New math for Proficiency Bonuses

[Ironwedge]

Correct me if I am wrong, but in 3.5/Pathfinder 1e, using a weapon that you are not proficient with was a -4 penalty. In 2e, being untrained is level -2, which, you will always have at least one level, so really it's a -1 or less when applied. My question is why not stick to the -4 in some way. Maybe Untrained should be level -4? In this way, it will always be a -3 or less when applied to a roll. Additionally, I think Legendary should be more than a +1 away from Master, and really set itself apart from the rest of the pack. I think this could be done by making Legendary Level +4, like the opposite of Untrained, which I think should be Level -4. In play, untrained just seemed too close to trained and expert. And I think keeping some numbers from Pathfinder 1 is good, especially when they make sense and won't break the game when used. What do you think?

[GLD]

I think the proficiency mechanic is both my favourite and least favourite part about this game. There's a lot of potential, but it seems misdirected. You're right that the gap should be widened between proficiency tiers.

I think that the tiers should be a lot more impactful overall. Achieving master or legendary status should really feel like a big step.

[master_marshmallow]

The math behind proficiency does seem to break when actually playing the game, the level -2 resulting in a -1 modifier is more than likely intentional, and I do agree that it's pretty rough. I think the increments by which proficiency changes things should be higher, by margins of at least 2 given how the rest of the math works.

Untrained -5 (resulting in a -4 when level is applied)
trained 0
Expert +1
Master +3
Legendary +5

This way the actual spread of untrained to legendary actually mathematically changes you from a success to a critical success, it also makes combat a lot smoother when you need to land crits to justify taking crappy feats because there are no good ones (expect Double Slice and its line, which is the defacto best melee style in the game.)

[pauljathome]

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

One of the stated goals of the new system is to keep all the characters in a party within "reasonable" distance from each other.

I certainly agree that PF1 has a major problem. The character maximized in just about anything (to hit, AC, diplomacy, stealth) etc quickly becomes absurdly better than the character who dabbles in that same thing. My Level 5 bard diplomancer makes the fighters efforts to have a decent diplomacy skill absolutely irrelevant. GMs have a choice of building monsters capable of hitting the tank (which makes AC almost irrelevant for anybody else) or letting the tank be nearly unhittable.

At first blush the difference in modifiers seems too low to me too. BUT, those are the modifiers that Paizo thinks work AFTER doing some playtesting. I certainly want to playtest them myself since I don't always agree with Paizo on such things BUT I'm definitely willing to take their playtested values over my first impressions. At least until I get actual experience with them

[Staffan Johansson]

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

I agree. My instinct says that they are either too low, or there should be more secondary benefits to increasing your proficiency - not just gated behind skill feats, but inherent to increasing your proficiency. Perhaps Survival would let you forage for more people, Stealth would let you move faster, various knowledge stuff would give you more information, and so on.

Or perhaps the crit range should expand on higher proficiency levels - instead of +10/-10 for critical success/failure, it could be +9/-11 for Expert, +8/-12 for Master, and +7/-13 for Legendary. That might be more cognitive load than expected though. It's easy to remember that if the DC is 13, 23 is a crit. But if the threshold for a crit can be anywhere from 20 to 23, it's not so clear anymore, and it can be particularly hard to remember what applies to this particular skill.

But as pauljathome says, that's the kind of thing that's best left to actual playtesting.

[master_marshmallow]

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

One of the stated goals of the new system is to keep all the characters in a party within "reasonable" distance from each other.

I certainly agree that PF1 has a major problem. The character maximized in just about anything (to hit, AC, diplomacy, stealth) etc quickly becomes absurdly better than the character who dabbles in that same thing. My Level 5 bard diplomancer makes the fighters efforts to have a decent diplomacy skill absolutely irrelevant. GMs have a choice of building monsters capable of hitting the tank (which makes AC almost irrelevant for anybody else) or letting the tank be nearly unhittable.

At first blush the difference in modifiers seems too low to me too. BUT, those are the modifiers that Paizo thinks work AFTER doing some playtesting. I certainly want to playtest them myself since I don't always agree with Paizo on such things BUT I'm definitely willing to take their playtested values over my first impressions. At least until I get actual experience with them

Here's the thing about this whole "you need to play test it first before you make judgement" argument: it's not true. How do I expect to build and play a character if I don't read and try and understand how the game is meant to be played?

The entire game is a thought exercise, with some dice added in for variety. If we can't get into the game because the math is designed so poorly that it's not fun to try and build a character, then why bother?

[Vition]

Like many here, I also feel like the difference is too insignificant. I actually think the numbers proficiency should give should change depending on what the are actually being applied to. For me it comes down to a trained person should be reliably better than someone who is untrained, while an expert should be [i]reliably[i] better than someone who is simply trained, etc...

