Legendary bonus not so legendary


Pathfinder Playtest


1 person marked this as a favorite.

Is it just me, or is a meager +3 not that impressive? By fourth level it is the smallest part of your total bonus, after your level and skill ranks. That +3 is lost in the shuffle by 10th level.


6 people marked this as a favorite.

I think the majority of the benefit of legendary proficiency will be in additional skill uses or specialized feats not the +3 bonus. That probably holds true for everything now.

Shadow Lodge

3 people marked this as a favorite.
Star Dragon Caith wrote:
Is it just me, or is a meager +3 not that impressive? By fourth level it is the smallest part of your total bonus, after your level and skill ranks. That +3 is lost in the shuffle by 10th level.

If your modifier was the be all and end all, I would agree. However, it appears that being "Legendary" in a skill is a gateway to awesome feats that lesser-proficient combatants cannot access. I'm hoping that modifiers are reined in so other aspects of the game can impact power level and success.


2 people marked this as a favorite.

Just as you need "trained" to even attempt some things, for at least some skills you'll need "legendary" for the same, beyond the feat options it opens up. The one we know for sure is that you can't craft legendary items (such as weapons with a +3 item bonus, and which quality is a prereq for some enchantements) without being legendary in crafting. There might also be legendary locks to pick or legendary distances to try to jump.

Paizo Employee

8 people marked this as a favorite.
Herremann the Wise wrote:
Star Dragon Caith wrote:
Is it just me, or is a meager +3 not that impressive? By fourth level it is the smallest part of your total bonus, after your level and skill ranks. That +3 is lost in the shuffle by 10th level.
If your modifier was the be all and end all, I would agree. However, it appears that being "Legendary" in a skill is a gateway to awesome feats that lesser-proficient combatants cannot access. I'm hoping that modifiers are reined in so other aspects of the game can impact power level and success.

The math is also much tighter in the new game, specifically so it doesn't end up with the "Player A always auto-succeeds on these things and Player B always autofails". By keeping the numbers tight, you make everyone more heroic while the 4 degrees of success system means that there is still a significant degree of difference represented by even a variance of +/-5. If you're talking about attack rolls, for example, a Legendary fighter vs. a Trained wizard is going to have a higher chance of critting, a higher chance of hitting at all, a lower chance of failing, and a lower chance of critically failing. Despite the Fighter having essentially the benefits of Greater Weapon Focus and Improved Critical (all weapons) tagged onto his chassis, the Wizard could still potentially have a reasonable chance of landing some kind of weapon attack, which means that arcane-archer-type characters get to be more valid in the new system without needing a bunch of add-ons. That also leads into supporting characters playing into their strengths; the fighter with his higher proficiency is going to be more willing to lean into additional attacks, while the arcane archer wizard might only want to make one weapon attack a round and then use the other actions for movement or spellcasting. A battle cleric will probably want to use an action or two to buff before making a single attack. So on and so forth.

The tight math and new action economy mean that a bonus of +3 goes a lot farther and actually supports a broader array of playstyles and characters than you can do effectively in the current edition, particularly at lower levels where many characters don't have many (if any) swift/immediates to round out various playstyles. The difference between classes isn't "Should I even be trying to do this thing?" it's "What kind of strategies do I have that will be effective here?". And each class will have more answers to that question with less fiddly bits to try and track or stockpile.


As an example, one spell entry from the PaizoCon Banquet was Discern Lies, a 4th level spell that gave a +4 bonus to perception checks to determine lies. If a +4 bonus is an example of the kind of bonus a 4th level spell might give a character, you can imagine how closely-structured the math is to keep rampant bonus accrual in check. Much like Starfinder and D&D5, this edition of Pathfinder is going to be noted less by how a character pumps their bonuses, and more by the cool abilities and options a given character can utilize.

Personally, I like it more that way, anyway, because just gaming the numbers treadmill gets boring after a while, and every bard with a +20 Diplomacy and 4th level and Barbarian with a +15 to hit at 5th level starts feeling pretty cardboard to play, because the most special thing about them ARE their numbers.

