Wednesday, February 22, 2023

Overview: Skills (2023)

Today I want to do a comprehensive(-ish) look at the skills system.

Skills in The Next Project essentially come in two flavours: Core skills, and Knowledge skills. In either case, making a skill check always involves rolling at least 1d10 and 1d6, and adding the results together. You can increase your odds of success by gaining training (i.e. advantage) and/or mastery on the d10 roll, or by increasing your "ranks" with that skill -- allowing you to roll additional d6's, but only adding the highest.

In the latest paradigm, the d10 roll is considered to be the actual check, while the d6 is considered to be a bonus (albeit "always on") to the check, known as your skill rank bonus.

Core skills (including skills which can be used in combat) fall under both an Attribute and a Skillset. If you have training with both the attribute and the skillset associated with a given skill, you gain mastery on that skill -- which applies only to the check itself, not the skill rank bonus. This is a change from earlier ideas, where "skill mastery" would allow you to reroll 1s on any dice rolled as part of the check; it turns out that applying mastery to the d10 roll gives about the same odds, so having a separate mastery mechanic for skills was deemed unnecessary.

The idea behind using attributes and skillsets sort of comes from the ethos of using a given skill proficiency with a different ability modifier than the default; the classic example is adding Strength instead of Charisma, when using the Intimidate skill. (This showed up later on in 4e but is a bit more common in 5e, particularly when it comes to the tool proficiencies.) In a weird way, what ends up happening is that the idea reverse-engineers itself, to the point where the intersection of a skill with a different ability modifier creates (in fact) an entirely different skill.

The other influence on skillsets was the idea of taking class skills and being able to level them up as a group, without being dependent on ability modifiers. For example, in 4e pretty much every Fight-man class had Athletics (STR), Endurance (CON), Heal (WIS), and Intimidate (CHA) as part of their list of class skills. So what if we took all of those skills, and put them together in a skillset, and said, "you can use your bonus to this skillset in place of the ability modifier, for this skill"? That's essentially what the idea is with core skills in TNP. So, instead of needing to be equally good at STR, CON, WIS, and CHA in order to be equally good at all of the Fight-man skills, you instead need only be good at the skillset. This also sort of covers the design space of "skill synergies" as it existed in 3.5 D&D.

By structuring core skills this way, you're actually able to fit more skills into the "grid" -- as long as a skill can be described by the intersection of one attribute and one skillset, it can exist within the system. This works because you're always levelling up skills as a group, rather than having the overhead that comes with trying to manage 30+ distinct skills.

Another name for a check using a core skill is an "ACIDS test" which comes from the acronym for both the attributes and the skillsets:

Attributes (usually listed in reverse order):

  • Agility
  • Charisma
  • Intelligence
  • Dexterity
  • Strength

Skillsets:

  • Athletics
  • Communication
  • Infiltration
  • Detection
  • Subtlety

In general, at the start of character creation, each class gets training with some combination of two attributes and/or skillsets; this is one of the first areas where class categories start to make a distinction. "Skill Expert" classes will gain training with another attribute or skillset at 1st level; there is also a feat to gain training with an attribute or skillset of your choice, but only "Skill Expert" or "Jack of All Trades" classes may take this feat.

Now, knowledge skills are not tied to attributes or skillsets, but instead are linked to power sources. Generally speaking, each class will get to pick one power source at character creation, and either upgrade it or add another one at later levels -- this varies a bit by class category. When you gain a power source, you have the option to gain training with two of its associated skills, or training with only one but mastery with all three; upgrading a power source gives training and mastery with all three of its associated skills. As mentioned in an earlier post, each power source will have one unique skill, and two skills which are shared with other power sources. Ranks in knowledge skills will be gained as part of a universal progression, but can also come from feats or as a class category feature.


It's important to understand what the math (roughly) looks like for skills. A straight 1d10+1d6 against a DC (or "target number") of 10, is a 45% chance to succeed; this is the same rate as rolling 1d20(-2) against a DC10. Using a trained skill boosts your chances of success to 66.83% (an increase of just over 20%) which is comparable to 1d20+2, at 65%

Adding a 2nd rank to either a trained or an untrained skill increases the chances of success by approximately 10%; additional ranks produce diminishing returns, about 5% for the 3rd rank, and so on. This is why it makes sense to level up core skills by being able to add only a single rank to the attribute and to the skillset. Adding mastery to an untrained skill gives it a flat 10% boost, whereas it's a little bit less for a trained skill.

One of the things I wanted to keep (while still adding more skill gradations) was "a meaningful chance of failure." With that in mind, a maxed out skill (advantage, mastery, and 3 ranks) caps out at a success rate of 89.44% -- and classes which aren't in the "Skill Expert" or "Jack of All Trades" categories are only going to be able to (at most) gain mastery on one combination of an attribute and a skillset (i.e. one or two skills.) This means under most circumstances, the failure rate won't dip below about a 20% chance, which feels like a meaningful enough number.


When deciding on which skills to include, part of the idea was to "unpack" some of the skills which had been condensed down in 4th Edition D&D.
For example, Thievery (DEX) was broken down into:

  • Sleight of Hand (Dexterity, Subtlety)
  • Lockpicking (Dexterity, Infiltration)
  • Disable Device (Intelligence, Infiltration)
  • Forgery (Intelligence, Infiltration)

By doing this, we can have a larger set of more-precise skills, while controlling the bloat by limiting it only to skills which can fit into an attribute AND a skillset.

