Is there any good game mechanics out there for a Splinter Cell type TTRPG experience?

So my current project game is in desperate need of some decent stealth mechanics and I'm hoping there's some board members here that can help me out with it.  I'm looking for something more dynamic than D&D stealth rolls but not something I need to do massive bookkeeping to manage.

When I run stealth in my games, I either run it fast and loose or I have a very carefully devised script map that takes into account every possible player action I can think of.  I don't think either of these concepts are going to work for my project, particularly because stealth should be more of a central part of the play experience over combat.  I'm looking for something more dynamic than D&D stealth rolls but not something I need to do massive bookkeeping to manage.

My current clunky rules set I have is include a stealth and awareness stat block for NPCs and monsters. I'm playing running stealth sort of like combat with each PC having stealth hp they loose and gain based on decisions, rolls, and the actions of NPCs and monsters.  It kind of works but requires a lot of bookkeeping on my part as the GM.

So does anyone know if there's a RPG game or supplement that does stealth action well?  Is there something you think does a good job at capturing the tense feeling of almost getting caught of Splinter Cell, Dishonored, or Commandos?  Anyone have good house rules I could see?

I have found the Wrath & Glory rules for Stealth interesting and useful for gaming. You roll your stealth score (after including modifiers for environment, gear, special abilities, etc.) and your successes (it's a dice pool system) give you a stealth score. This is the score opponents need to beat with their passive awareness to detect you. The trik is that various actions taken can increase or (more commonly) decrease your stealth score. If the GM wants to keep you on your toes, they can switch the flat increases/reductions to a randomized value and keep track of your score out of players' sight so they don't necessarily know the exact moment they are detected.

Just spitballing here, but I think you're on the right track with treating stealth a bit like a combat encounter.

One of the problems with a lot of TRPG stealth is that satisfying stealth gameplay is often more about planning than execution: scoping out guard positions, preparing your route, etc. Planning requires a certain level of predictability to be rewarding, and that goes right out the window if a single bad dice roll can blow the whole plan. In a lot of RPGs, a successful infiltration is going to require a long series of dice checks, and you have to succeed on every single one of them. The laws of probability are going to make that not worthwhile, especially if that single fuckup is likely to mean death.

You could argue that's realistic, but honestly, the whole Splinter Cell/Thief/Styx style of stealth infiltration game is highly unrealistic, so I think you have to suspend your disbelief on the whole genre.

You can mitigate the dice-luck issue a lot by flipping it so that instead of having to make lots of rolls and fail none, you just have to pass most of the rolls in a sequence. That retains an element of randomness while still averaging out to more consistent results. It'd go a long way towards making stealth characters as reliably good at their specialty as combat characters are.

The fip side there is you don't want to make passing each guard as involved as a full combat encounter. I don't know, maybe you could blow up the scale so that an entire infiltration plays out like a single combat encounter, rules-wise?

I think the best stealth mechanics I've encountered recently are in a Stars Without Number supplement "Darkness Visible" which is about espionage campaigns.

Basically, you get everyone to roll their Stealth skill normally. The person with the highest Stealth skill has their skill level (Traveller-rated, so 4 is the highest) in free passes that they can dole out to the less ghostly party members. It keeps it light and fast, while also preventing one bad stealth roll from screwing the whole infiltration.

I do like the idea of a Stealth pool, where time and actions degrade the pool. I wouldn't use a HP mechanic, but I might use poker chips or something like that.

I guess I'd start by asking "What do you mean by Stealth?"


When I think of stealth, it either boils down to encounter stealth, or whole-scenario stealth. The D&D rulesets have an acceptable level of abstraction for encounter stealth, but really fall flat if you try to apply them to whole-scenario stealth.

Key principles I would suggest for designing a system for whole-scenario stealth:

* More players rolling dice should not necessarily lead to higher chance of failure
* Every character can contribute something unique to the encounter (not just "I roll Stealth")

In my mind, whole-scenario stealth is more along the lines of Ocean's Eleven or Mission Impossible. You need to define key roles for characters like face-man, technical (hacker, mage), muscle, diversion, etc. The roles need not be specific to the character but could be fluid to the scenario -- e.g. Russian character is a more effective face-man in a Moscow-based scenario, even if he's normally the strong-man who knocks out guards.

What you probably end up with is some kind of progress-clock minigame where you give players a certain number of moves to accomplish their objectives. I also like the idea of a stealth pool that automatically ticks down, which implicitly creates some tension and you can either force players to do certain things (hide, knock out a guard, make a diversion, etc) to maintain stealth, or complete objectives.

May be more abstract than what you are asking for, but I get results I like using d20-style stealth checks with a few considerations and habits thrown on top:

- Actively sneaking is about stealth. Rarely do I use opposed perception, unless the opponents are actively searching.  (Guards at their posts are not actively searching unless there is an alert.)

