D&D 5E Table practices for handling skills in 5e?

[Quickleaf]

I've come to believe that treating skills as operating in the same way with the same rules is not the best design because it creates a mismatch between the mechanics and actual flow at the table. This is a strong belief and my question is coming from this premise. If you prefer to question the premise or defend 5e's skill system, or you prefer to rag on 5e's skill system for a whole different reason... this is not a plus thread, so go for it - we can learn from each other.

However, the conversation I am interested in having is about how you handle one skill differently from another in your game.

For example, I've learned that not having players roll Stealth immediately, and instead waiting until there's a chance of being detected to have them roll Stealth, leads to a much smoother play experience both in terms of rules handling and in terms of immersing players in the suspense of the scene. IMHO it's an objectively better way to run Stealth that is different from other skills where there isn't that latency/delay between a player declaring an action (with risk) and resolution through dice. The rules don't explain to do this. It's a "table practice" that varies from group to group.

I'm not attached to which particular list of skills you have in mind, whether d20, 4e, or 5e, or if you have thoughts about tool proficiencies, but for conversation sake, here's the list of 5e skills as a touchstone / reminder...

Spoiler: List of skills

Athletics
Acrobatics
Sleight of Hand
Stealth
Arcana
History
Investigation
Nature
Religion
Animal Handling
Insight
Medicine
Perception
Survival
Deception
Intimidation
Performance
Persuasion

EDIT: I'm including this graphic for clarity of communication, since the conversation has been focusing on X-axis and Y-axis questions. Which is ok. I'm just interested in having more conversation around the Z-axis, especially with folks who are engaging in that axis.

 

[Stormonu]

I miss the days of having rolls like Stealth being rolled behind the DM's screen, I think my players would revolt if I tried to enforce that in this day and age. But, it would get rid of the "he rolled a 5, can I now try?" (which comes up for Perception moreso than anything else)

I'm fine with the player rolling at the time of declaration/start of the activity for things like Stealth, I can hold the result to compare it against events further down the road. Waiting until a moment of contact on a Stealth check just isn't my style.

Also, side note - I wish "History" had been named "Lore" instead - that's a bit more broad, and History could be a subcategory of Lore. Overall I wish there was more teeth to the skill system, with things like Expertise only working for a subcategory of a skill. For example, if a Rogue took Expertise in Acrobatics, they might take it in Acrobatics (Tumbling) or Acrobatics (Parkour). A Wizard might take Expertise in Arcana as Arcana (Spellcasting) or Arcana (Potioncraft).

 

[Shardstone]

I ended up removing skills in favor of Ability Score Prof.

 

[Swarmkeeper]

For example, I've learned that not having players roll Stealth immediately, and instead waiting until there's a chance of being detected to have them roll Stealth, leads to a much smoother play experience both in terms of rules handling and in terms of immersing players in the suspense of the scene. IMHO it's an objectively better way to run Stealth that is different from other skills where there isn't that latency/delay between a player declaring an action (with risk) and resolution through dice. The rules don't explain to do this. It's a "table practice" that varies from group to group.

Hmm... I actually think you are running Stealth as intended. The advice in the DMG (p. 237) is to only call for a roll if the outcome is in doubt and if there is a meaningful consequence for failure. If there is no uncertainty, the character can automatically succeed at their declared task - in this case, moving stealthily to avoid detection when there is no one around/able/caring to detect them. If they continue on the path and now there are some lookouts in range, well, sure, have them roll Dex(Stealth) against the lookouts' Wis(Perception). If the PC fails, they are detected and... consequences.

The other skills can be handled the same way: the DM should only call for an ability check roll when there is uncertainty in an outcome and a meaningful consequence for failure. Otherwise, auto-succeed (or auto-fail as the case may be) and move on.

Notably, this obviates the need for "secret" ability check rolls behind the DM screen as the dice only come out when the criteria are met.

TL;DR: we don't treat the rolling of different skills differently at our 5e table.

 

[Clint_L]

Skill checks are called for as needed, or when the players want to use one, and are mostly done publicly. Up to two players can roll, or one player with advantage if someone else is assisting and qualified to assist. Guidance, bardic inspiration, etc. have to be announced before the roll, not after.

