Critical Hits and Critical Failures

Friday, March 30, 2018

In the Pathfinder Playtest Rulebook, when you roll your d20, there's more than just success and failure on the line. You can also critically succeed or critically fail at a variety of checks, from attack rolls, to saving throws, to skill checks and beyond. Rules like these have always been a part of Pathfinder—for example, if you fail a Climb check by 5 or more you fall, and if you fail a Disable Device check by 5 or more you set off the trap—but they are uncommon and not universally applied. In the playtest, we have a unified mechanic.

The Four Degrees of Success

In Pathfinder Second Edition, every check is rolled against a particular DC. Your roll on the d20 + your proficiency modifier + your ability modifier + all your relevant modifiers, bonuses, and penalties make up your check result. If your check result meets or exceeds the target DC, congratulations! You succeeded, and you might have critically succeeded. Otherwise, you failed. If you exceeded the target DC by 10 or more, or if you rolled a natural 20 and met or exceeded the target DC, then you critically succeeded. If your result was 10 or more lower than the target DC, or if you rolled a natural 1 and didn't meet the target DC, then you critically failed. Collectively, success, critical success, failure, and critical failure are called the four degrees of success. You can gain special abilities that increase or decrease your degree of success, often due to having a high proficiency rank. For instance, if your class grants you evasion, you get master proficiency in Reflex saves and treat any success on a Reflex save as a critical success!

Examples

Let's start with a fireball spell. In Pathfinder First Edition, if you succeed the Reflex save, you take only half damage, and evasion allows you to take no damage on a successful save. In Pathfinder Second Edition, here are the degrees of success for fireball (and many of its old friends like lightning bolt and cone of cold) in the playtest.

    Success Half damage
  • Critical Success No damage
  • Failure Full damage
  • Critical Failure Double damage

Illustration by Wayne Reynolds

Any character who critically succeeds takes no damage, and characters with evasion count their successes as critical successes. What about someone legendary at Reflex saves with improved evasion? They count critical failures as failures and thus can never suffer the deadliest effects of a Reflex save, even on a natural 1!

Not all effects list all four degrees of success. If an effect doesn't list a critical success entry, that means there is normally no special effect for critically succeeding, so you just use the result for a success. Similarly, if an effect doesn't list a critical failure entry, there is normally no special effect for critically failing, so you just use the result for a failure. If a success entry is missing, that means nothing happens on a success, and if a failure entry is missing, that means nothing happens on a failure. Let's take a look at an example that combines two of these rules: the results of a basic attack called a strike.

Success You deal damage, which equals the weapon's or unarmed attack's damage dice plus your Strength modifier if it's a melee attack, plus any bonuses.

Critical Success You deal double damage—you roll twice as many damage dice and add double the ability modifier and double any other bonuses to damage.

Let's unpack what this means. You deal damage on a success and double damage on a critical success. Since there is no failure entry, that means normally nothing happens on a failure, and since there is no critical failure entry, that means a critical failure has the same effect as a failure, so nothing happens. But the fighter might have something to say about that! The fighter can use the special certain strike action, which lets him strike with the following failure effect.

Failure Your attack deals the minimum damage. (Treat this as though you had rolled a 1 on every die.)

So with certain strike, a failed attack roll isn't actually a miss—your fighter is so skilled that you still get a glancing blow on a failure and miss entirely only on a critical failure! Meanwhile, a fighter with the twin riposte reaction can use one weapon to parry and attack with the other weapon whenever an enemy critical fails an attack roll.

Save or Lose

One of the effects of the four degrees of success that adds the most fun to the game is what this means for save or lose effects—effects where if you fail your save, you're unable to continue the fight. These sorts of effects are tricky in almost every roleplaying game, and Pathfinder is no exception. In Pathfinder First Edition, even if your character has a 75% chance of succeeding at your Will save against a mummy's paralysis, chances are pretty high that four mummies are going to paralyze you. (Thanks a lot for that encounter in your Pathfinder Society Scenario, Jason!)

