The DM Rolls

[Pex]

Another old school DMing technique - the DM rolling for the player.

The purpose of it is to avoid metagaming. PC searches for traps. The DM rolls for it so that when he says the PC doesn't find any traps the player doesn't know if it's because there really isn't a trap or there is a trap but the character missed it because of a low roll. A DM would roll a PC's saving throw behind the screen because the PC/Player doesn't know someone or something happened to cause the saving throw to be made. Another reason for this is to stop chain rolling. A player rolls low on investigating something, a knowledge check, or insight so another player rolls. The DM rolls instead so that an answer of nothing or don't know doesn't tell a player anything about the die roll.

This was way popular in my 2E days. I hated it but accepted it because that's how I thought it was supposed to be done for immersion. I did it myself as DM and stopped when a player finally objected. I needed someone else to agree with me it was bad. The player should always roll his own fate. Now in 5E Passive score use is fine. The bad guys can roll stealth against the PCs' Passive Perception. That I find fair. However, player should always roll their own skill use and saving throw. The DM is correct to enforce no chain rolling if it's a bother, but he should trust the players when they roll low for searching for traps they'll open the door/chest or walk down the hallway when they missed the trap. DMs just have to accept it the player knows he had to make a saving throw even if the PC doesn't know the source.

[Khedrac]

I go for a much simpler solution: at the start of each session I get each player to roll 3 or 5 rolls, and I note down key skills and saves (so for D&D 3.5 I note down spot, listen, sense motive and their will save).
Then, as the play proceeds I just use the rolls in order if they are needed, if a player runs out I ask them to roll another set.

[OldTrees1]

Recently I ran a session where the party was travelling overland through a meticulous web of illusions. The party was going to cover a lot of ground and I had no way of knowing how long their journey would be.

Party of 6 x Roughly 8 illusions per mile x N miles (ended up being 3 days of travel by horse) = Way too many checks. Okay, what about passive? Well the DCs are rather high and staggered (meticulous web). Using the passives would reduce it to just the specialized 2.5 character but their passives are below the highest DC. Oh, and what if I only use 1 check per PC per set of illusions?

I had the players do the rolls for the first hour. After that the challenge became overcoming the scenario rather than doing the checks, so I took over the rolls so the players could focus on how to overcome this obstacle with their ingenuity and resources. This also removed the distraction of "A 16 noticed 4 illusions here but a 22 back there noticed 6" and let the party focus on "How do we get to our destination when the land itself is enchanted".

I think it is one of those tools that can be quite useful but some (not all) players don't like it. Don't use it if a player objects, but if the players are fine with it, use it when it is the right tool for the job.

[J-H]

I don't use passive Perception very often. They're adventurers in a dungeon, they're actively alert unless occupied with something like eating or memorizing spells. I use an Excel sheet with pre-rolled D20 values, 1 column per character, with the character names and modifiers noted at the top. I can usually glance at whichever line I'm on and go "Ok, anybody with a 12 or lower definitely didn't make the DC, what's the modifier on the higher rolls?"

Players appreciated it.

[PhoenixPhyre]

I don't use passive Perception very often. They're adventurers in a dungeon, they're actively alert unless occupied with something like eating or memorizing spells. I use an Excel sheet with pre-rolled D20 values, 1 column per character, with the character names and modifiers noted at the top. I can usually glance at whichever line I'm on and go "Ok, anybody with a 12 or lower definitely didn't make the DC, what's the modifier on the higher rolls?"

Players appreciated it.

Passive doesn't mean that the characters are passive, just that the player doesn't roll. In the case of perception, it reflects the baseline that you see unless you're actively distracted.

[Mastikator]

I go for a much simpler solution: at the start of each session I get each player to roll 3 or 5 rolls, and I note down key skills and saves (so for D&D 3.5 I note down spot, listen, sense motive and their will save).
Then, as the play proceeds I just use the rolls in order if they are needed, if a player runs out I ask them to roll another set.