Liberty's Edge

1 person marked this as a favorite.

Discern Lies also only has a duration of 10 minutes. So a +4 to one use of Perception (not all uses, just one) for 10 minutes is a 4th level spell.

That's pretty restrictive.


Deadmanwalking wrote:

Discern Lies also only has a duration of 10 minutes. So a +4 to one use of Perception (not all uses, just one) for 10 minutes is a 4th level spell.

That's pretty restrictive.

If anything else, this may give us a good idea of how intent Paizo is to keep magic from stepping on skill-users' toes. I'm now dying to know what "Knock" looks like:

"Arcana check in place of Thievery check, for one check, as if Trained?"
"...As if Expert?"

Liberty's Edge

2 people marked this as a favorite.
ENHenry wrote:

If anything else, this may give us a good idea of how intent Paizo is to keep magic from stepping on skill-users' toes. I'm now dying to know what "Knock" looks like:

"Arcana check in place of Thievery check, for one check, as if Trained?"
"...As if Expert?"

Well, they've also said that for the 'replace' spells (like knock or pass without trace) they usually give two options, one a replacement, the other a direct bonus to someone who actually has the skill.

So Knock could be "The target may use Arcana at your full bonus for one Thievery check to open a lock, or may do so with Thievery with a +3 conditional bonus."

Or something like that. So you're better off casting it to buff the Rogue than use it yourself, though if you don't have a Rogue it will also work to solve the issue on its own.


1 person marked this as a favorite.

If you do "chains" of skill rolls, you really do want to be good enough that some parts of the chain are auto-successes, or you will have really low probability of doing the full chain.

A very simple example: "I jump over the fence, sneak over the lawn, climb up the wall and listen at the window." That means a jump roll, a stealth roll, a climb roll and a listen roll - and if you fail at either one of these four, you fail to listen in at the secret meeting taking place behind said window.

Lets say you are not good enough to auto-succeed on any of these rolls. For simplicity's sake say that you succeed on a 5+ on all of them - i.e. an 80% chance. Well, then the chance of succeeding at the whole chain is 0.8^4, which is a 40% chance - not very good odds if failing is of any consequence.

Let's add the three rolls needed to make the trip back to the starting point. That means a jump roll, a stealth roll, a climb roll, a listen roll, a climb roll, a stealth roll and a jump roll. The odds of succeeding at 7 rolls in a row, each one at 80%, is a dismal 21%...

The group I play in tends to use skills *a lot*, especially when you do some kind of heist or intrigue play, and these kinds of chains show up all the time.

I have not seen anything about handling chained skill rolls like these so far in PF2. In PF1 we've handled it by simple being good enough to auto-succeed at select skills. (Remember that a 1 is *not* a fail or fumble at skill-rolls in PF1!)

Liberty's Edge

1 person marked this as a favorite.

Well, for one thing there's the Assurance Skill Feat. That lets you flat-out auto-succeed at checks with a low enough DC. That DC is 10 if you're only Trained, 15 if you're an Expert, and around 30 if you're Legendary. You just take the simple success instead of rolling.

In your example Assurance (Athletics) would cut the number of rolls to two (Stealth and Perception), since it's used for both climbing and jumping. Three total if you require one for sneaking away as well.

That helps quite a bit, and is a fairly likely Skill Feat for a traditional Rogue (for the climbing).

There's probably also advice not to force chained rolls like that. I certainly wouldn't require that many rolls for the described action series in my games (I'd probably require a single Climb roll, a single Stealth roll, and a single Perception roll), and games can effect things quite a lot by simply noting when it's appropriate to call for rolls.


1 person marked this as a favorite.
Mats Öhrman wrote:

If you do "chains" of skill rolls, you really do want to be good enough that some parts of the chain are auto-successes, or you will have really low probability of doing the full chain.

A very simple example: "I jump over the fence, sneak over the lawn, climb up the wall and listen at the window." That means a jump roll, a stealth roll, a climb roll and a listen roll - and if you fail at either one of these four, you fail to listen in at the secret meeting taking place behind said window.