I usually make my rolls publicly too, except in cases where the characters wouldn't immediately know if they are successful or not. For example, if you are trying to deceive the guard, you roll your deception check and I secretly make his insight roll if he has any reason to be suspicious.

For a lot of rolls, I will allow different options - various lore skills might be applicable to a piece of knowledge, acrobatics or athletics to climb a tree, etc. When in doubt, I try to give players the benefit.

 

[Quickleaf]

I miss the days of having rolls like Stealth being rolled behind the DM's screen, I think my players would revolt if I tried to enforce that in this day and age. But, it would get rid of the "he rolled a 5, can I now try?" (which comes up for Perception moreso than anything else)

I'm fine with the player rolling at the time of declaration/start of the activity for things like Stealth, I can hold the result to compare it against events further down the road. Waiting until a moment of contact on a Stealth check just isn't my style.

I wouldn't have thought much of the idea of rolling Stealth more like a saving throw until I tried it – it worked great for the players I tried it with. Tastes differ, but it actually preserves the mystery more like what you miss from the days of rolling behind the DM screen!

Edit: I've also found that reworking Perception to function more as a reaction to threat (i.e. saving throw) to avoid traps or ambushes at the last second vastly improved game flow in the sessions where I've been able to try it out.

Also, side note - I wish "History" had been named "Lore" instead - that's a bit more broad, and History could be a subcategory of Lore. Overall I wish there was more teeth to the skill system, with things like Expertise only working for a subcategory of a skill. For example, if a Rogue took Expertise in Acrobatics, they might take it in Acrobatics (Tumbling) or Acrobatics (Parkour). A Wizard might take Expertise in Arcana as Arcana (Spellcasting) or Arcana (Potioncraft).

All good ideas in my book. Do you handle Lore skills the same as other skills at your table? Roll to meet a DC, then GM spouts some lore, and player says something "ok yeah, I'll share that with the party"? I ask, because I find that immensely unsatisfying at the table.

Hmm... I actually think you are running Stealth as intended. The advice in the DMG (p. 237) is to only call for a roll if the outcome is in doubt and if there is a meaningful consequence for failure. If there is no uncertainty, the character can automatically succeed at their declared task - in this case, moving stealthily to avoid detection when there is no one around/able/caring to detect them. If they continue on the path and now there are some lookouts in range, well, sure, have them roll Dex(Stealth) against the lookouts' Wis(Perception). If the PC fails, they are detected and... consequences.

The other skills can be handled the same way: the DM should only call for an ability check roll when there is uncertainty in an outcome and a meaningful consequence for failure. Otherwise, auto-succeed (or auto-fail as the case may be) and move on.

Notably, this obviates the need for "secret" ability check rolls behind the DM screen as the dice only come out when the criteria are met.

TL;DR: we don't treat the rolling of different skills differently at our 5e table.

That may be, but in actual play? I NEVER saw anyone using Stealth more like a saving throw in 5e or any other edition. It was always the player declared "I'll try to hide behind the standing stone", and boom the roll was immediate. I get your point, but my counterpoint is regardless of what the rules say or do not say about rolling procedure in regards to skills, switching it up this way with Stealth is a very different way from handling sneaking rolls I've ever seen at the table.

Skill checks are called for as needed, or when the players want to use one, and are mostly done publicly. Up to two players can roll, or one player with advantage if someone else is assisting and qualified to assist. Guidance, bardic inspiration, etc. have to be announced before the roll, not after.

I usually make my rolls publicly too, except in cases where the characters wouldn't immediately know if they are successful or not. For example, if you are trying to deceive the guard, you roll your deception check and I secretly make his insight roll if he has any reason to be suspicious.

For a lot of rolls, I will allow different options - various lore skills might be applicable to a piece of knowledge, acrobatics or athletics to climb a tree, etc. When in doubt, I try to give players the benefit.

With your example of deceiving a guard... how has that played out at your table? Did the player say "I want to trick the guard into letting us pass by claiming we're the rotation duty?" and you said "sure, roll Deception" player rolled, succeeded, and you narrated happened? Did it begin with the player speaking in-character to the guard, then at a certain point in the conversation, you called for a roll, with the DC set irrespective of what the player said leading up to that?

