Balancing a conflict resolution system favouring players picking a spread of skills and attributes in a challenge, while rewarding some over others?

Question

Introduction

I'm hoping I can draw on the Game Designer experience of this stack to help be make a decision on a dice-based conflict resolution system I'm designing for an RPG that I've been developing on and off for a few years now. I'm hoping to get good subjective answers based on either personal experience or knowledge gained from people in the industry.

I've set my mind on characters resolving conflicts in this system by what I'm calling 'engaging an obstacle' (e.g. monster, crossing a rickety bridge, hacking into a network, convincing a guard to let a street urchin go, reversing an arcane ritual...) and to do so picking three actions per turn.

These actions are special abilities with their own rules/resource requirements, or a basic 'skill action' all character with that skill can perform.

Each of the three actions will usually specify an attribute and a skill that goes towards the total dice pool, with success/failure and any special rules influencing the remaining actions on your round.

Attributes

Character's attributes range from 1 to 6, and additionally at character creation they pick one attribute to be their strength and one to be their weakness.

The attributes I've picked (the names of which aren't necessarily fixed) are split into three categories:

• Body - Brute and Finesse
• Soul - Guile and Heart
• Mind - Wits and Will

The attribute that's your strength makes skill checks easier to succeed, and the one that's your weakness give more reward/resource for success and/or failure.

As an aside, I haven't figured out how to represent the resource tokens/weakest-strongest aspect mechanically, but either reducing/increasing target numbers or that it just must be the highest/lowest attribute.

Skills

Characters skills range from 1 to 6 as well. I haven't pinned down exact skill names, but I'm expecting unlike say, Storytelling/er systems, each skill can be used with any attribute. I'll just make up skill names below, as they aren't as important as the attributes for this question.

I'm considering letting skills be considered a characters strength/weakness as well.

The idea here is make the skill and attributes more thematic aspects of a character than just numbers. A character with a strength in Brute succeeds by brute force where other character fail, whether that breaking down a door or brute forcing a keypad combinations. A character with a weakness in arcana doesn't succeed initially, but the resources for trying open up more opportunities later on.

---

Goals

My goals for this system are as as follows:

• Reward tactical choices around picking the three actions, especially the order
• Reward using a spread of skills, attributes and abilities
• Preventing players 'novaing' challenges; using their most powerful abilities early on to end the combat early (encouraging using their 'weakness')
• Making each action itself simple, and only having emergent complexity

My concern is that if one attribute/skill is easier to use (i.e. it lowers difficulty checks), or provides more rewards (failures provides resources regardless of failure) then that will be favoured - there's not spread of skills/attributes used.

What disincentivizes a player using the Talk skill on each three action with their best ability, if it makes the challenge easier (Talk around to get info, Talk to get past the governor's doorman, Talk to the Governor)? What disincentivizes the player from using their worst ability to get more resources until the last action or just building them up and unleashing them in future challenge?

I also want to hit the right level of complexity and scale, without having to force that on players or games masters in extensive rules text - such as arbitrary limits as to when special abilities/strengths/weaknesses can be used.

Attempted Solution

I was considering simply saying that repeats of skills and/or attributes are not allowed, and (somehow) getting the player to roll all their actions at once, and then secretly distributing successes between each action. That allows your to force a variety of skills and attributes be used. And any action that produces more success than required, they can spill over onto later actions.

However players would likely always use their best skill+attribute, but that could still help later actions. The issue is you then favour using your best ability on the easiest action if you know the successes needed in advance - that leads to players using their most powerful abilities first to guarantee success later on.

My goals seem opposed to each other.

How do I achieve my goals, and balance them against each other?

Answer 1

Model 1 - Traveller task chains: Some inspiration can be drawn from the task chain rules in various Traveller editions, which have some points in common with what you are trying to achieve and how. Looking at the recent Mongoose editions in particular:

• Each skill is used with the attribute relevant to the task, so if your character is trying to socialise with a bunch of gun nuts you might roll Gun Combat skill using the Social Standing attribute, if trying to remember a fact about a gun then roll Gun Combat using the Education attribute, if shooting at a target then roll Gun Combat using Dexterity. So this aspect is in common with your design.
• To promote teamwork and/or more rounded characters, difficult tasks can be broken into task chains. So, if searching a planet's surface for the McGuffin, there might be a Science (Physics) task to work out what the sensors should be tuned for, the positive or negative Effect of which would modify the success of an Astrogation roll to plot the optimal course to conduct the sensor sweeps, the Effect of which would modify the success of the final Electronics (Sensors) roll to detect the McGuffin. If task difficulties are set correctly, a character or team with skills in Physics, Astrogation and Sensors are much more likely to succeed (or succeed faster) than a character or team that is relying on character who only knows how to operate the sensors. The Traveller system does not have a fixed number of tasks in a chain, but three is quite reasonable.