- Hiding is passive.  In that case, the other side is using active perception, assuming they even know to look.  If you are just sitting quietly, hiding, the other side can't find you unless they do something active.

- In my game, getting moderately competent to excellent at stealth or perception is a definite opportunity cost.  You get that because you didn't get something else.  Moderate perception costs more than moderate stealth, and so on up the competency ladder. 

- Sending out scouts is practically required.  That means that if you don't and someone blows the roll, sucks to be you.  If you do, the scouts are probably at least competent at stealth and/or perception.  Once you know where the opponents are, it's a lot easier for the incompetent sneakers to hide, find another way in, or even plan a distraction or bluff or other alternative.  Key here from a general game design perspective is that the whole group blindly sneaking through a populated area won't work even if it doesn't come down to one person blowing a roll--and it really shouldn't work that way if there is any challenge. 

- Even with the above, I had one glaring hole, which I fixed by letting "stealth" ability stack with perception when trying to perceive an ambush by stealthy foes.  That is, people good at stealth should have some boost in detecting others trying to use stealth on them.  Because it works fairly well to have the possibility of good perception but lousy stealth, but not vice versa (unless played for laughs).

- Finally, let a single roll carry through a situation unless something major changes.  A lot of distance to go or poor cover should usually be a negative modifier on a single roll (or likewise a positive modifier when the circumstances are conducive to stealth).  That "unless something major changes" does carry some weight.  Just don't get to hung up on this first leg is 50 yards and the next one is 20 feet.  If the first leg is in shadow with lots of underbrush, while the next one is timing a dash across a cleared area, then they are different enough to have another roll.  It's also a chance for the players to do the first one, see what they are up against, and then chicken out.  And if they don't back off, at least all that underbrush is a chance to sneak off successfully.

Doing all that, I'm using the straight d20 + mods, a few key abilities, and getting tense, exciting stealth scenes.

First, thanks for the feedback.  I will definitely get a hold of Wrath & Glory and Darkness Visible.

For more context, I'm making a game where the players are ninjas.  In direct combat, players are going to be a lot more squishy than many of their opponents.  I am aiming at multiple layers of covert play ranging from infiltration to target stalking.

A lot of this can be done with simple skill checks and narrative development but I am trying to develop a tactical stealth action play that is at the center of the game rather than more traditional combat. 

A lot of my thinking is in line with the kind of thing you guys are saying.  A single failed roll should not immediately result in completely blowing the player's cover but a series of mistakes and failed rolls should result in increasing difficulty and increased chances of getting caught.  Getting caught will lead to conditions that are very undesirable from blowing the mission to getting caught in mortal combat at a disadvantage.

Quote from: ForgottenF on May 12, 2024, 11:29:49 AMYou can mitigate the dice-luck issue a lot by flipping it so that instead of having to make lots of rolls and fail none, you just have to pass most of the rolls in a sequence. That retains an element of randomness while still averaging out to more consistent results. It'd go a long way towards making stealth characters as reliably good at their specialty as combat characters are.

With stealth as with other skills, what I like to do in skill-based systems (as opposed to class-based) is that if you have the skill, you're not rolling to see whether you succeed, but how well and how quickly you perform the skill.

So for example with stealth,

• fail - you move slowly and noisily, you can be mistaken for an animal moving around in the area, or a person friendly to the watcher
• success - you move quickly or quietly. If quickly, you're still noisy, but they can't get a precise bead on you. If quietly, you're quiet but it'll take you a long time to get from A to B
• great success - you move both quickly and quietly

with this in mind, only a great failure is a traditional "fail." Two of the other possible three results give the player a choice - and it may be a difficult choice. Let's say you're doing a prison break and trying to sneak by a guard - you can make noise but move quickly, or be quiet but be slow - but both you and the guard are outside the building, he's on the opposite side, if you choose to be slow he might come across you soon, but if you're noisy while he might ignore you, his mate up on the tower might notice you, and...

Indeed, if the player is willing to have their character do something slowly enough, they can always do it well - without even rolling. Got mechanics? Want to fix an overheated engine? Got a week? No worries, hold your dice, pass the cheetos. Want to do it in ten minutes? Want to do it so it won't burn out again after another hour of driving? Better roll.

Combat skills, by the way, always need to be done well and quickly. That's why you always roll for combat, and why the results are either/or with success/failure. That's combat.

You can tell that I've been using dice pool systems for a while. My intuitive response is to say that players with good rolls should accumulate excess successes which they can defer to players who roll less well, presumably letting characters good with stealth act as mentors to the un-stealthy members of the party.

