Not sure if that's the right phrase but what I'm asking is do PCs generally make rolls despite being the subject or object.

For example if my PCs are trying to sneak they roll stealth.

If an adversary is trying to sneak by them does the adversary roll stealth or do the PCs roll perception?

Specifically I'm wondering where in the rulebook this is discussed.

Thanks for any help.

I normally don't answer questions like this because I am the n00biest n00b to ever n00b... but we've actually had a few threads on this topic so now I are a ECKSPURT!!!

There's no right answer! It depends on your GM's style. Some groups play the PC centric way all the time. Others will make some checks secretly if the PC's shouldn't know the results or should only know them if they succeed like when an NPC tries to sneak on them. Some GM's don't like secret rolls because they want to give the players the opportunity to contribute to the story. Other groups prefer that there's some suspense for the players, too.

For the most part, it's just a matter of personal preference. There isn't a hard rule.

It should be noted, though, that the mechanics system is "actor-biased". Roughly speaking, the "good dice" are more positive than the "bad dice" are negative - the two sides are not mathematically equal. Thus, whichever side is rolling the dice in a conflict is given a statistical advantage. Whichever side is being rolling against is at a statistical disadvantage.

So if an enemy attempts to sneak by the PCs, the PCs are more likely to spot him if you roll their Perception against his Stealth and less likely to spot him if you roll his Stealth against their Perception.

I think this system tries to shy away from bajillion opposed roles like other systems, but sometimes its necessary.

If you dont want to make an opposed role, just try to determine and assign a set difficulty based on the situation; remember to include options for setback or bonus dice.

So maybe it was really easy for the ambushers to pass their [AVERAGE] stealth check, but the victims made a successful [HARD] perception check that mitigates some of the advantage the stealthers got.

Except for combat, I generally want the players to do the rolling, so in the OP's case they'd roll their Perception against the NPC's Stealth. Yes, the positive dice are slightly favoured, if that's bothersome it's easy to increase the difficulty or add setback dice.

I think, as others have said, that it's a matter of personal choice. Nonetheless, I use some patern to determine the course of rolling:

If both actors (like the PCs and NPCs) are being active with their skills - actively searching for something and the other sneaking around - then I would call for a competitive check (even though, as per RAW, you only do this when both actors want the same thing, same objective).

If one is active and the other is passive, like a guard just fulfilling his service time , then I would call for an opposed roll, being the active character the one rolling the test.

Not sure if that's the right phrase but what I'm asking is do PCs generally make rolls despite being the subject or object.

For example if my PCs are trying to sneak they roll stealth.

If an adversary is trying to sneak by them does the adversary roll stealth or do the PCs roll perception?

Specifically I'm wondering where in the rulebook this is discussed.

Thanks for any help.

If it isn't in a scene then don't be afraid to skip the dice and just have what you want happen. It's faster and doesn't ruin your plans with randomness. Also having the PCs suddenly make a perception check for no apparant reason sort of gives the game away. If it is during a scene then the character doing the action would normally get the roll, although it might be opposed in which case both are rolling.

Letting the players roll is a good idea most of the time. It's what I try to go with. Some players might not be used to it and always expect danger, madness and craziness when ask for checks... In this case I stop asking for rolls and let them ask when to roll, this has varied success, even if I let them know I won't ask for rolls even when they might benefit from it.

In certain situations though I prefer to roll, sometimes secretly even. I prefer to keep this to a minimum, but sometimes its for the best I think.

Sometimes I want my NPCs to have a more surprising effect that will be decreased if they are asked to roll and therefore expect trouble.

Thanks all. It sounds like this is not really covered in the rules so I really appreciate everyone's suggestions.

I keep the rolls to a minimum. Who makes the roll is important for both probability and who has the potential for despair so that's why I ask.

The style of our game is very open world and very cooperative story so my PCs can handle making a random roll and RPing what their characters really know instead of metagaming.

Think I'll keep it PC centric.

In certain situations though I prefer to roll, sometimes secretly even. I prefer to keep this to a minimum, but sometimes its for the best I think.