What I'm driving at is the uniqueness of social skill checks compared to other skills, at least from what I've observed.

 

[Clint_L]

With your example of deceiving a guard... how has that played out at your table? Did the player say "I want to trick the guard into letting us pass by claiming we're the rotation duty?" and you said "sure, roll Deception" player rolled, succeeded, and you narrated happened? Did it begin with the player speaking in-character to the guard, then at a certain point in the conversation, you called for a roll, with the DC set irrespective of what the player said leading up to that?

What I'm driving at is the uniqueness of social skill checks compared to other skills, at least from what I've observed.

It sort of depends on the context. Did the players come up with a great plan where they stole uniforms and are at pains to fit in? Is the place on high alert? If the guard has no strong reason to doubt them, it'll be a low DC (usually I like to announce the DC beforehand, but in a case like that I would keep it secret). On the other hand, if the PCs are acting sketchy I am probably going to make it a contested check.

Contested skill checks are fun, so I use them when I can. But reading a book and looking for a piece of lore wouldn't be contested - it's a one-sided effort. Opening a lock isn't normally contested, either. In both those cases, I will announce the DC in advance: "you can see right away that this is a very complicated lock, and the DC will be 20."

Some social interactions aren't contested - making an entertainment check, for example, is usually not.

 

[payn]

For example, I've learned that not having players roll Stealth immediately, and instead waiting until there's a chance of being detected to have them roll Stealth, leads to a much smoother play experience both in terms of rules handling and in terms of immersing players in the suspense of the scene. IMHO it's an objectively better way to run Stealth that is different from other skills where there isn't that latency/delay between a player declaring an action (with risk) and resolution through dice. The rules don't explain to do this. It's a "table practice" that varies from group to group.

I like this, yoink.

I have not run a 5E game yet, but I do try things like this out. When I played in a campaign it was usually a call for anybody with a particular skill, or retroactive kind of thing. Didn't come up much, but I also felt like the GM was on autopilot much of it.

I am a big fan of skill systems so when I run they matter. I do leave a lot int he hands of players to try and make a narrative case for skill use. 5E is pretty poor in variety for wing it reasons in this department. Which, is likely one of the few reasons the itch to run a 5E game hasnt become strong enough.

 

[Umbran]

I typically allow some stat substitutions on Persuasion rolls. Charisma is a good standard, but I will allow substituting Intelligence if the player can outline an argument based on logic and evidence, or Wisdom when they can base their presentation on intuition, emotional cues, or empathy.

 

[DEFCON 1]

I use the Alternative Ability Score variant rule first of all-- just wanted to start with that. I also tend to assume every character succeeds on what they are trying to do to a certain extent... and when I ask for ability checks (plus applicable skill) it usually tends to be for gaining additional information or success on top of what they've already gotten.

Now that will seem odd to some people when I say that... that the players are successfully getting stuff without needing to make checks... but here's my way of thinking from my perspective: I give out information to the players freely all the time-- all part of my descriptions. I describe the situation in front of them, and just by doing that the PCs have ostensibly "succeeded" on any number of invisible checks-- ones that didn't require rolls.

I describe an altar in a chamber. How do they know it is an 'altar' and why did I call it that? Because they succeeded whatever invisible Religion check would have been necessary to know it was an altar, and I just handed that info to the players free of charge. And now that the first level of info is out of the way... they are free to ask if there is any more info to be gained through any number of other checks-- and depending on the quality of the check I will give out additional info. And what this does is that it gets ME out of the mindset of saying "Oh well, you rolled a '2', you failed"... but instead thinking of it as "You gained no additional information over what you had already successful gained just by me describing the situation to you." In other words, the PC were already great... they just weren't superb in this particular instance.

And that way of thinking-- no PC ever truly "fails", they just don't succeed further than the baseline foundation of success that they got through how the scene has been presented... changes our player and DM dynamic. My players tend to be more interested in trying things because they know I'm not going to "enjoy it" when they roll poorly as though it's a Me vs Them situation and I get to cackle when they "screw up"... but rather that they know I WANT to give them more help and success, but will only do so if their PCs roll well enough to get it.