| 12 people marked this as a favorite. |
PF2 has tried to give skills a bigger role to play in the game, which is great. But there's been a lot of dissatisfaction regarding the way skills currently feel in the game. This dissatisfaction seems to stem from three issues:
As noted by Deadmanwalking (here and here), and many others, monsters generally have much higher skill modifiers than a comparable PC could have. This incentivizes players to pursue purely violent ways of resolving encounters -- why risk using bluff, diplomacy, or stealth to non-violently defuse encounters (attempts which usually put you in a worse position if you fail) when you're unlikely to succeed? Likewise, the DCs given for challenges in the rulebook seem too high, and presuppose access to skill increasing magic items. This runs counter the goal of reducing magic item dependence, and (more importantly) makes PCs to seem bumbling and inept.
As eloquently noted by Ascalaphus and ryric, among others, the lack of guidance regarding skill DCs, or even representative skill DCs, has a number of undersirable consequences. It makes advancing skills unsatisfiying for players, since they don't have any tangible feeling for what this advancement means. It makes it hard for players to know what kinds of things they should be able to expect to do with their skills. It makes it hard for DMs to know how to assign DCs to various challenges. And it makes it even harder for DMs to assign DCs to challenges in a way that's consistent from session to session.
As noted by Midnightoker, among others, skill proficiency levels don't currently seem to mean much. The only requirements on skill uses just require being Trained, which makes it hard to differentiate between the higher proficiency levels. The numerical bonuses beween the higher proficiency levels is small. And while there are different skill feats available at different levels, they might not get chosen, making it easy to have cases where there's no discernable difference between someone who's Trained in Stealth and someone who's Legendary in Stealth (for instance).
Although all three problems are important, I want to focus on the third problem here. Here are four ways of resolving the third problem:
The big con of this kind of approach is that it breaks the uniform skill/saves/attack set-up the developers have set up. It also monkeys with the underlying math in a number of ways (the details here depend greatly on what the proposal is).
Athletics Proficiency Levels:
The big pro of this option is that (unlike option 1) it keeps all of the existing math and mechanics the same; it just adds another level of rules on top. The main con of this option is that it requires some delicate decisions regarding what skill feats are supposed to do, since they're supposed to add something over and above what (say) legendary proficiency in a skill already allows you to do.
This big pro of this option is that (unlike option 1) it keeps the underlying math the same, and (unlike option 2) it avoids delicate decisions about what should be a skill feat and what should be a "unlocked" skill use. And it makes for clear, dramatic differences between levels of skill proficiency. The big con of this approach is that it removes skill feats as a separate element of the game, which would require at least some changes to the way things are set up (to remove dead levels, etc).
The big pros of this option is that it ensures (unlike option 1) that the underlying math is the same, it avoids (unlike option 2) further delicate decisions about what skill uses should require which proficiency levels, and (unlike option 3) it keeps the skill feats framework. The big con is that it would also require tweaking the existing rules slightly (since one would get twice as many skill feats). And it would require adding substantially more skill feats to choose from.
__________
I like all four options (though I think option 4 is the easiest to implement, and so probably the most attractive option from the perspective of the developers).
Do any of these proposals strike you as attractive options? Are there big pros/cons to these proposals that I've missed? Any different kinds of proposal that should be considered?
|
| 5 people marked this as a favorite. |
I would like #3 or else #4. Let's start at #4 and work back to #3.
#4: gain a Skill Feat whenever you increase skill proficiency
Right now, you can become Legendary in a skill, which is supposed to be amazing, but the only thing that changes is you get +1 to your score, so it goes from about 25 to 26. The next level you get a skill feat and might start doing something worth telling legends about. This is very unsatisfying. If upgrading a skill meant immediately unlocking a new ability, that would make increasing proficiency feel like it was really a big deal.
I would add that you can also take a lower-level Skill Feat in the same skill instead of the maximum; maybe your end goal is a Legendary diplomacy Skill Feat but on the way there you're picking up two Expert Skill Feats because they're more important to you than the available Master ones.
As a consequence, you probably should get fewer or no loose Skill Feats. You can still take more with General feats, and the number of General feats you get can be slightly increased.
#3: fold all Skill Feats directly into skill proficiency levels
I think this is even a bit better because it lets us cut away a whole layer of probably unnecessary complexity. You can eliminate the entire concept of Skill Feat as a separate thing. Yay, more straightforward game.
Another advantage is that this will make it a lot easier to come up with high-DC tasks for each skill. I've written at length about why a lot of objective DCs for skill uses would be good for the game - they empower players and fire up the imagination in a way that Skill Feats currently are touted as doing. But I think the existence of Skill Feats as a game concept creates an unhealthy competition here; the temptation is to squirrel away high-level uses for skills behind Skill Feats. Drawing lines between generic high-level uses of a skill and stuff that ought to be gated behind Skill Feats would be hard.
---
If we take approach #3 then some Skill Feats need to be reconsidered, particularly the ones that require only Trained Proficiency. Arcane Sense for example requies Arcana-T to give you at-will Detect Magic. That might be a bit too much for a proficiency that people take at level 1, but if we move it to requiring Expert Arcana then it's perfectly fine.
Catfall (one of my favorite designs for skill feats right now) would also drop in people's laps for free at level 1 if they train in Acrobatics, not sure if that's ideal. Gating it behind Expert but otherwise leaving it unchanged would probably suffice however.
| 8 people marked this as a favorite. |
I've mentioned a little bit of what I think about this topic elsewhere, but I'll use this thread to go into some more detail about it. Using your scheme, it's somewhere between #2 and #3. Basically, things that you can do with a skill have some combination of gate-keeping and scaling based on proficiency. And then skill feats come in and allow you to do things with that skill that are unusual or mind-blowing. This system also greatly benefits from an unchaining of results from the +/- 10 system, but still works even when chained.
For a clear example of what I mean, I'll take a look at the Diplomacy skill, and how I would adjust it to fit my vision.
Gather Information: You canvass local markets, taverns and gathering places in an attempt to learn about a specific individual or topic. The GM determines the DC of the check and the amount of time each check takes. Success: You collect information about the individual or topic, per GM discretion; Critical Failure: You collect incorrect information about the individual or topic, per GM discretion.