As for what's appropriate, testing should be done, my gut says that skill proficiencies should have a greater distinction than armor, weapon, or saves.

My thoughts for skills:
Untrained: -4 (that whole +1 for level thing doesn't matter here, as everyone gets it so it becomes a wash and whatever penalty is given is the penalty)
Trained: 0
Expert: +3
Master: +6
Legendary: +9

Armor, weapons, saves:
Untrained: -2
Trained: 0
Expert: +2
Master: +4
Legendary: +6
(the -2/0/+1/+3/+5 is also tempting here)

[thflame]

I'd like to see proficiency's effect increased with the level bonus decreased to compensate.

Something like d20 + attribute Modifier + 1/2 level + Proficiency

Untrained = -3
Trained = 0
Expert = +3
Master = +6
Legendary = +9

I don't like it when the "Legendary" character has a 30% chance to perform worse on a basic skill check than an Untrained character of the same level. (Ignoring nat 1s and nat 20s)

The above suggestion would have a Legendary character succeed on a check over an Untrained character of the same level 0.7% of the time, which I think is a good number for "the idiot randomly knows something the professional doesn't" funny moments. (Again, ignoring nat 1s and nat 20s.)

[Paradozen]

master_marshmallow wrote:

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

One of the stated goals of the new system is to keep all the characters in a party within "reasonable" distance from each other.

I certainly agree that PF1 has a major problem. The character maximized in just about anything (to hit, AC, diplomacy, stealth) etc quickly becomes absurdly better than the character who dabbles in that same thing. My Level 5 bard diplomancer makes the fighters efforts to have a decent diplomacy skill absolutely irrelevant. GMs have a choice of building monsters capable of hitting the tank (which makes AC almost irrelevant for anybody else) or letting the tank be nearly unhittable.

At first blush the difference in modifiers seems too low to me too. BUT, those are the modifiers that Paizo thinks work AFTER doing some playtesting. I certainly want to playtest them myself since I don't always agree with Paizo on such things BUT I'm definitely willing to take their playtested values over my first impressions. At least until I get actual experience with them

Here's the thing about this whole "you need to play test it first before you make judgement" argument: it's not true. How do I expect to build and play a character if I don't read and try and understand how the game is meant to be played?

The entire game is a thought exercise, with some dice added in for variety. If we can't get into the game because the math is designed so poorly that it's not fun to try and build a character, then why bother?

See, that is your opinion on the matter. Paizo seems to have a different one.

Vic Wertz wrote:

• Tell us about your actual game play. Theory is all well and good, but everybody’s got theories, and we’ve probably heard most of them already. Tell us how things are actually working in play, not how you think things will work.

[master_marshmallow]

Paradozen wrote:

master_marshmallow wrote:

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

One of the stated goals of the new system is to keep all the characters in a party within "reasonable" distance from each other.

I certainly agree that PF1 has a major problem. The character maximized in just about anything (to hit, AC, diplomacy, stealth) etc quickly becomes absurdly better than the character who dabbles in that same thing. My Level 5 bard diplomancer makes the fighters efforts to have a decent diplomacy skill absolutely irrelevant. GMs have a choice of building monsters capable of hitting the tank (which makes AC almost irrelevant for anybody else) or letting the tank be nearly unhittable.

At first blush the difference in modifiers seems too low to me too. BUT, those are the modifiers that Paizo thinks work AFTER doing some playtesting. I certainly want to playtest them myself since I don't always agree with Paizo on such things BUT I'm definitely willing to take their playtested values over my first impressions. At least until I get actual experience with them

Here's the thing about this whole "you need to play test it first before you make judgement" argument: it's not true. How do I expect to build and play a character if I don't read and try and understand how the game is meant to be played?

The entire game is a thought exercise, with some dice added in for variety. If we can't get into the game because the math is designed so poorly that it's not fun to try and build a character, then why bother?

See, that is your opinion on the matter. Paizo seems to have a different one.

Vic Wertz wrote:

• Tell us about your actual game play. Theory is all well and good, but everybody’s got theories, and we’ve probably heard most of them already. Tell us how things are actually working in play, not how you think things will work.

And their products will suffer as a result, especially when most of the theory comes from walking away from session zero with no progress being done because my players don't want to play these characters because of how bad the system is.

I may not even get to play test because of this now.

[Voss]

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

Why? It is a math problem. It's a simple comparison of bonuses vs DCs.