Lets say you are not good enough to auto-succeed on any of these rolls. For simplicity's sake say that you succeed on a 5+ on all of them - i.e. an 80% chance. Well, then the chance of succeeding at the whole chain is 0.8^4, which is a 40% chance - not very good odds if failing is of any consequence.

Let's add the three rolls needed to make the trip back to the starting point. That means a jump roll, a stealth roll, a climb roll, a listen roll, a climb roll, a stealth roll and a jump roll. The odds of succeeding at 7 rolls in a row, each one at 80%, is a dismal 21%...

The group I play in tends to use skills *a lot*, especially when you do some kind of heist or intrigue play, and these kinds of chains show up all the time.

I have not seen anything about handling chained skill rolls like these so far in PF2. In PF1 we've handled it by simple being good enough to auto-succeed at select skills. (Remember that a 1 is *not* a fail or fumble at skill-rolls in PF1!)

They have also promised that DC don't scale based on level, even as our proficiency does. Meaning if jumping this was DC 10 at level 1, the same fence will still be DC 10 at level 20.

So you will reach points where there's pretty much no chance of failure on specific tasks, but it depends on how tough the task is. If you want to reach that point before level scaling gets you there, that seems to be what the Assurance feat is for.


1 person marked this as a favorite.
Captain Morgan wrote:
Mats Öhrman wrote:

If you do "chains" of skill rolls, you really do want to be good enough that some parts of the chain are auto-successes, or you will have really low probability of doing the full chain.

A very simple example: "I jump over the fence, sneak over the lawn, climb up the wall and listen at the window." That means a jump roll, a stealth roll, a climb roll and a listen roll - and if you fail at either one of these four, you fail to listen in at the secret meeting taking place behind said window.

Lets say you are not good enough to auto-succeed on any of these rolls. For simplicity's sake say that you succeed on a 5+ on all of them - i.e. an 80% chance. Well, then the chance of succeeding at the whole chain is 0.8^4, which is a 40% chance - not very good odds if failing is of any consequence.

Let's add the three rolls needed to make the trip back to the starting point. That means a jump roll, a stealth roll, a climb roll, a listen roll, a climb roll, a stealth roll and a jump roll. The odds of succeeding at 7 rolls in a row, each one at 80%, is a dismal 21%...

The group I play in tends to use skills *a lot*, especially when you do some kind of heist or intrigue play, and these kinds of chains show up all the time.

I have not seen anything about handling chained skill rolls like these so far in PF2. In PF1 we've handled it by simple being good enough to auto-succeed at select skills. (Remember that a 1 is *not* a fail or fumble at skill-rolls in PF1!)

They have also promised that DC don't scale based on level, even as our proficiency does. Meaning if jumping this was DC 10 at level 1, the same fence will still be DC 10 at level 20.

So you will reach points where there's pretty much no chance of failure on specific tasks, but it depends on how tough the task is. If you want to reach that point before level scaling gets you there, that seems to be what the Assurance feat is for.

But will that work in practice? Sure the DC to jump a fence or break the simple wooden door may not change, but will we still deal with those challenges? In my experience, the jump lengths get longer and the doors worth opening get made of more adamantine.


Of course you're going to face tougher challenges as you level up. You'd get pretty bored if you kept facing the same old challenges...


2 people marked this as a favorite.
Khudzlin wrote:
Of course you're going to face tougher challenges as you level up. You'd get pretty bored if you kept facing the same old challenges...

And if the challenges scale to your level, then in effect the DC does.


1 person marked this as a favorite.

My biggest concern with Legendary skills is that they will put restrictions on what non-Legendary options people have.

For instance, Legendary Medic allows a PC to use heal to remove certain negative conditions in an hour.

This of course implies that without Legendary proficiency in Heal, you will be unable to remove negative conditions in an hour.

We can stretch this interpretation further, obviously. Can someone Trained in Heal remove these negative conditions, but take longer? Or are they simply unable to remove them at all?

A conservative interpretation might make the latter conclusion, which in turn curtails characters' options.

This is already a problem in Pathfinder, where the existence of certain feats causes restrictions on what PCs without those feats are able to do.