Realistically, I would also include an element of quest design to this. If the party got captured and is escaping prison, they obviously have to do a whole party check, or come up with clever ways for PCs to avoid needing to make checks. But if the PCs are doing what they usually are doing and going out and doing something dangerous and must be sneaky, they really should split the party and only have the sneaky characters proceed.

Also real special forces do not expect to accomplish most missions stealthy from start to finish. Some, yes, but most SWAT actions--which are more like the RPG experience, anyways--"go loud" after being discovered, meaning that you have almost infinite time provided you aren't discovered, but once you are you get into an encounter which basically continues until you have eliminated all hostiles or completed all your objectives and left the area.

So it makes perfect sense for a squad of 5 to roll up with unsuppressed battle rifles and sneak around for 15 minutes before going coocoo for cocoa puffs on the guards completing their rounds.

The Angry GM: Tactical Infiltration Action

Quote from: The Angry GMIt's like this: if you take everything I already taught you about running a game like a True Game Master and you combine it with the crap about stealth and sneaking and perception and awareness I over-explained above — and which I previously explained as part of my speech on Problematic Actions — you'll get the secret sauce that makes stealth scenes work. That sauce is called dynamism. Good-feeling stealth moments and infiltration scenes — whether the players are doing the hiding or the seeking — are a back-and-forth game of "this is what you know; now how do you react?"

I like Savage Worlds Stealth rules. Effectively, if you're being Stealthy, you are. And unless people are *actively* looking for concealed people (GM's call) the players simply make basic skill checks as normal. Which is pretty straightfoward and "easy" in SW.

When people are actively searching, or have special abilities like Alertness, it's a contested check. Otherwise its pretty breezy. Of course if you wanna sneak up and kill someone you *always* have to make a contested roll, if you get a Raise you get "The Drop" which is effectively a very powerful Hit+Damage bonus that will likely kill most targets.

One problem with stealth is that there are a lot of different levels of stealth, and they don't always work together.

You have your ninja/sniper/scout/thief level of stealth, which is a small number of people hiding and sneaking around, and probably doing most of their movement under cover of darkness, or slowly creeping into a contested area.

But there are also stealth aspects to modern regular infantry operating with night vision.

Thirdly, stealth is also present at the level of an entire army maneuvering for tactical surprise. They're not going to sneak up on Ninja Guy, but they can sneak up on another army.

Likewise, there is a strong element of stealth in tank warfare. Tanks can and do sneak up on other tanks, even though they will rarely sneak up on alert infantry.

Then you have stealth among fighter aircraft, both actual stealth fighters and regular fighters using stealthy tactics like attacking out of the sun or flying below radar.

There is stealth in submarine warfare, and in space warfare modeled on submarine warfare.

In all of these situations, too, the non-stealthy operators can combat the stealth units, either with dedicated antistealth units like antisubmarine aircraft or satellites or spotters or scouts, or with tactics like sweeps of areas and skirmish lines, and artillery strikes on sniper positions, or even just good intelligence work.

Making a single system that can model all of these things seems like it would be difficult. And you might not even want to. If your characters are Army infantry, you don't need submarine warfare rules. But to get to the point where you can simplify your rules set, you need to make good campaign design choices. 

Lastly, yes, I am aware that I overthink things.

Quote from: Lurkndog on May 13, 2024, 11:19:00 PMOne problem with stealth is that there are a lot of different levels of stealth, and they don't always work together.

You have your ninja/sniper/scout/thief level of stealth, which is a small number of people hiding and sneaking around, and probably doing most of their movement under cover of darkness, or slowly creeping into a contested area.

But there are also stealth aspects to modern regular infantry operating with night vision.

Thirdly, stealth is also present at the level of an entire army maneuvering for tactical surprise. They're not going to sneak up on Ninja Guy, but they can sneak up on another army.

Likewise, there is a strong element of stealth in tank warfare. Tanks can and do sneak up on other tanks, even though they will rarely sneak up on alert infantry.

Then you have stealth among fighter aircraft, both actual stealth fighters and regular fighters using stealthy tactics like attacking out of the sun or flying below radar.

There is stealth in submarine warfare, and in space warfare modeled on submarine warfare.

In all of these situations, too, the non-stealthy operators can combat the stealth units, either with dedicated antistealth units like antisubmarine aircraft or satellites or spotters or scouts, or with tactics like sweeps of areas and skirmish lines, and artillery strikes on sniper positions, or even just good intelligence work.

Making a single system that can model all of these things seems like it would be difficult. And you might not even want to. If your characters are Army infantry, you don't need submarine warfare rules. But to get to the point where you can simplify your rules set, you need to make good campaign design choices. 

Lastly, yes, I am aware that I overthink things.

Good post. Thank you.