[ChibiNyan]

The non-proficiency adds up to -1 at level 1, yes. But compared to trained, there's still a -3 gap with, say, a weapon you're actually proficient with. A -3 difference in Pathfinder 2 is at least equal to the -4 difference in PF1, if not greater. The goalpost is not at 0 anymore, it's still a lot specially since characters get Expert quite fast.

[master_marshmallow]

ChibiNyan wrote:

The non-proficiency adds up to -1 at level 1, yes. But compared to trained, there's still a -3 gap with, say, a weapon you're actually proficient with. A -3 difference in Pathfinder 2 is at least equal to the -4 difference in PF1, if not greater. The goalpost is not at 0 anymore, it's still a lot specially since characters get Expert quite fast.

Right, but that difference doesn't show up until level 15, so I guess we'll here from you in 45 weeks?

[FitzTheRuke]

I've played two sessions now and what I've discovered is that every +1 to a d20 roll is worth at least double what you imagine it to be, because of the four-rank success system. I know, I didn't believe it when I read it either. We imagine a +1 to "just" give you a 5% chance of success. In practice, it stops you from fumbling, makes you succeed, and makes you crit. There are THREE rolls on the d20 where it makes a difference. Not one.

[master_marshmallow]

FitzTheRuke wrote:

I've played two sessions now and what I've discovered is that every +1 to a d20 roll is worth at least double what you imagine it to be, because of the four-rank success system. I know, I didn't believe it when I read it either. We imagine a +1 to "just" give you a 5% chance of success. In practice, it stops you from fumbling, makes you succeed, and makes you crit. There are THREE rolls on the d20 where it makes a difference. Not one.

The problem more comes from things like saying a feat like Dual-Handed Strike is good because you have a better chance to crit with it thanks to the new proficiency system.

Math says this is false, observe:

At the cost of an action, I essentially get to make a two-handed attack with a +2 to the damage. We're being told the numbers need to be smaller to accommodate for how impactful crits are, but when the odds of that being a crit are only 10-15% better than say making a second attack whose chance to hit is only 25% worse, I have to compare is that 10% chance to crit and do 4 damage better than the 15% difference in chance to hit to possibly land another attack doing a bare minimum of dice+STR which more than likely will always exceed 4 damage.

Proficiency needs a wider distribution because of the way the game is being designed around the new +10/-10 system, otherwise these action taxes will prove to be useless feats and options by comparison to simply not using them and playing a blank character.

As written, proficiency doesn't solve this problem.

[Unicore]

One issue I am having with the proficiency system is that the different levels are supposed to feel like more than numbers, but they don't really yet. The gate is real with some skills (especially at very high levels) but it takes level 4 or 7 before the difference between those levels really feel meaningful (trained to expert and expert to master), and even then. Characters just don't get enough options for proficiency = feat gate to feel as relevant.

My proposal would be to tag on a rider to each proficiency level that is automatic. Some specific proficiencies get this (like class boosts to saves) but if every proficiency level gave you one free reroll a day when making those proficiency checks, it would already feel massively more significant.

[GLD]

Unicore wrote:

One issue I am having with the proficiency system is that the different levels are supposed to feel like more than numbers, but they don't really yet. The gate is real with some skills (especially at very high levels) but it takes level 4 or 7 before the difference between those levels really feel meaningful (trained to expert and expert to master), and even then. Characters just don't get enough options for proficiency = feat gate to feel as relevant.

My proposal would be to tag on a rider to each proficiency level that is automatic. Some specific proficiencies get this (like class boosts to saves) but if every proficiency level gave you one free reroll a day when making those proficiency checks, it would already feel massively more significant.

One free reroll per tier of proficiency seems like a lot, but I do like the direction you're going.

Expert = Free reroll a day.
Master = 1 Proficiency surge a day, where you double the prof bonus for a single check.
Legendary = Turn one success into a crit per day.

Those are off the top of my head and probably not the best suggestions, but a nice benefit to really emphasize mastery of a skill for each tier would cool.

[pauljathome]

Voss wrote:

pauljathome wrote:

I'm strongly of the opinion that what the proficiency modifiers should be is one of those things that can ONLY be determined by actually PLAYING and not by theory crafting.

Why? It is a math problem. It's a simple comparison of bonuses vs DCs.

I understand the math quite well. What I do NOT know is how my diplomancer bard AND my somewhat diplomatic fighter are actually going to feel like in play across multiple levels. Will my bard be sufficiently better than the fighter AND will the fighter feel sufficiently diplomatic to validate the investments they both make in diplomacy?

Obviously, even AFTER playtesting I won't COMPLETELY know. But I'll have one HECK of a better idea than I do now.

Also, pragmatically, I expect my feedback to be taken a whole lot more seriously by Paizo if is based on actual experience.