It's tempting to just decide the solution is not to have save or lose effects, but that really cuts off a wide variety of classic feats, monster abilities, and spells from the game. The flip side of those abilities is that if they don't just win, chances are that many of these effects are just wasting a turn. So you either cast the save or lose spell and win, or you cast it and waste the turn. Having those as the only two outcomes is not a great proposition, and of course, players and GMs often maximize their DCs and saving throw bonuses in order to tilt the outcome to their side as much as possible.

But with four degrees of success, suddenly the design space broadens significantly. You can still suffer an effect that takes you out of the action entirely on a critical failure, and you can completely ignore the effect on a critical success. But on a failure, you suffer a powerful effect but not one that takes you entirely out of the fight in one go, and even on a success, you suffer a milder effect that is still useful for the spell's caster. For example, if you critically fail your save against dominate, you are completely under the spellcaster's control, but if you only fail, you can try to break out of the effect each round. On a successful save, you aren't controlled, but you still lose an action on your next turn as you struggle to fight off the mental commands, which could be a serious problem—you might not be able to step away before casting a spell, or have time to raise a shield.

Some Mysterious Critical Effects

I'm closing out with some cool critical effects that result from critical successes on your attack rolls or skill checks or from critical failures on your enemy's saving throws. See if you can figure out where they come from!

  • The creature is banished and can't return to your home plane by any means for 1 week.
  • The creature takes the full collapse damage and falls into a fissure.
  • The target believes the fact for an unlimited duration.
  • The target's intellect is permanently reduced below that of an animal, and it treats its Charisma, Intelligence, and Wisdom modifiers as –5. It loses all class abilities that require mental faculties, including all spellcasting. If the target is a PC, she becomes an NPC under the GM's control.
  • The creature is pushed 30 feet in the direction of the wind, is knocked prone, and takes 2d6 bludgeoning damage.
  • You grant a +4 circumstance bonus.
  • Per a failure, except the target believes that everyone it sees is a mortal enemy. It uses its reactions and free actions against these enemies regardless of whether they were previously its allies, as determined by the GM. It otherwise acts as rationally as normal and will likely prefer to attack enemies that are actively attacking or hindering it.
  • The target must succeed at a Fortitude save or die. Even on a successful save, the target is frightened 2 and must flee for 1 round.
  • Your target regains Hit Points equal to 2d10 + your Wisdom modifier.
  • Per a success, but even afterward, the target is too scared of you to retaliate against you.

Mark Seifter
Designer

More Paizo Blog.
Tags: Pathfinder Playtest Wayne Reynolds
551 to 567 of 567 << first < prev | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | next > last >>
Sovereign Court

Pathfinder Starfinder Society Subscriber

If you are invested in the skill through your skill feats, you don’t have to roll against super low DCs.


3 people marked this as a favorite.

You shouldn't pay a feat tax to not fail a DC 5 check when your result was 101.

Liberty's Edge

1 person marked this as a favorite.
Wultram wrote:
You shouldn't pay a feat tax to not fail a DC 5 check when your result was 101.

Which is one reason why I prefer the older method of 1 reducing the result by 1 step, and 20 increasing it by one step. In that case, if the DC is 5 (and your GM is still making you roll at all for some reason) and you rolled a natural 1, but got a result of 101, you'd still succeed. 101 is well over 10 higher than the DC, which would be a critical success, which the natural 1 turns into a normal success instead.

Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

Sovereign Court

Pathfinder Starfinder Society Subscriber
Wultram wrote:
You shouldn't pay a feat tax to not fail a DC 5 check when your result was 101.

If you haven't minimally invested in your skill, you should have a chance to fail. Your skill modifier is not the only form of investment.

