[3.5/PF, PEACH] Solving the Binary vs Gradual Defenses Conundrum

[Maginomicon]

I've been wrestling with the conundrum of binary defenses versus gradual defenses (http://www.minmaxboards.com/index.php?topic=1533.0) for quite a while now. I wanted a robust and extensible solution to the problem rather than just cherry-picking bad apples to ban as is generally the norm here.

Let's go through the various binary defenses:

Touch/Flatfooted/AC is a binary defense that generally protects the HP gradual defense (and to a lesser extent, also protects the nonlethal damage gradual defense and sometimes protects against debilitating conditions). Nothing wrong there. A binary defense and a gradual defense working together. Beautiful.

SR/PR is a binary defense a lot like AC (but for non-mundane effects) that often is unused or underused... and even when it is used, it doesn't intrinsically protect any sort of gradual defense (except HP on occasion). It's more of an extra binary defense that's stacked onto other binary defenses. This is... not so good, but workable generally since it's an uncommon defense.

Saving Throws are binary defenses that also do not protect any sort of gradual defense except HP, and most of the time it's not worth it to use effects that allow a save and affect HP. It's generally much more viable to use effects which cause debilitating conditions of some sort. This is bad balance-wise, but sometimes tolerable... unless you have an optimizer.

Then... there are direct offensive effects that don't offer a save at all. This category is the absence of the most common non-mundane defense. ... Yeah, these are terrible, as no-save effects are generally unimpeachable because even when paired with SR or an attack roll, those defenses are easy to optimize against into oblivion. Sometimes it's not even paired with SR or an attack roll. Optimizers know this, and so generally they choose no-save effects whenever feasible.

Saving throws and no-save effects tick me right the **** off. So... I found a way to turn them into gradual defenses. The most robust and extensible way to make them gradual defenses is to leave the binary defense as-is alone for the most part and pair them up with new gradual defenses that are unique and are never explicitly mentioned by any other effect. It is critical that they don't use existing terms which can be optimized against.

Thus, four new values are added to the character sheet:

Fortitude Points (FP) which represent your physical resilience,
Reflex Points (RP) which represent your reflexes,
Will Points (RP) which represent your willpower reserves, and
Strife (I'm open to other name suggestions for this) which is like nonlethal damage but for non-mundane effects.


Save Points (SPs)

Fortitude Points, Reflex Points, and Will Points have the following properties:

They start at and have a maximum equal to 100+(5 * that save's base save bonus),
They have a minimum of 0,
An average creature encounter has FP, RP, and WP equal to "50+(5 * that save's base save bonus)" as appropriate,
A randomly-generated creature encounter has FP, RP, and WP equal to "((3d6*5)-2)+(5 * that save's base save bonus)" as appropriate,
24 hours of non-restful non-strenuous activity (such as walking) causes a creature to recover "10 + half that save's base bonus (rounded down)" to each as appropriate,
A "full rest" (usually 8 hours) that doesn't result in fatigue/exhaustion causes a creature to recover "20 + that save's base bonus" to each as appropriate, and
A "full day" rest (i.e. bedridden) that doesn't result in fatigue/exhaustion causes a creature to recover "40 + (2 * that save's base bonus)" to each as appropriate,


Every time a creature succeeds on a saving throw, deduct half of the save DC (rounded down) from the appropriate SP. Then any normal effects on a successful save resolve.

Every time a creature fails a saving throw, first deduct the entire saving throw DC from the appropriate SP. Then roll a d%.

d% > SP = Save Attempt Failed (normal effects for a failed save resolve, although see the section on strife below)
d% ≤ SP = Save Attempt Succeeded (normal effects for a successful save resolve, and treat your saving throw result as if it was equal to the save DC)


Please Note:

Admittedly, there's some extra rolls involved here because we're using a d% to determine the result, but it's not as much of a hassle in actual play as you might think. It really isn't a hassle, not in actual play.

I created save points, and I was happy with that... but really, something had to be done about offensive no-save effects in-general. Because we have a concrete value for base save bonuses, we can afford to have a point value of 0 be essentially the same as not having the save point value on your character sheet at all. That is, having FP/RP/WP left over doesn't intrinsically negate an effect, it just gives it an extra chance at being negated by making it go through a gradual defense. Strife, however, is another story, as we don't have a "base value" for no-save effects.