- Untrained: You can attempt to gather information up to 3 times per day.
- Trained: You can attempt gather information up to 6 times per day.
- Expert: You can attempt to gather information up to 6 times per day. Treat a result of Critical Failure as a reslt of failure instead.
- Master: You can attempt to gather information up to 6 times per day. Treat a result of Failure as a result of success instead.
- Legendary: You can attempt to gather information any number of times per day. You always succeed at your attempt, and it never takes you more than an hour.
Make an Impression: With at least 1 minute of conversation consisting of charismatic overtures, flattery, and other acts of goodwill, you seek to make an impression on someone to make them temporarily agreeable. At the end of the conversation, attempt a diplomacy check against the target's Will DC, modified by any circustances per GM discretion. Success: The target's attitude toward you improves by 1 step; Critical Success: The target's attitude toward you improves by 2 steps; Critical Failure: The target's attitude decreases by 1 step.
- Trained: When you Make an Impression, you can compare your Diplomacy result to the Will DCs of up to 2 targets.
- Expert: When you Make an Impression, you can compare your Diplomacy result to the Will DCs of up to 4 targets.
- Master: When you Make an Impression, you can compare your Diplomacy result to the Will DCs of up to 10 targets. Treat the result of a Critical Failure as a result of Failure instead.
- Legendary: When you Make an Impression, you can compare your Diplomacy result to the Will DCs of up to 25 targets. Treat the result of a Critical Failure as a result of Failure instead.
{A} Request: You can make a request of a creature that's friendly or helpful to you. You must couch the request in terms that the target would accept given their current attitude toward you. Some requests are unsavory or impossible, and even a helpful NPC would never agree to them. Success: The target agrees to your request, but might demand added provisions or alterations to the request; Critical Success: The target agrees to your request without qualifications or agrees to a request that would ordinarily require it to have an attitude toward you one step better than it has; Failure: The target refuses the request, but may be open to further negotiation or other requests; Critical Failure: The target refuses the request, and its attitude toward you decreases by 1 step.
- Trained: You can make the same request of up to 2 targets at once.
- Expert: You can make the same request of up to 4 targets at once.
- Master: You can make the same request of up to 10 targets at once. Treat the result of a Critical Failure as a result of Failure instead.
- Legendary: You can make the same request of up to 25 targets at once. Treat the result of a Critical Failure as a result of Failure instead.
{A} Call the Guards: Calling for the guard requires a Diplomacy check modified by the settlement’s law modifier. It’s only a DC 5 check to call for the guard. Success: The guard arrives in 6 minutes. Success (+5): Reduce the time it takes for the guard to arrive by 1 minute. If the time was already reduced to 1 minute, reduce the time by 1 round instead.
{AAA} Direct a Crowd: You attempt to convince a crowd to move in a particular direction. The crowd must be able to see and hear you in order to be influenced. Success: The crowd moves slowly in the direction you indicate. If you are in encounter mode, it moves at a rate of 10 feet at the end of every round.
- Success (+15; Trained): Increase the speed the crowd moves by an additional 5 feet per round.
- Success (+10; Expert): Increase the speed the crowd moves by an additional 5 feet per round.
- Success (+10; Master): Increase the speed the crowd moves by an additional 10 feet per round.
- Success (+10; Legendary): Increase the speed the crowd moves by an additional 20 feet per round.
Bargain Hunter: You can gather information specifically about deals on items rather than other information. Name an item or a general category of items (such as "magic weapons") you're looking for and then roll your Diplomacy check. Any bonuses you have when Gathering Information apply. Success: You find a deal on the item you were looking for. You can purchase it at a discount equal to the value of a successful Practice a Trade check for a task of your level; Critical Success: You find a deal on the item you were looking for. You can purchase it at a discount equal to the value of a successful Practice a Trade check for a task of one level higher than your level.
- Expert: You can Practice a Trade with Diplomacy, representing spending your days hunting bargains and reselling for profit.
- Master: When you Practice a Trade with Diplomacy, treat any result of a Critical Failure as a Failure instead.
- Legendary: When you Practice a Trade with Diplomacy, treat any result of Failure as a Success instead.
Glad-Hand (Expert): First impressions are your strong suit. When you meet someone, you can immediately attempt a Diplomacy check to Make and Impression with a -5 penalty rather than needing to converse for 1 minute. On any Failure result, you can continue spending 1 minute of conversation to attempt a new check rather than accept the Failure result. You may not use this ability in encounter mode.
- Master: Reduce the penalty to -4.
- Legendary: Reduce the penalty to -2.
{AAA} Legendary Negotiator (Legendary): You use your incredible skill at persuasion to negotiate quickly in adverse situations. You attempt to Make and Impression and then immediately Request that your opponents cease their current activities and engage in negotiations.
Hypnotism (Expert): You use the power of suggestion and subtle psychic influence to alter a subject’s mind and dredge up repressed memories. Hypnotizing a creature requires 1 minute inducing a trance-like state in the subject, who must be willing to be hypnotized. Hypnotism can be used to either Recall Memories or Implant Suggestions.
Recall Memory: You can draw out forgotten memories from a willing subject. Make a Diplomacy check against the target's will DC. Once completed, the trance ends. Success: The target immediately attempts to Recall Knowledge about a topic they might possibly have once known or been exposed to.
- Success (+10; Expert): The number of topics you can cause the target to Recall Knowledge about increases by 1.
- Success (+5; Master): The number of topics you can cause the target to Recall Knowledge about increases by 1.
- Success (+5; Legendary): The number of topics you can cause the target to Recall Knowledge about increases by 1. You can attempt to hypnotize an unwilling creature as well. An unwilling creature must be restrained or fascinated for the duration of the check.
Implant Suggestion: You can implant a suggested course of reasonable action in the mind of creature, along with a defined trigger. Make a Diplomacy check against the target's Will DC. Success: You implant the course of action, as a suggestion spell with a duration of 10 minutes from when the triggering condition applies
- Success (+10; Expert): The duration of the suggestion increases by 10 minutes.