Sovereign Court RPG Superstar 2011 Top 32

1 person marked this as a favorite.
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

Grand Lodge

Pathfinder Adventure, Adventure Path, Maps Subscriber

The developers have been playing with this rule 1/20 crit fail/succeed rule for a couple of years. They started with the other version (the worsen/improve by one step on a 1/20 system) and ended up here. To me, that implies that this isn't a problem. I suspect that it makes abilities that trigger off the crit failure more common. You might choose to say it makes them appropriately common. It possibly leads to more interesting combats.

All the hand-wringing in the world isn't going to change this before we've tested it in the heat of battle. We will have 5 to 6 months to provide feedback after we've seen and played with the rules.

Liberty's Edge

ryric wrote:
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

I didn't say easy, I said insignificant. Does the result of the roll matter for the continuation of the story? If yes, then roll, if not, then don't. Making a nail? Not at all important, it's trivial, no matter the DC: if it's possible for the character to succeed, just say they succeed. Making the Grim Nail of Doom, which is required to seal the lid of the coffin of the Dark Vampire of Eternal Damnation so he can never rise again? That's vitally important, and should require a roll, no matter how low the DC, even if it's just to determine whether you succeeded or critically succeeded (do I need one hit with the hammer to nail the coffin shut because the nail is perfect, or will it take three?) And if there's a result on the die that causes failure, you need to check for that, because the success or failure of this one roll is very important to how the story plays out.


1 person marked this as a favorite.
JRutterbush wrote:
ryric wrote:
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

I didn't say easy, I said insignificant. Does the result of the roll matter for the continuation of the story? If yes, then roll, if not, then don't. Making a nail? Not at all important, it's trivial, no matter the DC: if it's possible for the character to succeed, just say they succeed. Making the Grim Nail of Doom, which is required to seal the lid of the coffin of the Dark Vampire of Eternal Damnation so he can never rise again? That's vitally important, and should require a roll, no matter how low the DC, even if it's just to determine whether you succeeded or critically succeeded (do I need one hit with the hammer to nail the coffin shut because the nail is perfect, or will it take three?) And if there's a result on the die that causes failure, you need to check for that, because the success or failure of this one roll is very important to how the story plays out.

Actually you have it very much backwards. If the story depends on it then there should be no roll. There should never be a situation where "Oh, you rolled a 1? Rocks fall everyone dies."

Sovereign Court RPG Superstar 2011 Top 32

1 person marked this as a favorite.

I am very much against determining whether or not a roll is worth doing based on something as arbitrary as plot circumstance. Also, such a system doesn't work for sandbox games at all, where there might not even be a "plot" as much as there is player-driven exploration.


2 people marked this as a favorite.

On the contrary, I find "I can think of absolutely nothing interesting that would happen as a result of a failure" is a good reason to skip rolls in a sandbox style game, since those put a heavy improvisational burden on the GM to begin with.

Liberty's Edge RPG Superstar 2008 Top 32, 2011 Top 16

ryric wrote:
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

The one exception I'd make is if there's an effect in play which triggers on a 1. Such as a curse that does something bad to you, or an environmental effect that if you roll a 1 at all in the area, you're targeted by random bubbling lava, an enemy which gains a bonus if their opponent rolls a 1, etc.


1 person marked this as a favorite.
JoelF847 wrote:
ryric wrote:
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

The one exception I'd make is if there's an effect in play which triggers on a 1. Such as a curse that does something bad to you, or an environmental effect that if you roll a 1 at all in the area, you're targeted by random bubbling lava, an enemy which gains a bonus if their opponent rolls a 1, etc.

I would prefer that in-game effects not trigger off meta-considerations. Given a DC of 15, if the PC with a +15 bonus rolls a 1 for a total of 16 and the PC with a -1 penalty rolls a 2 for a total of 1, why does the bad thing happen to the PC who actually beat the DC?

Liberty's Edge RPG Superstar 2008 Top 32, 2011 Top 16

Joana wrote:
JoelF847 wrote:
ryric wrote:
JRutterbush wrote:


Of course, you still shouldn't be making players roll for insignificant checks like that in the first place, but that's beside the point.