I'm stealing this technique.

[Elysiume]

As a player, I don't really see the point in pre-rolling some rolls for the DM to use for secret rolls. If you're doing a small set, it doesn't entirely remove the metagame aspect: if I roll 15, 18, 14, 19 as my set, I know I did pretty well until they ask me for more rolls. I still get to feel like they're "my" rolls, but I don't care about that all that much. If you're doing a big set — dozens or more — I don't understand how that's much different from them rolling on the spot. If I give my DM a list of a hundred actual rolls that they go through at their own pace, I wouldn't consider that meaningfully different from them just getting a fresh random roll. If I didn't trust my DM to do hidden checks fairly, I wouldn't be playing with them in the first place. Giving them a list of "my" rolls for them to run through gives me an illusion of control, but I wouldn't bother if I had the choice.

[Composer99]

I'd rather the PCs either roll themselves or use passives/take 10, myself.

I mean, most of the time the thing isn't trapped, whatever it is, so it's not a big deal if they flub the roll

[Pex]

I go for a much simpler solution: at the start of each session I get each player to roll 3 or 5 rolls, and I note down key skills and saves (so for D&D 3.5 I note down spot, listen, sense motive and their will save).
Then, as the play proceeds I just use the rolls in order if they are needed, if a player runs out I ask them to roll another set.

I've played with this before. Not a fan. The player physically rolls the dice, but it's still a secret result. The DM saying you fail the save at game play still feels like a lack of control despite having physically rolled a 2 an hour ago before the game session started. I want to roll the die at the moment it happens. This also won't work in 5E specifically because various class abilities can affect skill checks and saving throws such as Portent and Bardic Inspiration.

[Satinavian]

The technique itself is fine and fullfills its purpose.

But imho it is rarely worth it and i don't really see it used. It is just too much extra hassle for the DM just to prevent a possible bit of extra metagaming.




And then many games have rule elements that make it impossible anyway. Things like rerolls or metacurrencies to influence outcomes do exist. Or abilities that actuivate on good or bad rolls. None of that works well when the GM rolls.

[Pleh]

I don't think I would ever take away roll from my players.

Rolling their own dice they bought with their own money is a rather integral part of the experience for my group.

For specific cases where I want to avoid metagaming, I find the most success by dealing with the metagame in my DM prep. I'm treating the question as generally referring specifically to perception skills, but it should apply to most knowledge (thus meta knowledge) skills.

Can the party even afford to fail the check? Always make sure the game doesn't hinge on the success or failure of a skill check. At least know what happens in either outcome and make sure even the Failure State leads the players forward (if less optimally) so players aren't sticking around at a "locked door" rolling repeatedly to appease the RNG gods.

I am considering a skill check. What DC should it be?

Now compare this to my players' modifiers/passives. Does it fall off the spectrum, where none of the players can reach, or where no one can actually fail? If yes, either skip rolling entirely or reconsider what I am even bothering to include in my game or at least how it is presented.

Is the success of the check limited to the Specialized Perception Characters (SPC)? Do I want to hand this to the SPCs as a freebie, or will they still have to roll for it? If only the SPCs have a chance to find and they still have to roll, am I spending too much time prepping content the players are statistically unlikely to discover? I try to make sure this content has high value, but is not necessary for moving forward (high importance, low urgency). It shouldn't matter if they miss it, but they should feel proud of their characters for finding something especially useful to their party for their exceptional and timely perception. A hidden treasure chest inside an elaborate trap room (where the player goal is moving through the room to the dungeon beyond) is exactly this kind of content.

If the skill check is low enough for Non SPCs to reasonably make the check, will the SPCs still even need to roll, or are their Passives high enough? Does the party only need one person to Notice X, or do they gain more reward for the number of party members who make the check (such as in a monster ambush where players attempt to avoid surprise)? If only one PC needs to Notice X, will they be free to simply point it out to their party members?