Strife

Strife, as mentioned above, acts kind-of like nonlethal damage but for non-mundane effects. It's bad for it to go up and good for it to go down. Strife in a sense represents stress chipping away at your ability to shrug off debilitating effects. Strife is only relevant to any effect that has a "saving throw" line (even if that saving throw line reads "none") and directly affects a creature (a magic missile directly affects a creature, while a wall of force only indirectly affects a creature).

Strife has the following properties:

It starts at and has a minimum of 0.
Any time something would remove nonlethal damage (regardless of whether the creature is immune to or has any nonlethal damage) it also removes an equal amount of strife.


Any time a creature "fails a save" (not just the saving throw, but the entire save attempt) for an effect that allowed a save, the creature takes strife equal to half of that effect's save DC (rounded down).

Any time a creature would be directly affected by a no-save effect and that creature doesn't want to be affected by that effect, the creature compares his "strife + nonlethal damage" to half of his current HP (rounded down). If this value exceeds half of his current HP, the effect goes through.

If this value doesn't exceed half of his current HP, he takes strife equal to the save DC that the effect would have if it allowed a save (10 + level + stat + modifiers) and compares his new "strife + nonlethal damage" to half of his current HP again. If this value now exceeds half of his current HP, the effect goes through. Otherwise, the effect is negated.

Please Note:

HP damage and nonlethal damage are factored into whether a no-save effect goes through. Anyone can deal HP damage or nonlethal damage, so everyone can get in on beating up a creature so that a no-save effect can go through. This synergistic behavior where the party has to work together to beat down a tough creature for a while so the "final blow" no-save effect can get through is intentional.

Strife heals very quickly (at the same rate as nonlethal damage), and certain effects remove all nonlethal damage (and thus remove all strife). Strife makes both HP damage and nonlethal damage more relevant. That's a win-win situation so far as balance is concerned.

Save Points + Strife = No more "Rocket Tag" (for the most part anyway).

Conclusion

I'm not saying this "fixes" caster-archetypes. The goal here was to level the playing field as described in that article so that non-mundanes can at least feel like they're really helping. I believe that Save Points and Strife successfully address the two-fold problems indicated in the article I linked to at the start of this post (http://www.minmaxboards.com/index.php?topic=1533.0). Save Points and Strife force the most common binary defenses to interact with gradual defenses, and they give mundanes an avenue for affecting "binary" defenses themselves.

Personally, I also allow mundanes to inflict stacking debuffs through a Called Shots system (http://www.giantitp.com/forums/showthread.php?t=300755) (thus giving mundanes more access to "binary" defenses), but that's just me.

I will admit that even with the Save Points and Strife systems in place, situations where binary defenses are irrelevant (such as battlefield control, summoning, buffs, etc.) become far more common. I'm... pretty okay with that.

I also wanted to get rid of Rocket Tag. Rocket Tag isn't good for anyone balance-wise. Rocket Tag makes encounters short and boring (especially if the villain uses it).

I don't want the caster-archetypes to be one-hit wonders that make the rest of the party feel like they're wasting their time (even unintentionally, and even if it doesn't technically work out that way when the rest of the party just "plays along"). No one should have to feel like they're playing second-fiddle to the caster-archetype. No one should have to feel like "welp let's let the caster take up all of half the glory while we sit on our hands whoopty-doo". Mundanes should not have to feel like they're relegated to clean-up duty after the caster-archetype(s) throw out a debilitating first volley.

What I want to know is: (Please speak constructively!)

Constructive tweaks for Save Points or Strife to better accomplish my stated goals.

More fitting names for "Strife" than "Strife", given what it does.

...and of course, per the topic title...

Suggestions for other ways to solve the conundrum described in the article I mentioned.

What I don't want to hear: (No thread-crapping!)

That the original article was wrong or misguided. :smallfurious:

That people should just ban "bad apple" effects on a case-by-case basis.

That an extra d% roll is too much. :smallconfused:

That adding a few more values to the character sheet is somehow a bad idea. :smallannoyed:

Whining in-general (about certain build types becoming useless, about TO going out the window in some way, etc.).

That I should have stuck to the thread(s) where I somewhere already discussed these issues and/or solutions in some capacity (this is a combination thread that refines the results of my efforts, and besides-which, that would in some cases involve thread necromancy).

[JDL]

Isn't this thread more suited to Homebrew Design?