- Success (+5; Master): The duration of the suggestion increases by 10 minutes.
- Success (+5; Legendary): The duration of the suggestion increases by 10 minutes. You can also attempt to hypnotize an unwilling creature. An unwilling creature must be restrained or fascinated for the duration of the check, and receives a Will save to resist the suggestion (use the spell results for the saving throw) once the triggering condition applies, using your Diplomacy DC as the save DC.
-----------------------------------
And there can be other, interesting and oddball uses for Diplomacy - but I think I've gone on long enough. I would generalize this pattern to all of the skills. So some things you get access to just by increasing your proficiency. And things you can already do automatically get better for you as you increase in proficiency beyond where you acquired them. Always forwards, never stagnant. Is this more complicated to design? Possibly. I went through this off the top of my head without too much difficulty - the hardest part was sorting out how good each reward for a new tier of proficiency should be (whether I did a good job of balancing it is another matter entirely). I think the benefits of embracing this structure in terms of both fun and a definite feeling of progression are worth the additional thought required.
| 2 people marked this as a favorite. |
You could eliminate the Assurance feat and use a modified version of it for the TEML proficiency system. You roll the dice and
Expert: your minimum roll is 5
Master: your minimum roll is 10
Legendary: your minimum roll is 15and then unlike the current Assurance feat, you can add your bonuses to the roll.
Do you mean that that is the minimum value on the die (then add other things)?
Pretty sure Paizo doesn't want their skill system to work like that.
| 1 person marked this as a favorite. |
@Gaterie: That's another interesting option. It seems similar to the first option in a lot of ways, both with respect to it's advantages (effectively yielding conditional numerical bonuses as one gains skill proficiency levels, which allows for bigger effective differences with respect to the same DC tasks) and potential disadvantages (treating skills proficiency checks in a way that's disanalogous to how they're treated in other domains, like attacks). Certainly something worth considering.
@Ascalaphus:
Re #4: I agree that it would also be natural to allow someone to (say) spend the skill feat associated with the increase to legendary proficiency on a lower level skill feat if they desired. I considered that, but was worried that this would allow the initial worry to creep back in: it would be possible for someone who was Trained in Athletics and someone who was Legendary in Athletics to be virtually on a par (if the person who was Legendary spent all of their skill feats on silly things at the Trained level, say).
But I can certainly see the appeal of going the way you suggest!
And I agree that this option forces one to make a decision about whether to keep the number of total skill feats the same (in which case there are virtually no free skill feat choices, unless one burns general feats on them), or whether to make these mandatory skill feats bonus skill feats. I was suggesting the latter option in order to allow for more customization with respect to how one developed one's skills. (And because skill feats seem pretty weak in general, so adding more of them didn't seem to be particularly threatening from a balance perspective.) But I might have been too cavalier there... In any case, I think either option would yield something more satisfying than the current set-up.
Re #3: In my heart of hearts, I think I agree that this is probably the most attractive way to go. In addition to the advantages you describe, it would eliminate the possibility of people adding skill feats which retroactively ban certain uses of skills for those without that feat. (Something which many people found infuriating in PF1 -- "wait, now I need to have a feat to do this?!")
I worry about whether this is too radical a change for the devs to take on board. But If they do, I'm all for it!
@Leedwashere:
Wow. That's an impressively detailed proposal! And as someone who's pretty sympathetic to something along the lines of #3 (as i noted above), I find this approach pretty appealing as well.
My initial concern was that it might require too much work to develop, but after seeing your pretty comprehensive write-up, I think I might have been to pessimistic. Very interesting.
You could eliminate the Assurance feat and use a modified version of it for the TEML proficiency system. You roll the dice and
Expert: your minimum roll is 5
Master: your minimum roll is 10
Legendary: your minimum roll is 15and then unlike the current Assurance feat, you can add your bonuses to the roll.
This might work well with Porridge's idea of
Option 3. Merging Proficiency and Skill Featswhich I really like
I think eliminating (or heavily modifying) Assurance is a good idea no matter what. So I'm definitely on board with that.
This proposal strikes me as an interesting version of option 1: changing the numbers in way that makes higher proficiency levels more consistent.
It's certainly a better approach, I think, than versions of option 1 that just increase the bonuses proficiency levels provide. Those proposals run directly counter to the PF2 idea of allowing one to create challenges that can be both easy (but not automatic) for optimal skilled characters, and hard (but not automatically failed) for unskilled characters. This proposal (like proposals people have made about taking the best of multiple die rolls when you get a higher level of proficiency) avoid this problem.
And, as you say, it could be merged with option 3, as they're not mutually exclusive.
The main potential complaint one might raise is just that this treats skills in a different way from attacks and saves. But how big a cost that is is something I can see a lot of reasonable disagreement about.
(I.e., it's prima facie nice to have consistency across numbers in these different regimes, since it makes encounter construction easier, and the like. But it might also seem that to a large extent, skills just play different roles than attacks and saves, and so trying to treat them all the same way is a mistake.)
| 3 people marked this as a favorite. |
While I like a lot of the ideas presented here, I think the idea of not adjusting the the proficiency bonuses is skirting the psychological issues with the way the system was applied to skills.
To begin with the tight math being used and this absurd adherence to 50% success rate feels like "every one is a special snowflake" crap.
Some system try to fit a square peg into a round hole.
This system is trying to make every peg fit in the same hole.
And that leaves you with a lot of same looking pegs.
Trained should be a world apart from untrained.
There is a huge difference from gaining expertise in something and just knowing how to do it.
Masters are not just anyone.
And Legends are supposed to be unbelievable.
This:
-4/0/+1/+2/+3
...has very poor conveyance of actual progression.
While many of the solutions presented here do give a very good indication of how much you have improved at higher proficiencies, those things are not what I spend most of my time looking at while in actual play.
When I look down at my character sheet, the first thing I'm going to see every time is that measly +1, +2, or +3.
There is a lot of psychology to that is being ignored.
If the game play doesn't SEEM satisfying, all the perfect math and balance in the world will not hook people into it.
Everything just becomes bland, and is what I've heard is the pitfall 4e fell into.