Liberty's Edge

Mekkis wrote:
We can stretch this interpretation further, obviously. Can someone Trained in Heal remove these negative conditions, but take longer? Or are they simply unable to remove them at all?

Depends on the condition. Someone with Medicine cannot normally fix having been Enervated or permanently Blinded, no. However, they explicitly can treat poisons and diseases (it's mentioned in the Blog).

In short, I think you're worrying about the wrong thing when you talk about 'conservative interpretation'. What skills do seems like it's going to be very explicit, without much need for interpretation of the sort that causes the issues you hypothesize.


Maka Nashota wrote:
Captain Morgan wrote:
Mats Öhrman wrote:

If you do "chains" of skill rolls, you really do want to be good enough that some parts of the chain are auto-successes, or you will have really low probability of doing the full chain.

A very simple example: "I jump over the fence, sneak over the lawn, climb up the wall and listen at the window." That means a jump roll, a stealth roll, a climb roll and a listen roll - and if you fail at either one of these four, you fail to listen in at the secret meeting taking place behind said window.

Lets say you are not good enough to auto-succeed on any of these rolls. For simplicity's sake say that you succeed on a 5+ on all of them - i.e. an 80% chance. Well, then the chance of succeeding at the whole chain is 0.8^4, which is a 40% chance - not very good odds if failing is of any consequence.

Let's add the three rolls needed to make the trip back to the starting point. That means a jump roll, a stealth roll, a climb roll, a listen roll, a climb roll, a stealth roll and a jump roll. The odds of succeeding at 7 rolls in a row, each one at 80%, is a dismal 21%...

The group I play in tends to use skills *a lot*, especially when you do some kind of heist or intrigue play, and these kinds of chains show up all the time.

I have not seen anything about handling chained skill rolls like these so far in PF2. In PF1 we've handled it by simple being good enough to auto-succeed at select skills. (Remember that a 1 is *not* a fail or fumble at skill-rolls in PF1!)

They have also promised that DC don't scale based on level, even as our proficiency does. Meaning if jumping this was DC 10 at level 1, the same fence will still be DC 10 at level 20.

So you will reach points where there's pretty much no chance of failure on specific tasks, but it depends on how tough the task is. If you want to reach that point before level scaling gets you there, that seems to be what the Assurance feat is for.

But will that work in practice? Sure the DC to...

Some challenges will scale, but not all. You can see this in the way APs are written. Every now and then an encounter or obstacle is pretty trivially solved compared to other bits in the adventure.

I've been reading book 6 of an AP, and depending on circumstances the skill DCs can be as low as 15 or as high as 35 within the same section.

Edit: Actually, it can probably go 38+ for some of these.

And again, challenges can always keep scaling to your numbers, no matter how high they get. That's adventure design, not the rules themselves. You can never reach a point where you auto succeed on anything the GM throws at you in PF1, but you can reach the point where the things thrown at you would be impossible for anyone else.


Mekkis wrote:

My biggest concern with Legendary skills is that they will put restrictions on what non-Legendary options people have.

For instance, Legendary Medic allows a PC to use heal to remove certain negative conditions in an hour.

This of course implies that without Legendary proficiency in Heal, you will be unable to remove negative conditions in an hour.

We can stretch this interpretation further, obviously. Can someone Trained in Heal remove these negative conditions, but take longer? Or are they simply unable to remove them at all?

A conservative interpretation might make the latter conclusion, which in turn curtails characters' options.

This is already a problem in Pathfinder, where the existence of certain feats causes restrictions on what PCs without those feats are able to do.

That very facet is what prevents the concern that many people have about "the untrained level 20 fighter is better at stealth than the highly-trained level 1 rogue." Gating certain powerful uses of a given skill behind being a legend/master/expert or not is important to denote the level of effort someone has gone through to specialize. Since Paizo has gone the route of "everyone gets proficiency equal to level", then a necessary component of that is to restrict access to stuff other than basic use without proper training.

I'm unclear how that would be any different from saying, "you must be a spellcaster to cast spells" or "you must be a cleric (or similar class) to channel energy" which is what all versions of D&D have done.

