush". I mean, I make they to always assume that even when they know that its a fail they must deal with some consequence. If it's not too low, but a fail I try to make them to feel in the char perspective and if the players were reluctant I call for a out-of-game chat and say "you don't know if it has a DC nor if its a high one, theres no reason for you to be paranoid with it" and in some time they are showing awlays more engagement with a good roleplay.
When you say you ask stealth checks only when there's a chance of detection, you didn't meant you ban it from other situations right? I mean, I never ask that check when I know as DM that it isn't needed. But depending on the narrative, sometimes my playes ask for it. In those cases I make them roll even if there's no detection possible, so they can't know that is safe there and the metagaming don't become a problem.
I encourage too. I since then also changed how i handle Stealth checks, i now only have the player make a Stealth check when a possibility of detection occur. It avoid unecessary rolls in the event that no one comes across the hidden PC, and it avoid any possible metagaming based on the Stealth check beforehand.
It only prevents metagaming if the result is only valid for that one group of enemies. And if that's the case, rolling every time they come across an enemy makes it increasingly likely they'll fail. Also, if the observer is hidden, asking for the Stealth roll tips the players off, which also results in metagaming.
Rolling for them when they declare they're being sneaky and keeping the result secret sidesteps all of the problems. They can't metagame with the roll results, you can compare their rolls to passive Perception without them ever knowing, and if they roll well they get to keep it.
Rollback Post to RevisionRollBack
The Forum Infestation (TM)
To post a comment, please login or register a new account.
"I check the chest for traps"
Rolls a 2
"You don't find any traps"
But now the player knows he/she rolled badly, and still does not trust it. More realistic is an observant and perceptive person who knows they're doing will gain some degree of confidence after checking it. Had they rolled a Nat 20 they would know it is not trapped. Maybe the DM rolls behind the screen and adds the appropriate bonus? The idea has merit, but such a thing is taboo. Here's another one.
A moderately heavy object is in the way, let's say DC level 10 to lift it. The barbarian rolls a 1 and fails, then the wizard rolls a 20 and says, "Ha! Weakling." A comparable thing happening in real life is unrealistic.
What are some game mechanics with skill checks you think are lacking realism, and how do you fix it with house rules?
Replace all skill checks with passive scores, "taking 10" on check. Add +5 if they have helpful assistance from a teammate or otherwise might normally have advantage. Allow them to "take 20" instead by devoting an appropriately longer stretch of time. Where taking 20 is inappropriate because the situation is one with an element of risk/time sensitive, permit (or even require) a roll instead of taking 10, and forbid taking 20.
Its slightly more complicated for new players (which is probably why passive scores are so under-utilized in 5E vs prior editions), but its actually faster around the table, and leads to more predictable outcomes when the skill check doesn't have vital narrative weight. And where it does, you can still just roll checks as normal.
dndbeyond.com forum tags
I'm going to make this way harder than it needs to be.
The DM can always make the PC's check if he want to keep player in the dark and avoid metagaming second guessing
I think its up for the DM to make a good description and the players engagement to roleplay, more than a unrealistc mechacinc.
In the first exemple: the player make a check for traps in a chest and roll baddly. The DM announce "you get close to the chest and gently passes a needle between the slit in the lid, looking for loops or cramps that would trigger some trap, but finds nothing. You feel confidant that its safe." Than, the player know that the roll were low and its possible to have a trap there, but also it knows that the character are feeling safe and must roleplay as well.
In the secon exemple: First, if its ordinary the DM can always say "you lift the object" without ask for a check. If it isn't its becouse thats a situation that it can go wrong. If the barbarian roll bad, the DM must explain that like "You go confident as its a regular move for you. However, as you lift the object a few inches your dirty fingers slip and the object falls making a cracking noise trhough the chamber." Well, its totally plausible that a strong fellow with slippery fingers would fail that, and than they must be alert that making repetitive checks would make them to loose time and maybe call atention. Than the wizard goes and rolls pretty good. "Well, worried with the big noise the fall made, "char" as smart as he/she is, goes where he/she can use a physics at his/her advantage and easely moves the object clearing the passage." Dont' worry if its a STR check and you justify it by the chars INT, just make a plausible explanation.
Also, I trully recomend you to ban repetitive tests for a single situation. An exemple in another situation: the party is pursuing some fled foes and the ranger fails in a survival check to find its track. Than other player thows "let me roll, I also have a good bonus". You as DM must hold it and say "Well, you follow into the forest looking for footprints and broken branches in the way. After a while, the tracks become more deceptive and you are not sure if you are in the right path. You have passed trhough 1 hour of walking, you (the other player) can make another check."
Than if they want to make check after check untill they pass, they know that it will take time. You can also, rises DC and add random encounter while they keep repiting. Try to encourage them to roleplay more than think on dice results by talking to them and if they were failing to do it use that penalties as random encounters or loss of time.
Now, if they must pass that check or beeing stuck forever, I would say that it have a design problem. Make sure that they can always overcome a situation by different ways. So, if there is a heavy object in the way and the barbarian fails to lift it, the wizard may find a smart way to take it out of the way. And if the object in the way is something that the barbarian can try to lift over and over, well, there's no need to impose a check. Ask for checks only when a possible situational fail would generate another issue.