So I say the best solution is to buckle down and make the experience satisfying from both angles.
The spread should be like this:
-4/0/+3/+6/+9
This makes the increase feel real just by glancing at the numbers.
The level bonus is terrible.
Just because you watch someone do something a bunch of times does not make you good at it.
I can see a character being able to negate the untrained bonus over time, but not actually be better than a trained character.
Level bonus should be this:
level/4
At level 20 you are at least getting a +1 bonus instead of a penalty.
Hell, at that point I'd even give you access to the Trained Uses.
Now a character who never trained in a skill feels like they have become competent through observation and self-reliance over time.
The DCs would have to be overhauled, but that 10-2 table is fast becoming hated anyway.
If you combine this with one of the solutions presented above, you get a skill system that is functional AND satisfying from both a psychologically and practically point of view.
| 1 person marked this as a favorite. |
Option 2. Gatekeeping: Another suggestion is to add skill uses that require higher levels of proficiency to be performed. For example, one might add to each skill a chart like this:
Athletics Proficiency Levels:
Trained or higher proficiency is required for tasks such as maneuvering in flight, climbing surfaces without handholds, swimming (not just staying afloat) in rough waters, etc.
Expert or higher proficiency is required for tasks such as climbing surfaces with a negative slope, climbing (with equipment) up Mt. Everest, swimming down to the bottom of a deep lake, etc.
Master of higher proficiency is required for tasks such as climbing upside down or climbing one-handed, swimming up waterfalls or whirlpools, jumping to the top of a tree, or jumping out of a cliff and landing unharmed, etc.
Legendary proficiency is required for tasks such as doing things with both arms (e.g., firing a bow) while hanging upside down, diving down to the ocean floor, jumping over a tall tower, etc.
The big pro of this option is that (unlike option 1) it keeps all of the existing math and mechanics the same; it just adds another level of rules on top. The main con of this option is that it requires some delicate decisions regarding what skill feats are supposed to do, since they're supposed to add something over and above what (say) legendary proficiency in a skill already allows you to do.
This is something that is already referred to in the Rulebook under 'Proficiency-Gating' on page 336. It's the recommended method of running skill challenges.
I do kind of wish they employed it a bit more in the modules to give people a bit more visibility to it though!
|
|
| 1 person marked this as a favorite. |
Option 2. Gatekeeping: Another suggestion is to add skill uses that require higher levels of proficiency to be performed. For example, one might add to each skill a chart like this:
Athletics Proficiency Levels:
Trained or higher proficiency is required for tasks such as maneuvering in flight, climbing surfaces without handholds, swimming (not just staying afloat) in rough waters, etc.
Expert or higher proficiency is required for tasks such as climbing surfaces with a negative slope, climbing (with equipment) up Mt. Everest, swimming down to the bottom of a deep lake, etc.
Master of higher proficiency is required for tasks such as climbing upside down or climbing one-handed, swimming up waterfalls or whirlpools, jumping to the top of a tree, or jumping out of a cliff and landing unharmed, etc.
Legendary proficiency is required for tasks such as doing things with both arms (e.g., firing a bow) while hanging upside down, diving down to the ocean floor, jumping over a tall tower, etc.
The big pro of this option is that (unlike option 1) it keeps all of the existing math and mechanics the same; it just adds another level of rules on top. The main con of this option is that it requires some delicate decisions regarding what skill feats are supposed to do, since they're supposed to add something over and above what (say) legendary proficiency in a skill already allows you to do.This is something that is already referred to in the Rulebook under 'Proficiency-Gating' on page 336. It's the recommended method of running skill challenges.
I do kind of wish they employed it a bit more in the modules to give people a bit more visibility to it though!
It's alluded to a bit in Rose Street Revenge, that only an Expert in Perception can notice some traps. Of course, after your third level general feat, everyone is an Expert in Perception...
| 2 people marked this as a favorite. |
Option 2. Gatekeeping: Another suggestion is to add skill uses that require higher levels of proficiency to be performed. ... . This is something that is already referred to in the Rulebook under 'Proficiency-Gating' on page 336. It's the recommended method of running skill challenges.
I do kind of wish they employed it a bit more in the modules to give people a bit more visibility to it though!
Yeah, good. That makes it seems like Option 2 is the option the developers are currently intending to employ to distinguish between proficiency levels.
Given that, the lack of guidance regarding skill proficiency requirements, or even representative proficiency requirements, has a number of very unhappy consequences:
I.e., it basically raises all of the same problems that arise for the second big skills problem ("Skill bonuses/DCs don't mean anything").
As Ascalaphus and ryric have convincingly argued, the only viable solution to the second big skills problem seems to be providing, for each skill, explicit guidelines regarding what kinds of challenges require what kinds of DCs, and/or providing a number of representative DCs for particular kinds of challenges.
And if gatekeeping is how proficiency levels are going to be distinguished, then it seems the same needs to be done here: the book needs to provide, for each skill, explicit guidelines regarding what kinds of challenges require what kinds of proficiency levels, and/or provide a number of representative proficiency requirements for particular kinds of challenges.
| 4 people marked this as a favorite. |
Two follow up thoughts:
1. If the developers stick with option 2, and do provide detailed guidelines about what kinds of tasks require what kinds of proficiency levels, then I'd be fine with it.
But it's hard for me to shake the feeling that this way of distinguishing between proficiency levels raises a number of uncomfortable issues. For example, it becomes very important to spell out, for each skill, what the difference is between a high DC and low proficiency skill challenge and a low DC and high proficiency skill challenge. And it's not clear (to me, at least) how to do that in a principled manner.
2. Echoing a comment I made in Ascalaphus' thread, if option 2 is how we're distinguishing between proficiency levels, then assigning concrete DCs and concrete proficiency requirements to different tasks is something it's really important for the developers to work out -- much more important (IMO) than pretty much any other aspect of the game.
Here's why. There are a lot of things about PF2 I'd like to see tweaked (dying rules, mandatory magic items, resonance, sorcerer spell flexibility, etc). But almost all of these are things which I could introduce relatively simple house rules to change.