Sometimes I want my NPCs to have a more surprising effect that will be decreased if they are asked to roll and therefore expect trouble.

I've moved entirely away from secret rolls. If the rolls are secret, why bother? Just do what you want. That said, the "NPC surprise" issue is valid. The way I've gotten around that is to have the PCs roll for no reason…well, it's not really "no reason", it has a whole bunch of useful outcomes: on a success I can say "you're absolutely sure there is no one around, go ahead with your sneaky activity"; on advantages, maybe it's an opportunity to breath a little easier and reduce some strain across the party; on threats, maybe it's a way to ratchet up the tension with a little strain; on despair, well, there wasn't anyone there before, but there is now!; on triumph…you get the picture.

But the main point is if you mix things up a bit, the NPC surprise problem kind of goes away, since they never know what's up. I can hear the alarm bells ringing in the ears of the "fewer rolls" crowd, and that's valid too, but I've only had to do this maybe once per session to scramble player preconceptions.

Fair point. Still, as I said, they do most of the rolling and report to me the results.

As for rolling secret = why bother? I do not agree with this, honesty and a social contract about having the most fun is at the core of any roleplaying game, this included.

Your suggestions are good though, but too much rolling annoys some of my players - particularly the new ones, and one of my players has a supernatural ability to always fail... when we played d20 she always fumbled. In her 15 years of gaming she's had one (1) critical hit in any game all together. Her luck changed slightly when we started EotE, but now the curse has gotten used to these funky new dice and she's back to failures, threats and despair. It's like magic. She's a statistical outlier, she rolls a d6 10 times and the average is 2, and it stays there, we've tested.

Anyway, I'll remember some of your thoughts, might work in the long run.

When possible I like the PC's to role. Usually if somebody is trying to sneak by I'll have them roll perception, with the difficulty being the stealth skill of the sneaker, modified by environmental circumstances. Occasionally I'll do the rolling for one reason or another, but I can't think of one off the top of my head.

I wanted to chime in just to say that, from a purely statistical perspective, you aren't necessarily doing your players any favors by handing them the dice. The odds of an evenly matched set of dice is pretty close to 50/50 with a slight tilt (1-5%) toward failure in most situations. The difference is more glaring in 1v1 or 2v2 scenarios, but those ought to be rare once the characters have developed some.

That said, I agree with what's been said here already. Nobody wants to sit around and watch the GM play with his dice. Player-centric rolling is a good policy, even if it doesn't afford a statistical advantage.

For my group we determine who rolls by who's the active party. So if a PC says they wants to search a room for a hidden opponent then they roll Perception vs. Stealth. If the NPC is trying to hide and the PC doesn't even know to look or doesn't ask to look then it's me who rolls Sealth vs. Vigilance, and depending on the situation I'll roll this behind the screen without mentioning it and only revealing the results if there relevent (a failure etc.)

I also have interaction rolls (Coerce, Charm, Negotiate, etc.) made by the PC whose Player initiates the action, even if their not the PC with the highest skill. So if player A says "we'll tell the merchant that we're representing Boo Boo the Hutt to try and get a better price" then they roll even if their Negotiate is 1, but if Player A says to Player B, who has Negotiate 3 "Tell the merchant that we're representing Boo Boo the Hutt" the Player B's PC rolls.

I do this to help Player's stay in character, plus it can make for some entertaining outcomes

I like PC-centric rolling.

The story is about the PCs and it reflects that. Do they see the baddie trying to sneak up on them? Do they resist his/her attempts to intimidate them? Etc.

Also, if everything goes wrong, they can't blame you, they rolled the dice

I mostly let my players roll. Only when the NPC(s) are clearly the ones using an active skill vs. the players' passive one do I make rolls myself (outside combat, of course). And the dice are tilted ever so slightly in favour of the active character - the number of successes and failures are the same on all dice sets, but the greens, blues and yellows all have more Advantage symbols than their counterparts.