Wheb prepping, I find a common DM impulse to gamify everything that *could* be described with a skill check as a check, both/either because we want to honor the rules, and/or because it isn't much of a game if everything hidden is trivial or impossible to find.

But I feel I run better games when I let this half baked flow chart help me verify that the skill check in question makes sense to run as an actual check.

When it turns out I shouldn't be making this content into a check, but want to preserve the game of rolling dice for the players, that's when I can start exploring how to change my plot to *make it* into a logical skill check. How can I make my trivial content nontrivial to find? How can I make the impossible content actually accessible? Is the entire idea just rubbish and should I move on or replace it with a better idea?

[Quertus]

Such a deceptively big topic. Yet y'all have already covered most of the sub-topics. Kudos!

So, yes, there are times when there are just too may rolls. For example, I have written software to, for example, handle rolling mass hit locations for Battletech mechs with 40+ Machine Guns, or 10+ SRM launchers. I think one had the option to also roll "to hit" / number of missiles rolls, too. Or to handle masses of attacks from undead / summoned minions. Or even to handle initiative (for systems where it isn't a fun minigame, like it is in 2e D&D).

Metagaming? Honestly, that's arguably worse when the GM is just making the rolls - why is the GM rolling dice? OTOH, if an effect is supposed to be subtle (and, no, "make a reflex save vs death from the giant boulder rolling over you" is not something subtle that you have to worry about the player metagaming) and the players cannot roleplay not knowing about it, it's time to realize that the players and the adventure don't match, and pitch one of them.

Having a bunch of pre-rolled dice is... meh. As others have already covered.

Perhaps I should add, the GM choosing unilaterally to use a particular technique is the same hubris as the GM choosing to role for the players, or the GM choosing to Railroad their story. Have an honest, open conversation with the players about the various tools available for the job, and the problems that you would like to solve via those tools, and crowdsource to see what the players have to say about this.

Actually, don't. I'm going to roll your Wisdom checks for each of you to see which of you will enact which tactics at your tables, and Charisma checks for each of you to see how well it will go over.

[King of Nowhere]

I go for a much simpler solution: at the start of each session I get each player to roll 3 or 5 rolls, and I note down key skills and saves (so for D&D 3.5 I note down spot, listen, sense motive and their will save).
Then, as the play proceeds I just use the rolls in order if they are needed, if a player runs out I ask them to roll another set.

as a player, i've done something similar occasionally. when i go scouting, i preroll some spot checks and tell the dm those are for the next times i will need them. so in case i miss i won't know i'm missing something.
then again, with a maxxed out spot check, i rarely failed anyway

[PhoenixPhyre]

I don't think I would ever take away roll from my players.

Rolling their own dice they bought with their own money is a rather integral part of the experience for my group.

For specific cases where I want to avoid metagaming, I find the most success by dealing with the metagame in my DM prep. I'm treating the question as generally referring specifically to perception skills, but it should apply to most knowledge (thus meta knowledge) skills.



Wheb prepping, I find a common DM impulse to gamify everything that *could* be described with a skill check as a check, both/either because we want to honor the rules, and/or because it isn't much of a game if everything hidden is trivial or impossible to find.

But I feel I run better games when I let this half baked flow chart help me verify that the skill check in question makes sense to run as an actual check.

When it turns out I shouldn't be making this content into a check, but want to preserve the game of rolling dice for the players, that's when I can start exploring how to change my plot to *make it* into a logical skill check. How can I make my trivial content nontrivial to find? How can I make the impossible content actually accessible? Is the entire idea just rubbish and should I move on or replace it with a better idea?

I like this thought process immensely. I've done similar things and it's always worked well.

[Duff]

Another way to reduce metagaming is to roll when the result becomes apparent, not when it's triggered.
For example:
The party turns a corner and finds a 60' long corridor...