That's not true for skill DCs and proficiency requirements. There's no simple house rule I can slap on to get representative DCs and proficiency requirements for different skill challenges (to say nothing of working out how high DC challenges and high proficiency requirements are supposed to be different!). This is something that requires a lot of time and design expertise to do well. And that's precisely the kind of thing you want professional designers to do for you.
|
| 2 people marked this as a favorite. |
Linking skill increases or making skill feats automatic seems to be a decent option, though I am not entirely sure how that will affect future options.
I think having a fair number of set DCs (like jumping or climbing a rope) will be pretty critical longterm, while others really should depend on the challenge at hand.
Diplomacy is a great example since the RAW rules are usually ignored in favor of custom DCs at least in something like PFS scenarios.
I am not entirely sure how much gatekeeping will be useable (thinking about organized play and modules) since you can never be sure that you have a character with that proficiency level at the table, and the adventure needs to have a way to continue regardless.
Part of me would prefer some sort of automatic unlock in each skill (not unlike untrained and trained skill uses) whether those are taken from skill feats or written for each skill might not matter all that much.
Since they are also used offensively, increasing the actual bonus might be off the table, but messing with the way you roll the dice (rolling twice, rerolling etc.) might be an option to make the increased proficiency rank feel better.
Of course, altering or replacing Assurance could solve a lot of those problems, right now the feat is better for you if you are terrible at the skill (low attribute bonus, ACP) which just feels rather unrewarding.
| 1 person marked this as a favorite. |
Linking skill increases or making skill feats automatic seems to be a decent option, though I am not entirely sure how that will affect future options.
I think having a fair number of set DCs (like jumping or climbing a rope) will be pretty critical longterm, while others really should depend on the challenge at hand.
Diplomacy is a great example since the RAW rules are usually ignored in favor of custom DCs at least in something like PFS scenarios.
I am not entirely sure how much gatekeeping will be useable (thinking about organized play and modules) since you can never be sure that you have a character with that proficiency level at the table, and the adventure needs to have a way to continue regardless.
Part of me would prefer some sort of automatic unlock in each skill (not unlike untrained and trained skill uses) whether those are taken from skill feats or written for each skill might not matter all that much.
Since they are also used offensively, increasing the actual bonus might be off the table, but messing with the way you roll the dice (rolling twice, rerolling etc.) might be an option to make the increased proficiency rank feel better.
Of course, altering or replacing Assurance could solve a lot of those problems, right now the feat is better for you if you are terrible at the skill (low attribute bonus, ACP) which just feels rather unrewarding.
I certainly agree that replacing Assurance would be a good idea! I think that the kinds of worries motivating the third kind of skill problems I described ("Skill proficiency levels don't mean much") will still be around, though.
As for how to deal with that problem, the kind of "skill unlock" suggestion also sounds like a nice option. It'd sort of be a more conservative version of option 3.
Likewise, linking skill increases and the level of skill feats one has sounds like an attractive option to me (that's more or less along the lines of option 4).
Gloom's point that the devs seem to be adopting option 2 has made me try to reconcile myself to that thought. But I'm starting to get a bit more worried about this option, for both reasons you've raised, and reasons raised in the "Skill Gating - Common Misconceptions about Skill Proficiency" thread.
|
|
| 2 people marked this as a favorite. |
I've thought it over and my preference is actually #4: get one skill feat every time you raise a skill. Why? Growth opportunities for the system.
If we say that when you gain a new skill level you gain all the new options, then you basically need to codify all of those options in the CRB, otherwise you get characters getting more powerful just because a new book was published. Or actually, new books probably won't publish new skill uses that you gain automatically. They'll probably make general feats that look suspiciously like skill feats, but that you don't get for free.
With the #4 approach, we basically mandate that the CRB includes at least one skill feat for every proficiency tier for every skill, and preferably two, so that there's some choice. Seems like a good thing to me.
Ideally, we also get defined skill check DCs for non-feat uses of skills going up to DC 50 (as I argued in my own thread). Those would become a sort of guarantee: "these things you can already do without a skill feat; everything not in here is up in the air". If a player wants to try something that's clearly part of the skill, but not in the table, the GM can set a DC by comparison, or it could be something they come up with a skill feat for.
| 1 person marked this as a favorite. |
I've thought it over and my preference is actually #4: get one skill feat every time you raise a skill. Why? Growth opportunities for the system.
If we say that when you gain a new skill level you gain all the new options, then you basically need to codify all of those options in the CRB, otherwise you get characters getting more powerful just because a new book was published. Or actually, new books probably won't publish new skill uses that you gain automatically. They'll probably make general feats that look suspiciously like skill feats, but that you don't get for free.
With the #4 approach, we basically mandate that the CRB includes at least one skill feat for every proficiency tier for every skill, and preferably two, so that there's some choice. Seems like a good thing to me.
Ideally, we also get defined skill check DCs for non-feat uses of skills going up to DC 50 (as I argued in my own thread). Those would become a sort of guarantee: "these things you can already do without a skill feat; everything not in here is up in the air". If a player wants to try something that's clearly part of the skill, but not in the table, the GM can set a DC by comparison, or it could be something they come up with a skill feat for.
Yeah, I think those are compelling reasons. I've been moving toward option #4 as well.
I've been trying to find easier ways to capture something like this idea, and was toying with something like this:
(b) Whenever you get a skill feat, you can choose a skill feat in a skill that's up to one proficiency level higher than some other skill feat you have in that skill. (So if you have an expert-level skill feat in Stealth, and a trained-level skill in Thievery, you can choose a master-level skill in Stealth or a expert-level skill in Thievery.)
The main con of this version of 4.1 is that it disincentivizes getting more than one skill feat of a given proficiency level (e.g., a second expert-level Stealth skill feat) because such a choice won't yield an additional "+1" numerical bonus with any of your skills.
So I think I probably like the variant of option 4 sketched above better. (That whenever you increase your proficiency in a skill, that increase automatically comes with a corresponding skill feat of the same proficiency level (e.g., if you become an expert in Stealth, you also get an expert-level Stealth skill feat).)
I still think we need the option to "slum" for a lower-grade skill feat than the maximum you're entitled to, especially because there's a lot of Trained skill feats around. But apart from that, I prefer the 4.0 option most now.