[Sewercop]

+1 to the homebrew

[Maginomicon]

{{scrubbed}}

[ahenobarbi]

Together those seem to add a fair amount of book keeping. Also they somewhat overlap so I think it would be better to use one of them (and strife seems cleaner and more to the point).

Also why not "successful save doesn't reduce save points; failed save reduces save points. If you have positive save points after save treat it as it it was successful"? It's simpler, more similar to HP and gives more protection. I'd also make those more scarce (100+ is a lot) and faster to regenerate (with healing magic or something) so you can handle a failed save or two a fight.

[Maginomicon]

Together those seem to add a fair amount of book keeping. Also they somewhat overlap so I think it would be better to use one of them (and strife seems cleaner and more to the point).

Also why not "successful save doesn't reduce save points; failed save reduces save points. If you have positive save points after save treat it as it it was successful"? It's simpler, more similar to HP and gives more protection. I'd also make those more scarce (100+ is a lot) and faster to regenerate (with healing magic or something) so you can handle a failed save or two a fight.

They're mechanically and thematically different... intentionally.

For things that allow saving throws, having it be a simplistic "negate as long as you have points" value is wayyyyy too powerful. It's not like you can spam abilities that allow saves. You have limited ammo. There needs to be a way for your first attempts to still MAYBE work, not just fail because the target had points left (or you'll just be spinning your wheels for the most part). It would be intrinsically unfair to just make having points left mean that it negates the effect, even if you drop the number of points down to something low. That's fine for the first attack or two if the target happens to be at full strength, but that's because it takes days to recover and nothing can help you recover save points faster (this was also intentional).

Also, Strife counts upward, like nonlethal damage. It doesn't do what you think it does. Stress (from strife) makes sense to go away after a few hours. If your reflexes (from Reflex Points) are shot, you need rest.

Basically, you can't just throw any old gradual defense at this issue and say "it negates" without thinking about how fair that would be. Rudimentary complexity is needed to make it reach even a base level of fairness.

[TuggyNE]

I'm intrigued by this, since it rather closely parallels the thinking I've been doing in the last few months for my own system. The biggest difference, of course, is that one is trying to patch 3.x, and the other is a ground-up design with a somewhat different game feel.

However, beyond that, this system, while it does indeed allow a certain amount of cooperation between mundanes and casters, does so only in the case of spells that either a) do HP damage (already handled by the default system) or b) have no save at all. Save-or-X spells are somewhat nerfed, yes, but there is no real way in most cases for a mundane combatant to contribute to whittling down RP or WP, or even FP. The net result, I think, is that blasting is useless if it offers a save (since it must now go through not one but two gradual defenses), and debuffs are also useless in all cases; they can only land when the foe is largely defeated anyway.

The scaling of the new defenses, and the damage done to them, is unusually slow and static, which feels weird next to d10+Con/level HP and 3d6+1.5 Str damage. I suppose that's a concession made to the existing system though.

[Sewercop]

{{scrubbed}}

Yup, It annoys me to see it in this forum. Rationality? nope
I even saw your thread on the minmax boards earlier.
I just don`t like it forced upon me, makes me want to pick it apart in a non constructive way.

If you keep hit points, you need to call fortitude something else then resilience, its just weird. Description might as well have been hit points.

Besides all that.. I can just spam you with spells to make you go to 0 save points and hit you with anything and win.... bad

[Psyren]

{{scrubbed}}

But if you're designing homebrew (I've never heard of "strife" before, so I assume it is), that's indeed where it belongs. They're not wrong for pointing that out.

I do agree it's unfortunate that subforum gets so little traffic compared to this one; I keep meaning to go over there.

[Fax Celestis]

{{scrubbed}}

[D-naras]

All in all I agree with you and my own system actually functions in a similar fashion.

Regarding Strife: How about naming it Stress?

Another possible way to make all defences gradual can be effects similar to the Death Domain power. Specifically, all spells with a save will check the target's current hit points on a failed check (say 1d6 per caster level. if the total roll is over the target's current hp, he is affected).

[OldTrees1]

Gradual Defenses are a good idea.

However there are some things that do not sit right with me:
1) A DC 10 Reflex save per attack can eat through a Reflex defense despite being passed by Binary Defenses many many levels ago. Should 100 CR 1 traps really tax(weaken) the Reflexes of a 20th level Swordsage(say +20 Ref Save)?