GM "How do you want to deal with this?"
The party describe how they deal with it - Send the rogue right to the end first checking for traps while being as sneaky as possible. The others will wait at the end, ready to rush in if the rogue is attacked.

"GM - "Roll stealth and detect traps"
Rogue - "16 on stealth, 8 to find traps"
GM - "As you get to the end of the corridor you start to feel woozy. You suspect gas may have been flowing into the corridor for a minute or 2. Make a fortitude save"

[MrStabby]

Another way to reduce metagaming is to roll when the result becomes apparent, not when it's triggered.
For example:
The party turns a corner and finds a 60' long corridor...

GM "How do you want to deal with this?"
The party describe how they deal with it - Send the rogue right to the end first checking for traps while being as sneaky as possible. The others will wait at the end, ready to rush in if the rogue is attacked.

"GM - "Roll stealth and detect traps"
Rogue - "16 on stealth, 8 to find traps"
GM - "As you get to the end of the corridor you start to feel woozy. You suspect gas may have been flowing into the corridor for a minute or 2. Make a fortitude save"

I use this one quite a bit. You say you are being stealthy, you roll when you meet someone to determine if they are prepared. You craft a potion - roll the alchemy check when you drink it...



Generally I have resigned myself to not being a perfect DM. I have limited attention, limited prep time and all this is a budget I have to spend at the table. It is easier for me to just get players to roll and to describe whatis happening than to manage it all myself. I would rather spend my attention thinking about the impact of their actions and how to convey it in the adventure than handle some of this.

I think it is part of a wider problem though; there are a lot of things to metagame this way. If I have an enemy in 5th edition deal 14 points of psychic damage players will sart thinking it is an illusion - because I declared is psychic. Now i it is described as fire damage and they see me roll 14 but they are resistant to fire damage... then they are able to make some additional inferences from how you played the game mechanics. I tend to just suck it up. Its fine.

Its usually easier and better to let the players do stuff. Not perfect for giving out the right amount of information, but good enough. I also subscribe to the belief that winning is not valuable without the risk of losing. Survival is meaningless without the prospect of death. Open rolling from everyone is a signal that you won't fudge any die roll; it adds to the tension.

Indeed a bit of metagaming can be fun. If players think there is something they don't know, then it can be an OK way to raise the tension, if used sparingly.

[Luccan]

I think it's very rarely going to ruin anything if the PC knows they need to roll. Moreover, I'm fairly certain there are scenarios in 5e where they have to know in order to properly make use of their abilities.

That said, I have heard that for PbP DM rolling is generally faster. Of course, you can do this in such a way that rolls are made public, so players can provide the input they need to on their rolls. Additionally, this is done for the sake of expediency, whereas DMs rolling for PCs at the table seems to largely be about hiding the results from players or not trusting them to report honestly. Though it's possible it would speed the game up a little for DMs to roll all the dice at the table, I'm willing to bet the same would be true if PCs rolled all the dice and doubtful as many DMs would agree to that as they would to DM-rolling. Or at the very least that the crossover would have notable decline.

[Quertus]

I think it's very rarely going to ruin anything if the PC knows they need to roll. Moreover, I'm fairly certain there are scenarios in 5e where they have to know in order to properly make use of their abilities.

That said, I have heard that for PbP DM rolling is generally faster. Of course, you can do this in such a way that rolls are made public, so players can provide the input they need to on their rolls. Additionally, this is done for the sake of expediency, whereas DMs rolling for PCs at the table seems to largely be about hiding the results from players or not trusting them to report honestly. Though it's possible it would speed the game up a little for DMs to roll all the dice at the table, I'm willing to bet the same would be true if PCs rolled all the dice and doubtful as many DMs would agree to that as they would to DM-rolling. Or at the very least that the crossover would have notable decline.

Lol. I've done the "the players make all rolls" thing. I've handed people the stats of monsters (or artillery, or...), and just let the players roll against themselves. Made the game a lot faster than waiting for 1 GM to roll the dice for *all* the monsters (or artillery, or...). It works great.