Yeah, I agree. So there's just the tricky question of how to best do that... Here's one thought (different from the one sketched above):
1. Let your proficiency level in a skill = the highest level skill feat in that skill you have. (To ensure that anyone who's legendary in Stealth can actually do something legendarily stealthy, etc.)
2. Replace the current skill feat and skill proficiency increases with a single disjunctive option gained every other level: You can either (a) pick *2* skill feats up to your current proficiency level in a skill, or (b) pick 1 skill feat one level higher than your current proficiency level in a skill (which would thus increase your proficiency level in that skill).
If you choose (b) every time, you'll end up with the same number of skill proficiency increases and skill feats as in the current system. If you choose (a) sometimes, you'll end up with fewer skill proficiency increases than the current system, but more skill feats.
And it would allow you to choose how to balance the quality of skill feats you can get and the quantity of skill feats you get. How does something like that sound?
| 1 person marked this as a favorite. |
Trained- This is pretty well covered. You get effectively a +4, access to additional skill options, and can take skill feats.
Expert- This is rather lame. You get a +1, and can take expert level skill feats. Not all skills have at least one expert skill feat, and most of them are highly situational at best.
Master- This is even more lame. You get a +1 and you can take master level skill feats. Except only half of the skills have master level skill feats and most of those are highly situational.
Legendary- This is rather lame. You get a +1 and you can take legendary level skill feats. At least more skills have legendary level options than master level options, but they are still very situational.
Remember, there are a limited number of skills that you can take beyond trained. All classes except rogue have 9 skill increases, so if they specialize, as the skill charts expect them to do, they can have at most 3 skills at each expert, master and legendary levels. Right now there is very little incentive to go past expert in any skill. In fact it is probably numerically superior to spend these skill increases increasing the number of trained skills at your disposal than increasing proficiency level. About the only skill worth maxxing out is thievery, and that only because we know there are proficiency gates that have to be overcome. Unless every skill is expected to have extensive gating there needs to be some incentive to increase your proficiency.
Some suggestions:
Expert: 1 free reroll per day
Master: Treat successes as critical successes or
Treat critical failures as failures
Legendary: Treat successes as critical successes or
Treat critical failures as failures or
Roll twice and take the better result for all checks
As far as the numbers go, I think a simple change to the bonuses would help make the increases feel better.
Untrained: 0 (no level bonus, or if we eliminate level bonuses entirely make this -4)
Trained: +1
Expert: +3
Master: +6
Legendary: +10
Then you can also roll in skill feats/gates into those skills if you like. What's nice about this kind of setup is it forces players to make a decision between eliminating a large penalty on a skill, or getting increasing returns on upping an existing skill.
| 1 person marked this as a favorite. |
Untrained: 0 (no level bonus, or if we eliminate level bonuses entirely make this -4)
Removing or reducing the Level Bonus from Untrained Skills specifically says to a player 'I don't want you to possibly compete on a skill challenge with untrained skills.' One of the bigger issues here as well is that removing the level bonus would skew testing and experiences, causing higher level players to have much worse chances on a skill check than lower level players. All in all a terrible experience if you ask me.
This stands true for extremely basic checks.
Having a -4 penalty is already putting them at a 20% disadvantage to every other player who has their skill at a trained level of proficiency.
With the skill DCs as tight as they are right now having a -4 penalty is a huge difference that most people won't realize how extreme it is until it's their turn to make a roll.
If your concern is that you do not want someone who is untrained in a skill to be able to make a roll on more nuanced uses of a skill because they are not trained then that is already the case.
If your concern is that you do not want someone who is untrained in a skill to be able to make a roll on more nuanced uses of a skill because they are not trained then that is already the case.
That's not really the goal I'm trying to get across. -4 is also fine, and personally I think the level bonuses in their entirety are pointless. IF you remove it from the game the math remains largely the same.
What I want is to have the initial skill proficiency to feel worth it, but also specialization to feel worth it as well just from a bonus perspective. The math being as tight as it is means players are commonly feeling 50/50 on all their rolls. While I've not gotten a chance to play level 7+, I've yet to feel like any of the modifiers are meaningful even with level bonus. Even playing a level 4 paladin, I missed more often than I hit. And I feel similar outcomes as a result of skills.
So either the gates need to be more integral and in the forefront, or the bonuses need to start feeling more potent.
| 3 people marked this as a favorite. |
All for #3 lite, each proficiency unlocks more skill uses. I think that solves the PFS problem of not having anyone capable of making the check.
I think getting a free feat with each skill increase is very flavorful as demonstrating increased mastery in that skill by training towards a feat matches up with a lot of good stories.
I still would provide some general skill feats to allow for choosing those below your proficiency. Perhaps you can only spend those on feats that require proficiency level below the current skill level.
Yeah, if we're keeping the proficiency numbers the same, then there are basically two routes to make higher skill proficiencies feel like they mean something:
The current rules are a weird mixture of the negative and positive approaches, largely focusing on the negative. Being trained allows certain new skill uses, but other than that, higher proficiency only serves as a prerequisite for certain kinds of skill challenges.
But the more I think about it, the more I'm inclined to think that a fully positive approach is better. As sadie and others in that thread point out, this pre-requisite function isn't very fun or "special"-feeling for those who can meet the prerequisite (so the super special thing I can do by being legendary in Thievery is get to attempt to pick a really hard lock? whee...), and feels punishing for those who can't.
Whereas an approach which gives you something special and proactive that you can do when you get a new proficiency level feels much better -- you get some new special thing that you can do with the skill (over and above just the ability to attempt certain skill challenges).
TLDR: Like you, I've become more and more persuaded that some version of the Positive Approach would be the best way to make skill proficiencies feel meaningful.
One thing I notice is that the 4 degrees of success has LOTs of room for tweaking how a certain skill check feels and resolves.
Range
Number of Targets
Time between uses, bolstering duration
Actual result, like reducing conditions by 1 or more stages/steps.
Failure cost.