Sovereign Court

2 people marked this as a favorite.

I think it is an extremely good thing to curb down the mathematical creep of the endless bonuses.

Way less calculation and more realist goals. YAY !


It's probably far more valuable than it looks even if you drop Skill Feats. Exactly how much more will depend on how widespread Critical Successes and Critical Failures are vis-a-vis the PF2 Skill system


ENHenry wrote:
That very facet is what prevents the concern that many people have about "the untrained level 20 fighter is better at stealth than the highly-trained level 1 rogue."

Well, that and the fact that a Level 1 character can't really be considered 'highly trained' by any reasonable measure. At best, they might have more theoretical knowledge or practice, but lack experience using any of it in practical terms.


Crayon wrote:
ENHenry wrote:
That very facet is what prevents the concern that many people have about "the untrained level 20 fighter is better at stealth than the highly-trained level 1 rogue."
Well, that and the fact that a Level 1 character can't really be considered 'highly trained' by any reasonable measure. At best, they might have more theoretical knowledge or practice, but lack experience using any of it in practical terms.

And that this is what people are getting upset with, despite with BAB being a thing its been in the game since forever. The level 20 Wizard is better at fighting that the level 1 Fighter. BAB is really just skill points that are automatically invested in the Fight Stuff skill afterall.


I think another point at the materies of skill one should consider is the +1 per level

I dont know the gateway level so i just make some up for the example

lets say you are at level 5 and get expert
that makes +6 under the cut - +5 from your experience and another one from a few special tricks you learned, which also enable you to do more stuff
next step at lets say 10 where you become master - which makes +12 under the cut, which is a lot more then most people will get in their lifetimes


1 person marked this as a favorite.

I want to also emphasize the idea that it's not the bonus, but what comes with the rank that is really important.

A good example would be Evasion, which is just greater proficiency in Reflex saves. As a Master in Ref, a Success is always a Critical Success, a Failure is a success and a Critical Failure is just a Failure. Never taking double damage from Fireball is worth TONS more than just +2 to reflex saves.

This is also born out with the fact that at the same level, any given character will have no more than +16 higher on a roll than any other (stat at 10 vs 22, untrained -2 vs legend +3, +5 skill boost item)


Seisho wrote:

I think another point at the materies of skill one should consider is the +1 per level

I dont know the gateway level so i just make some up for the example

lets say you are at level 5 and get expert
that makes +6 under the cut - +5 from your experience and another one from a few special tricks you learned, which also enable you to do more stuff
next step at lets say 10 where you become master - which makes +12 under the cut, which is a lot more then most people will get in their lifetimes

I think the skills' levels have come up pretty well, though I don't remember where they specifically came up. Expert: 2, Master: 7, Legendary: 13

So looking at it that way... at the level you can become an Expert, if you take that option you are 50% better than others of your level who didn't take that option. When Master unlocks you're a bit less than 1/3 better than Trained, and at Legendary a bit less than 1/4 better than Trained. And that is just numerically speaking, not counting all the other stuff you get for those higher proficiencies.


We may have already meantioned this somewhere in the thread, but the four tiers of success also play a role here. We have some pretty good ideas of how this works for attacks and spells, but very little on how it interacts with skill checks. What does a critical success in a stealth check look like? I have no idea, but it is probably better than a normal success. Similarly, critical failures on stealth checks will probably be worse than just normal failing.


I'd imagine it would probably depend on the skill, and possibly what skill feats you have. For Stealth for instance, I could see a critical success letting you move at full speed rather than 1/2. Likewise with everything that has a 'speed'. Critical diplomacy could move their attitude two steps. Critical intimidate either does the same for influence (or maybe increases the duration, since that was a concern with influential intimidate in PF1e), or increases the severity or duration for a Demoralize. Critical [Knowledge] give more information. Stuff like that basically. And then Skill Feats give more options, or else state what happens when they give more skill uses.

Community / Forums / Paizo / Archive / Pathfinder / Playtests & Prerelease Discussions / Pathfinder Playtest / Legendary bonus not so legendary All Messageboards