2) If they reach 0 save points then it reverts to a Binary Defense.

[Person_Man]

{{scrubbed}}

I read the homebrew forum every week, have been honing my writing there for years, regularly get very helpful feedback from the Playground, and am thankful for it.

[Maginomicon]

{{scrubbed}}


All in all I agree with you and my own system actually functions in a similar fashion.

Regarding Strife: How about naming it Stress?

Another possible way to make all defences gradual can be effects similar to the Death Domain power. Specifically, all spells with a save will check the target's current hit points on a failed check (say 1d6 per caster level. if the total roll is over the target's current hp, he is affected).

I considered naming it "stress", but the word "stress" is much too ambiguous to be used for something that's a defense against non-mundane effects.

As for the 1d6 per caster level option, I like it, but when I was building the Save Points variant the biggest point of grief I got from people was the notion that there was too much dice-rolling added in. All that was added was a d%. Adding in an indeterminate number of d6 rolls would thus be problematic on its face.


Gradual Defenses are a good idea.

However there are some things that do not sit right with me:
1) A DC 10 Reflex save per attack can eat through a Reflex defense despite being passed by Binary Defenses many many levels ago. Should 100 CR 1 traps really tax(weaken) the Reflexes of a 20th level Swordsage(say +20 Ref Save)?

2) If they reach 0 save points then it reverts to a Binary Defense.

Both of these were intentional.

A level 20 swordsage has 100+(5*12)=160 Reflex points. 100 CR 1 traps will indeed eat away at his reflexes at a rate of 5 per trap, but this makes sense because doing anything that taxes your reflexes will wear you down given enough time. Besides, the "expected value" on those traps will cause him to fail the save (and thus lose RP equal to the full DC) 5 times by the time he gets through 100 traps. Even then, it's eating through a buffer.

As I said in the OP, we can afford save points being a relatively simple buffer above the original binary defense because we have a stat (base save bonus) that upgrades but is fixed between level-ups. We can't do that, for example, with caster level, save modifier, etc. because those can fluctuate wildly.

[Fax Celestis]

{{scrubbed}}

[OldTrees1]

Both of these were intentional.

A level 20 swordsage has 100+(5*12)=160 Reflex points. 100 CR 1 traps will indeed eat away at his reflexes at a rate of 5 per trap, but this makes sense because doing anything that taxes your reflexes will wear you down given enough time. Besides, the "expected value" on those traps will cause him to fail the save (and thus lose RP equal to the full DC) 5 times by the time he gets through 100 traps. Even then, it's eating through a buffer.

As I said in the OP, we can afford save points being a relatively simple buffer above the original binary defense because we have a stat (base save bonus) that upgrades but is fixed between level-ups. We can't do that, for example, with caster level, save modifier, etc. because those can fluctuate wildly.

I kinda expect characters to eventually outgrow some threats (even if it takes a 19 level spread). So I would expect a 20th level Swordsage to unfazed by 100 CR 1 traps. (Similar to how low level skills checks eventually stop being rolled by high level characters)

Note: I am in favor of the buffer(especially strife).

[Just to Browse]

A system that requires me to subtract a 2-digit number from a 3-digit number more than once a session is not a good system for most tables.
Any system that requires me to add a (potentially) 3-digit number to a (potentially) two-digit number is not a good system for most tables.
Any system that requires me to track 4 separate HP-like character resources is not a good system for most tables.
Any subsystem that encourages focus-firing a single character resource is not a good system for most tables.

A system which combines all four of those traits needs to be reworked from the ground up.

[pwykersotz]

I'm intrigued by this. Maybe I'll playtest it at one of my tables.

[Vadskye]

A system that requires me to subtract a 2-digit number from a 3-digit number more than once a session is not a good system for most tables.
Any system that requires me to add a (potentially) 3-digit number to a (potentially) two-digit number is not a good system for most tables.
Any system that requires me to track 4 separate HP-like character resources is not a good system for most tables.
Any subsystem that encourages focus-firing a single character resource is not a good system for most tables.

A system which combines all four of those traits needs to be reworked from the ground up.

Succinctly put, and completely correct.

[Maginomicon]

Succinctly put, and completely correct.

Heaven forbid we use basic addition and subtraction in Dungeons & Dragons.

Seriously. This is the bread and butter of D&D. This is one of those "Assume X is true, debate viability of Y which was derived from X" situations. That's why I said in the OP that it'd be threadcrapping to assert that the original article was misguided or wrong.