As for secret rolls, I only do these when I don't want to clue the PCs in on what the NPCs are doing. If I gather up a bunch of dice and ask whose Perception is higher, they know someone is sneaking around. My players are very good at not metagaming, but still. It's also useful if I don't want them to know exactly how good an NPC is at any particular thing.

the number of successes and failures are the same on all dice sets, but the greens, blues and yellows all have more Advantage symbols than their counterparts.

That's incorrect. Here are the expected outcomes of the different dice:

The Boost die averages 0.333 successes and 0.666 advantage.

The Setback die averages 0.333 failures and 0.333 threat.

The Ability die averages 0.625 successes and 0.625 advantage.

The Difficulity die averages 0.500 failures and 0.750 threat.

The Proficiency die averages 0.833 successes, 0.666 advantage, and 0.083 triumph effects.

The Challenge die averages 0.750 failures, 0.666 threat, and 0.083 despair effects.

So Ability dice have more successes than Difficulty dice have failures and Proficiency dice have more successes than Challenge dice have failures. Also Ability dice have fewer advantage than Difficulty dice have threat.

the number of successes and failures are the same on all dice sets, but the greens, blues and yellows all have more Advantage symbols than their counterparts.

That's incorrect. Here are the expected outcomes of the different dice:

The Boost die averages 0.333 successes and 0.666 advantage.

The Setback die averages 0.333 failures and 0.333 threat.

The Ability die averages 0.625 successes and 0.625 advantage.

The Difficulity die averages 0.500 failures and 0.750 threat.

The Proficiency die averages 0.833 successes, 0.666 advantage, and 0.083 triumph effects.

The Challenge die averages 0.750 failures, 0.666 threat, and 0.083 despair effects.

So Ability dice have more successes than Difficulty dice have failures and Proficiency dice have more successes than Challenge dice have failures. Also Ability dice have fewer advantage than Difficulty dice have threat.

Except that you have to overcome all the Failures, plus one! If all the successes and failures cancel each other out, you still fail. I'm no good at probability, but that's a pretty big damper on there, especially with a small number of dice overall.

the number of successes and failures are the same on all dice sets, but the greens, blues and yellows all have more Advantage symbols than their counterparts.

That's incorrect. Here are the expected outcomes of the different dice:

The Boost die averages 0.333 successes and 0.666 advantage.

The Setback die averages 0.333 failures and 0.333 threat.

The Ability die averages 0.625 successes and 0.625 advantage.

The Difficulity die averages 0.500 failures and 0.750 threat.

The Proficiency die averages 0.833 successes, 0.666 advantage, and 0.083 triumph effects.

The Challenge die averages 0.750 failures, 0.666 threat, and 0.083 despair effects.

So Ability dice have more successes than Difficulty dice have failures and Proficiency dice have more successes than Challenge dice have failures. Also Ability dice have fewer advantage than Difficulty dice have threat.

Except that you have to overcome all the Failures, plus one! If all the successes and failures cancel each other out, you still fail. I'm no good at probability, but that's a pretty big damper on there, especially with a small number of dice overall.

Exactly, the success-side dice need those extra successes just to break even. That's why the chance for success on 1v1 and 2v2 rolls are something like 35% and 40%, respectively. After that, it starts to approach 50/50 with a slight tilt either way, depending on the mix of d8s and d12s.

This is all, of course, assuming that the dice are evenly matched, which rarely happens in a session.

It depends on how important the NPC doing the sneaking is, or how important the action being taken is to the central plot. If it's an important plot character or a nemesis that you have actually created stats for - absolutely use that NPC's stat to it's full advantage against their perception. In fact, I would even treat that Nemesis as an actual PC that you the GM play.

If you don't already have stats for something already made, and in the process of playing, the narrative just happened to make an arbitrary saboteur try to stealth by with a some thermite for the party's ship - just have the PC's roll Perception and apply as many Difficulty Dice as you deem necessary - 2 Average, 3 Hard, etc - and then apply Bonuses and Setbacks accordingly. No need in making things harder on yourself than they need to be.