I’m running an Amber game for the household, and I’m using the following system. It’s a super rough writeup, but I wanted to get the draft version up somewhere.  There’s a very minor power system I’m not getting into right now, but the heart of it revolves around a iceless system where you roll dice after success to determine the position the success leaves you in.

Core Stats

The 4 stats^[1] are:
Fire (Spades) – Covers actions relying on physical speed & nimbleness.

Air (Diamonds) – Covers actions relying on intellectual horsepower or pure perception.

Water (Hearts – Covers actions of personal presence or understanding what is known.

Earth (Clubs) – Covers actions of physical strength or toughness.

When taking an action, if it falls under one of these stats, the character gets the stat as a bonus. If two stats might apply in different ways (such as using clubs or spades to hit someone with a stick), use the one that aligns with the player’s description of the action. When two or more stats both seem applicable (such as spades and diamonds in a question of how quickly you notice something) use the lower of the two. 

Characters start with 2 points in each stat, and have 12 more points to spend. This number is the default for an Amber like game, but might be adjusted up or down. 

Descriptors

Characters have 3 descriptors. They’re freeform, so go nuts. If a descriptor is applicable to a situation, then you get a +3 in that action. If a second descriptor is applicable, the bonus drops by 1, so you get +5 (that is, +3 and +2). If all 3 are applicable, then it drops again, and it’s +6 (+3, +2 and +1).

NPCs

NPCs have descriptors, but very rarely have stats. However, they have another mechanic to give them a little variability – Tiers.

NPCs can have a tier from 1 to 5, and the tier represents:

1. How many descriptors they can have
2. How big their starting descriptor bonus is. 

So, a Tier 4 NPC might be a, I dunno, Brave, Bold, Knight, General, and in a fight he’s going to be running around with a value of 10 (4 + 3 + 2 +1) so he’s KIND of a badass. 

Tier	Represents
0	Inexpensive furniture. Lunchmeat. Very small rocks.
1	The woefully inadequate. Children. Exceptionally fierce squirrels.
2	Most People
3	Heroes, adventurers, general badasses
4	Demigods, paragons, Isekai protagonists
5	Dragons, Gods, all that jazz.

As a double cheat, the GM is not obliged to sketch out all of an NPC’s descriptors, and they’re free to re-use them for simplicity. Which is to say, you don’t need to flesh out that Tier 3 Bandit. Just give him a 6 on banditing. That’s what he’s there for. IMPORTANT: This is just a trick for sketching out nameless characters. Anyone important enough to have a name merits a little more attention to detail, especially because for those characters, it can start mattering what descriptors they have or don’t have. 

Resolution

Higher number wins.

I mean, yes, sometimes there’s no an opposing number, in which case the number is “what kind of person could successfully do this?”, and then higher number wins. If you need guidelines, consider that a tier 2 human doing the thing they’re best at has a big old 3, so 3 is a pretty good default. 5 is good for something pretty hard. 9 is pretty much best in the world sort of stuff. 

Couple rules of thumb:

• Set the difficulties as if they were for NPCs, and that allows the characters to benefit from their stats, which is what they’re for.
• There’s a whole thing here about the role of narration in resolution. If players are clever or use tactics or find other ways to shift the situation, you can very reasonable make up a few points difference. This is, however, very subjective and that’s deliberate. If you don’t want it to be subjective, you probably want more numbers. 
• Tools done provide bonuses, they just change the difficulty. A cliff might be really hard (5) to climb on its own, but much more doable (3) with rope. The same logic applies to things like taking time, having help and all the other ways that people tackle problems. 
• For all that I’m committing page inches to them, static difficulties are largely boring. Even if players can’t immediately overcome such a challenge, a little extra effort will usually be enough, so try to assume success, and just fold in things like extra time.
• Opposition, on the other hand, tends to be a little more interesting, at least it can be. 
• MOST of the time, there’s no actual “resolution”. You know what the character’s numbers are, and you are going to stop to trigger the system when the logic of the narrative is obvious.

BUT WHAT IF IT’S INTERESTING?

When it’s interesting, or there are uncertainties, the GM can call for a roll. Critically, you only call for a roll when the character is going to succeed. If not, stick to narration and description. But when success is certain, but other questions are open, that’s the time to call for a roll. Possibilities include:

• Going into town to ask around about a person
• Crossing swords with a notable opponent
• Breaking into a well guarded villa
• Relying on magic for…well, most things.

The purpose of the roll is to see how the situation unfolded – has it revealed new opportunities, or has it sprung unforseen complications?

A roll is made with 2df, so the potential results range from -2 to +2, and the results are interpreted as follows:

Roll	Means
+2	Everything’s Coming Up Milhouse. A happy coincidence introduces a major opportunity.
+1	A lucky break. Things go well, and a minor opportunity is revealed.
0	Business as usual. The scene plays out by the numbers.
-1	Bad Break. Minor complication
-2	Oh @$^%! Major complication.

OPPORTUNITIES & COMPLICATIONS

Opportunities & complications are both elements that move the scene forward – they either introduce a question that the player must answer (like: The building is on fire, what do you do?) or an opportunity for action that had not previously been available (like: You spot a secret door, do you want to go through it?). 

Opportunities are generally easier to adjudicate. A minor opportunity might take the form of a little bit of extra information, a friendlier reception, a nice tip or the like. A major opportunity is a full on lucky break – run across a friend, discover a significant piece of information or generally be in the right place at the right time. 

Complications are trickier. They can suck, but it’s important that they not negate the underlying success. If the complication offers a choice, the choices should not include (effectively) retroactively failing.

As an example, the character sneaks into a castle to steal the Maltese Frankfurter. They have the skills to succeed, but there are a lot of variables in this, so the GM calls for a roll, and the player gets a -2. 

The GM’s first instinct is “You grab the Frankfurter, but then every alarm in the place goes off. You have only moments to flee ahead of the guards”.

Now, that might be a fun scene, but it raises the question of what the player was trying to accomplish. If their goal was GETTING the Frankenfurter, this is probably fine. But if their goal was passing undetected, this pretty much flips the bird to that. For that character, the twist emerging after they get out might be more appropriate. 

A minor complication sours the result a bit. You succeeded, but…perhaps you took a minor injury, or the prize had a catch, or you didn’t end up right where you hoped. A minor complication tends to not be about a choice but is instead just a little extra badness that was outside of the character’s control.

A major complication take one of two forms. First, they might be a consequence – an injury, a loss of resource or some other complication they need to proceed in the face of. These are fine, and if you can think of a good one, they’re cool. Otherwise though, consider a force: the complication create a situation the character must respond to, such as an imminent threat, and asks “What now?”

Options:

• Roll for all the good stuff – If a scene is particularly cool, such as a duel with a nemesis, the outcome might be sure to be failure, but it might be fun to allow them to throw in a die roll to see where it goes. 
• Bad Stuff – If using this rule, characters are allowed to take one more descriptor (though they can still only use 3 of them on a given action). This descriptor is flagged as “Bad stuff”, and in any situation where the character actively uses it, a roll is appropriate, and if the roll take a bad turn, the bad stuff descriptor is almost always the problem.

Anaway, as I said, it’s a work in progress, but I want to capture the bones of it here.

1 – I could not tell you how many variants on these 4 stats I have used throughout the years.