The tricky thing there is determining what constitutes an easy check that shouldn't require a roll. It's obvious when someone has a +20 rolling against a DC5 check - but much less so when the check is DC15 or 20, especially if someone else in the same party is making the same check but with a +12.

To me, the natural threshold should be "if the PC's bonus is greater than the DC, no roll is needed." But that makes the "nat 1 = autofail" rule basically meaningless since it will never come up.

The one exception I'd make is if there's an effect in play which triggers on a 1. Such as a curse that does something bad to you, or an environmental effect that if you roll a 1 at all in the area, you're targeted by random bubbling lava, an enemy which gains a bonus if their opponent rolls a 1, etc.
I would prefer that in-game effects not trigger off meta-considerations. Given a DC of 15, if the PC with a +15 bonus rolls a 1 for a total of 16 and the PC with a -1 penalty rolls a 2 for a total of 1, why does the bad thing happen to the PC who actually beat the DC?

I guess it depends on the bad thing in question. If it's related to some form of misfortune, unluck, etc, then tying it to a roll of a 1 makes sense. If it's essentially a random event with a 5% chance of happening, it might not, but it cuts out a die roll independently for the event if you tie it to existing die rolls.

Liberty's Edge

Joana wrote:
I would prefer that in-game effects not trigger off meta-considerations. Given a DC of 15, if the PC with a +15 bonus rolls a 1 for a total of 16 and the PC with a -1 penalty rolls a 2 for a total of 1, why does the bad thing happen to the PC who actually beat the DC?

In fairness, it does happen to both of them. And, indeed, happens worse to the one who rolled a 1 after modifiers (since he critically failed) while the one who rolled a 16 merely failed.

That's not necessarily all that much better, depending on your preferences, but it's not quite as bad as you're making it out to be.


1 person marked this as a favorite.
Pathfinder LO Special Edition, Maps, Pathfinder Accessories, PF Special Edition Subscriber; Pathfinder Roleplaying Game Superscriber; Starfinder Superscriber

The overriding principle in a role playing game is that the story is paramount. I would expect any competent gm, given that principle, to ignore a die roll that says "rocks fall, everyone dies" unless that's the story he and the players are trying to tell - which is doubtful.


It's really just the balancing act of maintaining tone. I have not gone a single campaign in years without killing two hands worth of PCs, but at the same time, I try to soft fudge things so that only happens when it's meaningful.

Is it a boss / miniboss fight? Are the players making stupid decisions? Are the characters going into a situation they've been warned (directly by NPCs or contextually by other cues) is dangerous? Does the party have the clear opportunity to flee or call for help and they're not taking it? Would it be a good story moment, like a heroic sacrifice? Then sure, the dice fall where they will and that sometimes means PCs die.

Is it a battle with weak mooks and the dice are just terrible tonight? Did the characters have a really great idea / plan that plays on in-character knowledge rather than meta and would improve the story / campaign if it went off, and it's something that feels like it should have been a lock but for whatever bad luck it's just falling apart? Did the PCs win the fight and could have saved the downed party member with one more round to get to him if not for the action economy? Then I'll sometimes make the enemies dumber, not take obvious opportunities for ganking, lower check DCs without telling the players, "forget" to remind the player to mark off that last hit point of bleed.

As long as you're good at the balancing act, and PCs do die when appropriate, the players will be invested and don't have to know about the times you fudged to keep the game from going south.

Lantern Lodge Customer Service & Community Manager

Thanks for all the lively discussion so far in this thread. At this time we've decided to close up the blog discussion thread. If you have comments, questions or other things you want to post that do not fit into any currently open threads, you are welcome to start a new thread. Thanks!

551 to 567 of 567 << first < prev | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | next > last >>
Community / Forums / Archive / Pathfinder / Playtests & Prerelease Discussions / Pathfinder Playtest / Pathfinder Playtest Prerelease Discussion / Paizo Blog: Critical Hits and Critical Failures All Messageboards
Recent threads in Pathfinder Playtest Prerelease Discussion