Affect on failure. Some of the skill feats do tweak these basic skill check results in one aspect or the other, but it
would be awesome if proficiency had more impact. For some things, Master and Legendary mention this on the skill, but very rarely does Expert matter.
That's a good point. Another interesting way for differences in proficiency level to have a more tangible impact without changing the underlying math.
I like many of the suggestions I've read about this "problem".
It's great to see many minds working together.
I have one more suggestion.
What do you think about this:
Untrained: -4
Trained: +0
Expert: +lvl/2
Master: +lvl
Legendary: +lvl+3
(restricted access to M and L)
I mapped that out on a per level basis and what I found is:
* Untrained and Trained stay static at all levels so your characters never improve as they level up.
* At level 1 there is no difference between an expert, a master, and a legendary; even if it is unlikely.
* At level 2 there is still no difference between an expert and a master.
* The bonus gaps between expert, master and legendary increase with level but the ratios between them and their differences are strangely staggered from level to level.
I have one more suggestion.
What do you think about this:Untrained: -4
Trained: +0
Expert: +lvl/2
Master: +lvl
Legendary: +lvl+3(restricted access to M and L)
It's on the same way of what I had in mind the first time I read the miserable modifiers of the current proficiency system, but I went a bit more extreme in my theorycrafting:
Untrained: 1/2x level + 1/2x ability score (max +12)
Trained: 3/4x level + ability score (max +20)
Expert: level + ability score (max +25)
Master: 5/4x level + 3/2x ability score (max +32)
Legendary: 3/2x level + 2x ability score (max +40)
As an alternative, without changing the maximum values, the last two ranks could be like this:
Master: level +5 + 3/2x ability score
Legendary: level +10 + 2x ability score
All this is based on the BAB progressions in PF1 and so it is somewhat a way to recover some of the past. Fixed DCs in the game should be set upon the Trained or the Expert score, so to have a standpoint to decide what is easy and what is hard.
The point is: Untrained has to really suck, Legendary has to really shine, every step forward along the progression should really be meaningful.
| 1 person marked this as a favorite. |
My personal preference, as this feels true to life, is to simply have "modifier" and "proficiency" be a 2-dimensional space where the former dictates how well you can do at tasks you are qualified for and the latter illustrates the extent of what you are qualified for.
Like someone can be really good at multiplying 3-digit numbers, but might not have a clue how to find the Galois Group for a given polynomial. Someone might be able to grill a tasty steak, but would be baffled if asked to make a soufflé. As I see it the difference between "changing your oil" and "rebuilding a differential" is more a difference of your conceptual and theoretical understanding of how cars work than anything else.
I don't want proficiency to have any more mathematical benefit for skill rolls than it does now, I want more things gated by proficiency level where you need acumen beyond training to even attempt them. I don't love the notion that being familiar with how to complete the advanced tasks means you are automatically much much better at the basic ones.
Problem with skills and their relative impact of training to general bonus is:
1st: +1 per level, and training above trained(as it gives +4 to check) will feel meaningless in comparison to +15 from level, or what not.
2nd: every +1 means little as long as d20 is used as reference for success chance. Having same chance to be average or to be best possible or worst possible is dumb. Things don't work that way.
Best efforts are rare. 5% is not rare. If we had 3d6 then best result would be 1/216 tries. That IS rare.
Also with 3d6 every +1 would be worth more than with d20. especially in the middle of the distribution curve.
Problem with skills and their relative impact of training to general bonus is:
1st: +1 per level, and training above trained(as it gives +4 to check) will feel meaningless in comparison to +15 from level, or what not.
2nd: every +1 means little as long as d20 is used as reference for success chance. Having same chance to be average or to be best possible or worst possible is dumb. Things don't work that way.
The system in Dungeons & Dragons 3rd Edition, which Pathfinder 1st Edition copied, is that the odds of success were 5%, 10%, 15%, ..., 95% depending on the Difficulty Class and the bonuses. This is not the same chance.
Best efforts are rare. 5% is not rare. If we had 3d6 then best result would be 1/216 tries. That IS rare.
Fluke luck is rare. Best effort is what a character can reliably accomplish well. 5% is too low a chance to be best effort.
Neverthelss, I see that you want some rare events to occur, but occur less often than 1 time out of 20. That is understandable. 5% is infrequent but not rare.
But you have to think about the mathematics of dice rolls and how bonuses interact with them in order to make rare events occur rarely. Rolling an 18 on 3d6 does occur 1 time out of 216, which is a good probability for rare. But if you give the 3d6 roll a +1 bonus, then a roll of 18 or higher occurs of 17 or 18 is rolled on 3d6, 4 times out of 216, also known as 1 time out of 54. A +2 bonus to the 3d6 makes DC 18 possible on a roll of 16, 17, or 18, 10 times out of 216, which is roughly 5%. We already decided that 5% is too frequent to be rare.
Using 3d6 to have rare 1 out of 216 events would mean scrapping the +X bonus system. GURPS uses 3d6 and their method is not bonuses, but changing the target number.
Also with 3d6 every +1 would be worth more than with d20. especially in the middle of the distribution curve.
The highest increase, going from rolling 11 or higher on 3d6 to rolling 10 or higher on 3d6 is 27/216 = 1/8 = 12.5%. That is not far from a +2 bonus on 1d20. Switch your rolls to 1d8 and get 12.5% for every +1 bonus.
The probability distribution of 3d6 from the minimum of rolling 3 to the maximum of rolling 18 is 1, 3, 6, 10, 15, 21, 25, 27, 27, 25, 21, 15, 10, 6, 3, 1.
Further compounding the issue is how the probabilities impact the characters. 5%, aka 1/20, is not much, but going from hitting an opponent on a roll of 19 or better on 1d20 to a roll of 18 or better is a 50% increase in average damage per attack. Yet, an attack that succeeds only 3 times out of 20 is either a desperation measure or a bonus on top of another tactic rather than a regular tactic. Probabilities have gameplay meanings, and a rare 1 out of 216 event is something that the party can never rely on.
I have one more suggestion.