>But now the player knows he/she rolled badly, and still does not trust it
Good. They can hang onto that feeling of distrust while I roll the damage. Why are we lingering in the moment between when the trap expert said "all clear" and when the party moved forward?
If there's no penalty for failure (ex., There is no trap for them to find), consider adding a cost. Just do it before they roll. For example: "The air down here tastes foul. You can take the time to check, but you'll have to save against poison." Or, instead of doing that, just use their passive numbers and don't let them check for traps at all. Traps in 5e are usually built as quick, thematic save-or-lose-some-HP surprises, not actual encounters or death traps that you need to avoid at all costs.
That't not a lie, but I don't recomend it. It can be fun-breaking for the players that shouldn't feel passive into the narrative.
I don't see problems for the DM to consider a passive result for irrelevant tasks (as lift a heavy object from the path when it have a narrative importance nor an issue upcoming from a fail). But when an active task performed by the characters whould have a significant result, I would say that it should be up to players to choose what to do and to roll that dice (unless the player agree to don't mind by the roll), but never the DM roll in secret to determine the player choice and result for it. I mean, if you have that check for traps exemple, the player would feel passive into the story creation if the DM roll for he/she.
Well, I'm not saying that it shouldn't be a option, but I think that it must have an agreement between players and DM to not make it a problem for players, and I also think that its better to encourage the players to focus more on roleplay their characters thinking and make choices in that way even when the player knows what could happen(of coure its easyer for players that also DM, but I think they always learn that its more fun after a while).
Giving an exemple: Last session in a table that I'm a PC, the DM said that he would roll my death saving trhows so the rest of the group don't know how I was going. I deny. I didn't allow him to roll my dices and I can keep secret from the rest of the team. I would feel bad if he didn't let me roll my dices and would say that he could play just with their NPCs if I couldn't be in charge of my own char.
Of course I have a bit of experience to not let metagaming overcome my choices in game but if it was I case of a check for traps for example I wouldn't let him roll for me in order to hide a possible bad result, and as DM I don't feel good to make the same with my players even when they are learning and I feel thats more my responsability to encourage them to roleplay that bad result as the char point of view and not from the players knowledge of the mechanics.
First of all, no such thing as skill checks in 5E, it's either an ability check, an attack roll or a saving throw. That said don't treat the rolls as binary "either you succeed or fail" situations but rather as a measure of how bad they succeed or fail or if the attempt has unforeseen consequences.
Take the lifting example. The barbarian rolls a one but they actually do lift the thing but strain a muscle doing so, 1D4 points of damage and disdvantage on Strentgth and Dexterity checks for the next couple of minutes (or until healed). The rogue rolling a 2 checking for traps? Yeah, they actually do notice something that looks like a trap, but when they try to disarm it, bam! Turns out it was a decoy that, depending on the roll to disarm, blows up in their face. The Ranger rolls a 12 instead of a 15 to navigate the dense forest? They still do it but it takes longer than expected so they party has to camp for the night or push through, risking exhaustion.
The basic idea is that a bad roll shouldn't prevent the story from progressing, but that progress might come at a price. Sure, meta-gaming will always be a problem, but that's a player issue. In any way, Matt Colville has done great video about the topic.
I do think some other systems handle failures/complication/success/brilliant success more interestingly than d20 systems do, though not necessarily as gracefully. Replacing single pass/fail DCs with tiered DCs is something that D&D occasionally does, when you see things like a special penalty for failing by 10 or more... but a savvy DM could roll them out in wider application, without changing much about how ability check rolls and skills calculate mechanically or are rolled.
dndbeyond.com forum tags
I'm going to make this way harder than it needs to be.
If you don't want to have the DM keep the result secret (which I absolutely encourage for stealth during exploration), change how you handle the rolls. Checking for traps isn't a discrete process. There's no clear endpoint since there's always a chance they overlooked something. Looking for things is a continuous process where you increase your level of confidence the more time you put in.
In my games, when a player is investigating something specific like a chest, I usually equate 1 roll to 1 minute of activity. They can keep rolling as many times as they want. When they're searching along large surfaces like walls or floor, then that roll and minute is limited to a 5 foot area. If they decide they're going to scrutinize every nook and cranny, I say it takes them 10 minutes or 1 hour or whatever order of magnitude makes sense and jump ahead to the part where they succeed, because if they look hard enough they'll eventually find whatever's there.
I don't use rolls for feats of strength. When I do a bench press there's no random chance I'll spontaneously fail to lift it. Unless you're doing a complex movement like an olympic weight lift, luck and skill barely factor into your capacity to move a weight. When lifting really heavy things, I set a minimum strength score and use rolls to account for endurance and fatigue.
Some other things I usually take into consideration:
The Forum Infestation (TM)
A player would be overacting if 1% of his check was fun breaking because they're sometimes made by the DM to avoid the player knowing the die result and having to rely on the DM narration to find out.