On the other hand, does 5e really need a vague Nature (INT) skill, as well as Survival (WIS) and Animal Handling (WIS)? I think the intent was to make characters who are strong at different ability scores still be able to contribute in these kinds of skill challenges. By divorcing these skills from ability scores (instead tying them to power sources, as knowledge skills) we can reduce it back down to two skills: Animal Handling (Primal, Martial) and Wilderness Survival (Primal).

When I had showed around a draft of the core skills list (to see if anything was missing) the answer I got back from a few people was the suggestion to add some kind of speechcraft and/or "decipher script" skills.
When I had decided to move Animal Handling over to a knowledge skill, there wasn't much reason left to keep Wisdom as an attribute, and it also meant that the skillset including Persuasion and Intimidation was left kind of lacking. Putting the linguistic skills into that skillset (under the Intelligence attribute) helped to even things out again; the name of the skillset went from Influence (when it included Animal Handling) to Presence (once it was dropped) and finally to Communication (when the Intelligence skills were added in.)


One other important thing with core skills is that the attributes are meant to be used as the fallback for actions that don't easily fit under one of the existing skills. This is where the distinction between Agility and Dexterity becomes most obvious --  and people always ask, when they see both being listed within the system.

When in doubt:

  • Strength should be used for actions that involve the arms/upper body/core
  • Dexterity should be used for actions that involve the hands/fingers
  • Agility should be used for action that involve the feet/legs/lower body


...

Anyways, I'll leave it at that for now. I realize this post has been sort of a long, disorganized mess -- but mainly the idea was to lay out some of the starting points for further discussion. Let me know if there is anything covered or touched on here that you would like me to talk about or expand upon.

Next post should be up March 5th so be sure to check back then.

Sunday, February 12, 2023

Interception!

Being that today marks the celebration of the Superb Owl, it seems appropriate that this post should touch on the mechanics of interceptions.

What are we talking about, in TNP terms, when we say "interception"?

  • you must be 'open'
  • as an off-turn action
  • when an enemy moves
Let's decompose this further.

Open means that you do not have any enemies adjacent to you.
An off-turn action is an action that can only be done when it is not your turn; you may only do one off-turn action per (other creature's) turn, and usually there is a trigger tied to these types of actions.

Now, if an enemy is moving, they're either trying to engage one of the PCs, or move away from a PC they are engaged with; generally, moving away from a hostile creature you're adjacent to will trigger an opportunity attack, which itself is an off-turn action. So potentially, you could have an enemy move away from you and you would have the option to make an opportunity attack, or to intercept the enemy to stop them -- but not both. Of course, since you need to be 'open' to intercept an enemy, this opportunity to pick from both types of actions only presents itself when the enemy moving away from you is the only enemy you're currently adjacent to.

I can't speak from experience, but I've heard that older (pre-3.x) editions of D&D would allow characters to use their movement speed throughout the round, allowing defenders to mimic these sorts of intercept actions. In the 4e-ish paradigm of "move action" rather than "move speed" you need to have some sort of method for allowing off-turn movement, which is what this mechanic is intended to do. (4e used the mark/punishment method for tanking, but positioning was always they key to being able to dish out opportunity attacks -- this sort of movement should also help with that.)


Cunning Stunts
A related topic that I've been agonizing over, is the 'tumble' mechanic. Now, for those who remember 3.x D&D, a tumble check could be made as part of any movement which would normally provoke an opportunity attack, in order to avoid provoking such an attack.

With 4th Edition, the balance, tumble, and escape artist skills were ostensibly combined into the single 'Acrobatics' skill. I say ostensibly, because the tumble skill completely vanished. This mechanic was replaced with 'shifting' and did not require a check; in 5th Edition, the Acrobatics skill remained, but shifting disappeared (aside from the 'Disengage' action, which is considerably more limited in its application.) I've always wanted TNP to re-introduce the tumble mechanic, largely to make Acrobatics a more relevant skill in combat. So how should we go about this?

Well, we need to add some more context.

TNP uses a "disengaging" mechanic, whereby you cannot be engaged by certain enemies until the end of your next turn; this effectively comes in 2 different flavours:
  1. Shift: disengage from the enemies in your current maelstrom
  2. Withdraw: disengage from ALL enemies in the encounter
Now, the question is, how should the acrobatics/tumble skill factor into using these actions?

Should you have to declare which method you're using, and then make the check -- burning that action if the check is a failure? Or does failing the check only mean you fail to avoid opportunity attacks?
Would a successful check let you 'step down' the action type (i.e. from Standard to Move, or Move to Minor) for whichever disengaging mechanic you're using? Or should this be a benefit only bestowed by class features?
Should the action fire off regardless, with a successful check only giving you something like... a save bonus against enemies you've disengaged from?

These are all options which I'm considering, and I haven't really nailed down or settled on one particular execution.

To bring the topic full-circle, perhaps tumble checks could also be used to avoid attempts at interception(?)

The problem that these sorts of off-turn mechanics can potentially create, is the situation where you have to go around the table and make sure no one "objects" to the action you're taking, before you can continue resolving your turn; this is especially troublesome in game formats like play-by-post. The problem is that 5e (for example) throws the baby out with the bathwater, swinging the pendulum too far in the other direction. So I'm trying to strike the right balance between having enough meaningful, impactful options during play, while not compromising elegance or game flow in the process.