Model 2 - OODA loop: The OODA loop is the process of Observe, Orient, Decide, Act. While the cycle involves four steps, there is nothing preventing two of the steps being combined as a game mechanic into the one task - I would suggest combining Observe and Orient or Orient and Decide. In a straightforward combat this could translate to a Perception check to allow the character to observe information, with the success level influencing a Tactics check to decide how to best use the information followed by the relevant Weapon skill check to see how well the devised tactics are executed. In a negotiation it might be Sense Motive followed by Psychology followed by Bluff or Diplomacy, depending on the negotiation strategy chosen. This model may not meet your needs in that the preferred sequence of actions is somewhat fixed, the player tactics are going to be more in the realm of selecting the optimal skill than the sequence in which to use them. Of course, it is up to the rules how fixed each step is - a character might choose to go with a sequence of Sense Motive to identify culture, Etiquette skill to establish rapport, then Diplomacy to close the deal, essentially not making a "Decide" task roll but making two rolls for the "Act" part of the OODA loop.

Both of the above models rely on the use of different skills at each step in the task chain - someone can't just "hack, hack, hack" in combat or "talk, talk, talk" in negotiation. The next part of the problem is how to make them use low-level skills and/or attributes.

Incentives: Why do people try new things that they aren't skilled at in real life? Often the answer is to learn. Learning a skill in real life generally involves diminishing returns. It will take a relatively short time to acquire the basics, more time to achieve competence and then vastly more time to become expert/elite. This is often reflected in roleplaying games which do not have "character levels" by making the purchase of each additional rank in a skill require more "experience points" or equivalent for higher levels. For example, purchasing a rank 1 skill may cost 2 points, then purchasing rank 2 may cost 4 points, rank 3 costs 6 points etc. Games that really want to cap skill ranks can make it exponential, eg costs of 1, 2, 4, 8, 16 etc.

Instead of (or combined with) an increasing cost to buy higher ranks in a skill, make the "experience points" (XP) received from a three-task chain inversely proportional to the rank of the lowest skill used (ie divide the base XP for the situation by the lowest skill rank used). This way you are not penalising the superb bladesman for using his rank 6 sabre skill in combat - why wouldn't he? - but you reward him if he uses his rank 1 perception skill to try to figure out his opponent's weakness first. The XP should be received regardless of success or failure - failure teaches as much or more than success.

Most players I have met are big on character development - if they will get six times more XP for using a level 1 skill with their two level 6 skills instead of using three level six skills then they will probably do it. They still get the choice - if the success or failure of the entire campaign is riding on how well a particular task chain goes, they can make the call to go all out with their best skills, as in real life. (A skilled sprinter who has just started learning karate facing multiple knife-wielding enemies should go with their strength to get out of there and save the learning experiences for later.)

Potential pitfalls:

• "Swingy" results: One observation I have made with Traveller task chains is that a good or bad roll at the start can make the second and/or third roll irrelevant - the character either can't fail (and success may have a ludicrous degree of success) or can't succeed. This can lead to lack of player engagement - it's easy to shrug off one bad roll and move on, but when you have to keep making rolls that can't succeed after the first (or second) then it leads to disengagement. The exact mechanics used may mitigate against this, but the danger of a bad = frustrating outcome is increased when you are encouraging the use of at least one low-rank skill in each chain.
• Three-trick ponies replace one-trick ponies: Characters are likely to have two good skills and one OK skill for each activity type they specialise in rather than just one skill. Personally, my preference is to encourage characters to acquire a variety of skills by repeatedly putting them in situations requiring a variety of skills rather than using rule mechanics.
• Potential player frustration: Looking at a computer-game example - I recently (albeit over a decade after most of the world!) played "Oblivion". I didn't mind the outdated graphics, but the levelling mechanics were extremely frustrating - including forcing repeated use of skills of no interest in order to level up efficiently - in the end I gave up on it. There are similarities with what you are trying to achieve, risking the creation of a game that players will not enjoy because the game mechanics force them to nobble themselves routinely or be penalised. One option may be to make cap the number of "XP-gaining" tasks in a session to be one-half or one-third of the expected total number of tasks and let players use their characters' strengths the rest of the time. Which is realistic - it's reasonable to practise using a language learnt in high school to order food while visiting another country, but reckless to use someone with such a low level as a translator in diplomatic negotiations.