[MoiMagnus]

Another way to reduce metagaming is to roll when the result becomes apparent, not when it's triggered.

This tip can be applied to a lot of unrelated situations.

Did you take the appropriate tools for the job? Roll to know if your character though about taking those before going adventuring.
[If you start using this method frequently, you might need to create a new skill for that.]

[Democratus]

When playing online (Roll20, for example) you can get the best of both worlds.

Set up a macro for skill checks where the player rolls, but the roll is sent to the GM only (even the player doesn't see the result).

That way the player is rolling the dice and the GM is telling them what the result is without metagame interference.

Note: I've also done this in real-life games with a dice tower the players can drop dice in - but the tray for the tower is behind my GM screen.

[Jason]

Use the passive perception rules, unless the players get suspicious and ask to actively search for traps/secret doors. If you players have spent the resources on getting a high enough passive perception score to spot all of your traps and secret doors then they should benefit from it.

If it's a case of spotting illusions then yes, the DM should make the rolls for them so that they don't know what was being rolled for or if they succeeded. You could do the "give me a bunch of rolls before play" method, but I prefer occasionally rolling dice for no real reason just to keep the players on their toes anyway, so I usually roll those myself.

[Pex]

When playing online (Roll20, for example) you can get the best of both worlds.

Set up a macro for skill checks where the player rolls, but the roll is sent to the GM only (even the player doesn't see the result).

That way the player is rolling the dice and the GM is telling them what the result is without metagame interference.

Note: I've also done this in real-life games with a dice tower the players can drop dice in - but the tray for the tower is behind my GM screen.

It's not about only who rolls the physical die. It's about the player knowing his own roll. I want to know I rolled a Natural 2 for whatever it is I need to roll. In the specific case of searching for traps let me play it out I don't think there's a trap whether one is really there or not.

[RedMage125]

I used to let them make all their own rolls. Then I experienced a player rolled really low for a Search for Traps roll, and everyone behaved like it was still trapped, instead of behaving like the search proved there were no traps (it WAS trapped, but that's neither here nor there).

From then on, and I make this clear to my players before beginning of my games, I make that roll for players, asking for their modifier, and just tell them "it is trapped" or "you find no traps". I will make a roll behind the screen, whether there is a trap or not.

On that note, I do something similar for Sense Motive/Insight. I will only use Passive Insight if the NPC rolls REALLY poorly for Bluff/Deception, or, for story purposes, is a very bad liar. Other than that, players have to ASK to make an Insight/Sense Motive check. But here's the fun part: I will roll a Bluff/Deception check behind the screen whenever a player asks for such, even if the NPC isn't lying. And my house rule is this: Any character who BELIEVES that what they are saying is true gets a +30 circumstance bonus to said Bluff check. So they ask for Insight, I roll Deception, and my usual response is "you trust him implicitly", or something. The best part is, there was an Adventure Path I was running once, and at high levels, there was this NPC that was helping the players to serve her own agenda. She appeared to be an elven sorceress, but was actually an evil, vampiric silver dragon. Players were very high level, and somethign she said made one player suspicious and he made a Sense Motive check, with a result in the 40s or 50s...really high roll. Said NPC had a REALLY high Bluff modifier, and just barely beat the player (by like 4 points). Given my usual responses on Bluff/SenseMotive checks in the past when said NPC was telling the truth, my players suspected nothing and were thus genuinely surprised when this NPC betrayed them. Worked great.

[Darth Credence]

Use the passive perception rules, unless the players get suspicious and ask to actively search for traps/secret doors. If you players have spent the resources on getting a high enough passive perception score to spot all of your traps and secret doors then they should benefit from it.

If it's a case of spotting illusions then yes, the DM should make the rolls for them so that they don't know what was being rolled for or if they succeeded. You could do the "give me a bunch of rolls before play" method, but I prefer occasionally rolling dice for no real reason just to keep the players on their toes anyway, so I usually roll those myself.