What do you think about this:Untrained: -4
Trained: +0
Expert: +lvl/2
Master: +lvl
Legendary: +lvl+3(restricted access to M and L)
It's on the same way of what I had in mind the first time I read the miserable modifiers of the current proficiency system, but I went a bit more extreme in my theorycrafting:
Untrained: 1/2x level + 1/2x ability score (max +12)
Trained: 3/4x level + ability score (max +20)
Expert: level + ability score (max +25)
Master: 5/4x level + 3/2x ability score (max +32)
Legendary: 3/2x level + 2x ability score (max +40)As an alternative, without changing the maximum values, the last two ranks could be like this:
Master: level +5 + 3/2x ability score
Legendary: level +10 + 2x ability scoreAll this is based on the BAB progressions in PF1 and so it is somewhat a way to recover some of the past. Fixed DCs in the game should be set upon the Trained or the Expert score, so to have a standpoint to decide what is easy and what is hard.
The point is: Untrained has to really suck, Legendary has to really shine, every step forward along the progression should really be meaningful.
I have played with the mathematics of different proficiencies advancing at different rates. My main reason is that I would prefer that proficiency increase at +1/2 per level rather than +1 per level, but +1/2 per level to everything has a major flaw: due to rounding down it alternates between +0 per level at odd levels and +1 per level at even levels. That gives the full problems of both +0 per level and +1 per level.
Mixing +1/2, +3/4, +1, +5/4, and +3/2 gives +1 to everything at every 4th level and +1 to almost everything at even levels, and +0 to almost everything at odd levels. That is hard to balance at individual levels.
Thus, I tend to use fractions such as (0.6)*level. I also like combining 0.5*level and 0.5*level+0.5, which alternate which one advances. The plan is that untrained, trained, expert, master, and legendary would not all advance at the same level. Of course, since some proficiencies advance about 50% of the time and others advance about 75% of the time and legendary advances 100% of the time, some overlap is inevitable. My latest table is:
level____ 1..2..3..4..5..6..7..8..9.10.11.12.13.14.15.16.17.18.19.20
untrained -3<-2=-2<-1=-1=-1<<0==0<<1==1==1<<2==2<<3==3= =3<<4==4<<5==5 Slope 8/19 = 0.42 using (0.4)n - 2.6
trained__ 1==1<<2==2<<3==3==3<<4==4<<5==5==5<<6==6<& lt;7==7==7<<8==8<<9 Slope 8/19 = 0.42 using (0.4)n + 1
expert___ 2<<3==3<<4==4<<5<<6==6<<7==7<<8<< 9==9<10=10<11<12=12<13=13 Slope 11/19 = 0.58 using (0.6)n + 1.8
masterful 3<<4==4<<5<<6<<7==7<<8<<9 10=10<11<12<13=13<14<15<16=16<17 Slope 14/19 = 0.74 using (0.75)n + 2.5
legendary 4<<5==5<<6<<7<<8==8<<9<10<11<12<1 3<14<15<16<17<18<19<20<21 Slope 17/19 = 0.89 mixing (0.75)n + 3.5 and n+1
which views as less scrambled in a fixed-width font. I have posted other tables, such as the one at What about having "remove +1 / level" as an optional rule, at least? on October 29. I keep experimenting with different tables.
And I confess that I am currently looking an another approach, merging pairs of levels together into one fat bilevel with staggered advancement in the level, discussed at Feedback and concerns on the math of Pathfinder 2 comment #57. That approach, alas, does nothing for the problem of differences between proficiencies.
| 2 people marked this as a favorite. |
I replied to this thread due the the mathematical discussions today, but I have an opinion on Porridge's original topic from October.
A few days ago, after someone made a remark about how skill feats gave a character something to learn besides combat, I wondered how much that was so. Some skill feats, such as Defensive Climber that means a climber is no longer flatfooted or Diehard that makes dying take longer, are about combat. What fraction of them were combat feats?
So I started cataloging the feats. I found they fell into 4 major classes:
1. Combat feats, that directly help combat. This was the smallest category.
2. Adventuring feats, that help with adventuring activities such as scouting, climbing, battlefield medicine, etc.
3. Professiona feats, that a townsfolk working with the skill might learn professionally.
4. Meta feats, which are feats that give feats or allow one skill to substitute for another.
I ran out of enthusiasm for this cataloguing at the letter F. Reading three pages of feats revealed is how trivial most feats are. For example, consider Bargain Hunter, a feat 2 that requires expert in Diplomacy. If a character spends all day shopping, they can get a discount on the purchase of an item. How big is discount? The discount equals the amount the character would have earned if he or she had Practiced a Trade. In other words, without Bargain Hunter, the character could have earned the extra money and then spend a few minutes shopping and walked away with the same amount of coins in his or her purse. All that Bargain Hunter saves is the time to look for a job.
And these trivial feats would make a good reward for simply increasing the skill to the prerequisite level. For example, trained in Diplomacy could automatically give Group Impression and Hobnobber and expert in Diplomacy could automatically give Bargain Hunter.
Many feats are not trivial, and those could be left as feats. For example, Glad-Hand (feat 2, expert in Diplomacy) allows a one-action Make an Impression at a -5 penalty. That alone is trivial, but upon failure, the character can try again with the usual one-minute Make an Impression. Rolling twice is a nontrivial feat.
Thus, we actually have the material to make a lot of gated abilities for Porridge's Option 2. Gatekeeping. They are currently behind two gates: gain the proficiency rank and spend the feat--when they are better suited for one gate: gain the proficiency rank.
I like your skill feat categorization.
And I agree that if PF2 is going option 2 — using gatekeeping to distinguish between proficiency levels — we definitely don’t want to use skill feats as gatekeepers. Indeed, my least favorite feats in PF1 were gatekeeping feats — feats that implied that you now had to have a feat in order to use a skill to X. Cutting out those kinds of feats would go a fair way to improving how interesting skill feats are to choose between.
Of course, the difference between a “gatekeeping feat” and a “feat that allows you to use a skill in some cool new way” isn’t always completely clear. It’s going to hang on what one thinks one should of normally be able to use a skill to do. But I guess I’d like to see them avoid requiring skill feats for anything that even remotely resembles an ordinary use of the skill.
Hrmm. Would be nice to be able to say something a bit more concrete about this distinction...