The PC check i could make as DM would be knowledge check which the result may not be apparent or obvious it is success or failure. Or Passive check could be used for the same reason, which also doesn't have playing rolling any check.
Another possible solution could be for the DM to have the player make the check in secret, so that no other player know the results and only have the DM narrative.
Death saving throw has no bearing wether the player know the result of his PC's save or not, there's nothing he can do about it so i don't see any value in making it as a DM. In my campaign, i have my player roll their PC's death saving throw in secret though, so that other players don't know the result. It creates more tension as a result.
Tension over death saves might not be what a player wants, though. It's additional stress, right? And the regular rules say they roll it themselves. A house rule that adds more stress for a player is maybe not something to assume is fine without asking. We play the game to have fun, after all. (For context, my table is using the same rule as yours for our Avernus campaign. I'm just playing devil's advocate here. It feels odd to call this hypothetical person the devil, but that's the phrase.)
To head back to OP's question, I think a good guiding principle is something like this: Never let yourself say "roll X" when what you mean is "let me think for a moment." It's a bad habit.
Yep. I see that you can think it's maybe overeacting, but when I said that I was thinking of how much people like to roll dices and be an active part of the narrative. Even if the possible results doesn't change by who roll the dices, people like to make that roll and specially newer players want that feeling. What I'm saying is that I see a better point in the DM to encourage players to roleplay more, than call some character rolls for him/her self.
That's why i prefer to have passive checks when situations can came with potential metagaming gessing (as much as hiden lies from an NPC for example) and let the players roll only when they call it. Also, it can be only 1% of the rolls called by the DM but it relies much on the campaing it self. I have sessions that most of the rolls can have a bit of metagaming revealed on the roll and the way I deal with it is chose the best words to make clear the players guess isn't the char guess. I think my players were improve well their roleplay in a way they wouldn't if I just call for me to roll that checks, you know (also, they've learnd to think as the char even in other situations).
Well, I already run out the topic of the forum lol. But I think you get my point right? Not that its a bad option to the DM roll for players, but I think that we have better options.
And yeah, I agree that I made a bad exemple, it was a fresh situation that came in mind. But I think I would deny even if it was a "check for traps" or "is he lying?" situation becouse I value the trust between players/DM and to improve the roleplay in the table an I see that it call the roll for him would just avoid the problem of metagaming thinking but not to solve it.
The main problem is when players are metagaming based on the result of their check, which prompt further actions. How many times do we often see party keep looking because a player rolled low on Perception or Investigation check. Proof is that when it's a high roll and they find nothing, nobody press further. That is the result of metagaming right there.
Where if DM handles knowledge roll internally, whenever he says you see nothing, it cut down metagaming
I encourage too. I since then also changed how i handle Stealth checks, i now only have the player make a Stealth check when a possibility of detection occur. It avoid unecessary rolls in the event that no one comes across the hidden PC, and it avoid any possible metagaming based on the Stealth check beforehand.
Yep. I ask for stealth checks only when thay say that want to go unnoticed (so they roll even if there is no one around) and keep that roll noted for until it needed or they change tatics. Otherwise, I assume they arn't being stealthy and any NPC with perception 10 or more notice them, or I roll for it.
Also I usually ask how they want to be stealthy, depending on their description I evaluate the better way to handle that situation without rolls. Unless it matters for the narrative to use that luck factor.
I mean, perception from NPCs
Oh i never assume PCs are stealthy unless they specifically say they try to and don't make them do any Stealth checks until it becomes necessary basically.
Also, about stealth checks, I describe their success based on the result. If the roll ar too low they aren't stealthy and also make clear noise or something that they know they fail, if its not too low or in the middle, they just assume that they are being stealth (in metagame they know that it relies on perception from other so), if its too high characters feel confident (as much as the players would).
In other checks, that relies on fixed DC, thats not so easy to do that, so when they roll badly I just say "you (character) know you fail, but you already loose some time in it and must rush". I mean, I make they to always assume that even when they know that its a fail they must deal with some consequence. If it's not too low, but a fail I try to make them to feel in the char perspective and if the players were reluctant I call for a out-of-game chat and say "you don't know if it has a DC nor if its a high one, theres no reason for you to be paranoid with it" and in some time they are showing awlays more engagement with a good roleplay.
Oh I didn't assume that.
When you say you ask stealth checks only when there's a chance of detection, you didn't meant you ban it from other situations right? I mean, I never ask that check when I know as DM that it isn't needed. But depending on the narrative, sometimes my playes ask for it. In those cases I make them roll even if there's no detection possible, so they can't know that is safe there and the metagaming don't become a problem.
It only prevents metagaming if the result is only valid for that one group of enemies. And if that's the case, rolling every time they come across an enemy makes it increasingly likely they'll fail. Also, if the observer is hidden, asking for the Stealth roll tips the players off, which also results in metagaming.
Rolling for them when they declare they're being sneaky and keeping the result secret sidesteps all of the problems. They can't metagame with the roll results, you can compare their rolls to passive Perception without them ever knowing, and if they roll well they get to keep it.
The Forum Infestation (TM)