I pretty much constantly roll dice. Giving descriptions, drop some dice in the tower. Listening to player plans, drop some dice in the tower. Get a slice of pizza, drop some dice in the tower. They will never be able to figure out if me dropping dice in the tower has meaning.

[gijoemike]

The GM Should NEVER roll for the players. There are feats, powers, spells in various systems that allow for manipulation of rolls after the result. Most commonly is the choice to reroll before the result is revealed. If a perception/spot roll is called for and a 2 gets rolled the player has the power to use a ability at that point. perhaps a +5 once a day power or a reroll accept the second result.

Secondly, If the player rolls a 14 and fails a skill check for the ninth time it indicates to the player that the GM sets DC quite high and that to keep up they need to max out rolls they wish to be good at which means fewer skills per PC. Open rolls show the power levels and expectations of all players in the game and ensure they are on the same page.

What about spells that specifically call for the GM to roll, e.g. Augury and Divination? There are fewer powers that affect those rolls but they still exist. Therefore the player gets to make the call for up or down. They need to have a base line otherwise it is blind coin flips. The ebb and flow of luck should lean in the heroes favor. Its one of the things that make them a BDH.

I had a GM for a very short time that would do all search/spot/perception/sense motive/attack rolls/climb/jump/bluff rolls for the PC. It got real old fast. Players want the dice to clatter on the table and see results, good or bad.

[Democratus]

The GM Should NEVER roll for the players. There are feats, powers, spells in various systems that allow for manipulation of rolls after the result. Most commonly is the choice to reroll before the result is revealed. If a perception/spot roll is called for and a 2 gets rolled the player has the power to use a ability at that point. perhaps a +5 once a day power or a reroll accept the second result.

This isn't a D&D 5e sub-forum. So much of that only applies to that particular edition of that particular game.

In OSR games, DM rolls for the characters aren't unusual at all. Many of the original DMs did a great deal of this, and the players still managed to have enough fun that it launched an entire industry.

[Duff]

It's not about only who rolls the physical die. It's about the player knowing his own roll. I want to know I rolled a Natural 2 for whatever it is I need to roll. In the specific case of searching for traps let me play it out I don't think there's a trap whether one is really there or not.

You get to play out not having not found a trap every time the GM doesn't say you found a trap

[Pelle]

In OSR games, DM rolls for the characters aren't unusual at all. Many of the original DMs did a great deal of this, and the players still managed to have enough fun that it launched an entire industry.

I prefer the OSR approach of just ditching perception skills altogether, and just giving the players information and clues directly, and let it be up to their skill to determine what they do with it and if they act upon it. For a 5e type game, I find it ok to use a hybrid approach of letting the perception type character skills determine how good clues the players get.

In games like Call of Cthulhu, however, it can be fun that the players roll and can fail to percieve something. The fact that the players do get the metegame information that something bad is up, and their characters don't know it, adds to the suspense and can improve the game.

[Democratus]

In games like Call of Cthulhu, however, it can be fun that the players roll and can fail to percieve something. The fact that the players do get the metegame information that something bad is up, and their characters don't know it, adds to the suspense and can improve the game.

Absolutely. The style of the game being run has a huge impact on the metagame.

Cthulhu is particularly interesting in that it purposefully leverages dual role of player of the player: participant (running a character) and audience (being outside the story for the sake of suspense and irony).

It also helps that your ultimate horrible fate (death or insanity) is a buy-in from the moment you roll up a character.

[Xervous]

On roll20 I have macros for secret whole party perception rolls. The intent is not to conceal the result of the roll so much as to conceal the event itself. I use calls for perception checks as a way of grabbing or directing attention, often when players get to durdling in a dungeon and the pace needs to be picked back up. If, pass or fail, something is happening RIGHT NOW that roll is in the open. Otherwise no need to disrupt flow for a non event.