Here's an analogy of what just happened, in-essentia:
1. I asked the reader to assume fruits are good for us.
2. I then asked the reader to ponder the rationale for a fruit smoothie.
3. JtB then asserted that fruits are stupid.

You can't have a reasoned discussion if the people involved... can't... or rather, won't... pay the curtesy of accepting the basic premise behind the discussion itself. Saying that having more gradual defenses is intrinsically bad (which is what he said) is utterly unhelpful to the discourse. :smallfurious:

[Bovine Colonel]

Heaven forbid we use basic addition and subtraction in Dungeons & Dragons.

Seriously. This is the bread and butter of D&D. This is one of those "Assume X is true, debate viability of Y which was derived from X" situations. That's why I said in the OP that it'd be threadcrapping to assert that the original article was misguided or wrong.

Here's an analogy of what just happened, in-essentia:
1. I asked the reader to assume fruits are good for us.
2. I then asked the reader to ponder the rationale for a fruit smoothie.
3. JtB then asserted that fruits are stupid.

You can't have a reasoned discussion if the people involved... can't... or rather, won't... pay the curtesy of accepting the basic premise behind the discussion itself. Saying that having more gradual defenses is intrinsically bad (which is what he said) is utterly unhelpful to the discourse. :smallfurious:

I would argue that requiring people to accept the basic premise behind a discussion in order to proceed is unhelpful to tying that discussion to any practical application. If I were to draw a conclusion from the idea that...I dunno, the sky is brown, the conclusion could be perfectly valid given the premise but if the premise is incorrect it's really not much use. Since most D&D discussions have some sort of practical application (even if it's just "hey look I can do a million damage and therefore get bragging rights") I would further argue that such discussions are not actually the bread and butter of D&D.

EDIT: Also, as far as I can tell JtB's post doesn't dispute the given article but focuses on the amount of bookkeeping required by your solution. By your analogy it's more like disputing the fruit smoothie for reasons of...I dunno, cost?

EDIT 2: Just to make it clear, I have not in this post implied that you believe the sky is brown or any such nonsense, nor do I actually believe that you do.

[Just to Browse]

That would be a terrible and inaccurate analogy. Please look at the word digit repeated multiple times in my post. Most of D&D's math almost never hits 3 digits (HP is the exception), and almost of all of it's math involves comparing two 2-digit numbers or subtracting/adding a one-digit number from/to a two-digit number. Adding four HP-like resources is above and beyond what D&D currently uses, and arguing that they're "the bread and butter of D&D" is intellectually dishonest.

Allow me to provide you with an alternative analogy:
Magicomicon recognizes that seeds in watermelons are a problem.
No one disagrees
Maginomicon says you can deal with this by smashing your head against watermelons until all the seeds fall out
I say that's a terrible idea, because it is.
Maginomicon tells me I hate fruit.

We haven't even gotten to the math of your solution (it doesn't work) or whether it even solves the problem at hand (it does not), because that's not relevant yet. If your solution is so clunky that 99.9% playgroups wouldn't even want to use it, it doesn't matter if it fixes the math because no one will use it.

[Maginomicon]

I would argue that requiring people to accept the basic premise behind a discussion in order to proceed is unhelpful to tying that discussion to any practical application. If I were to draw a conclusion from the idea that...I dunno, the sky is brown, the conclusion could be perfectly valid given the premise but if the premise is incorrect it's really not much use. Since most D&D discussions have some sort of practical application (even if it's just "hey look I can do a million damage and therefore get bragging rights") I would further argue that such discussions are not actually the bread and butter of D&D.
You completely missed my point. The bread and butter of D&D is basic addition, subtraction, and keeping track of numbers. Keeping track of four more in a fashion similar to HP isn't a stretch and certainly isn't anathema to "most tables". It's an utterly ridiculous assertion on its face. Hell, a paladin has a pool of lay on hands, that's one. Let's get really out there though: Anyone with a psycrystal or an animal companion or a familiar or a cohort has to keep track of far more numbers than the four suggested in the OP... and there's a complaint about having to do three digit addition and subtraction? (that will never exceed ~160 anyway) Are you kidding me?

[Bovine Colonel]

You completely missed my point. The bread and butter of D&D is basic addition, subtraction, and keeping track of numbers. Keeping track of four more in a fashion similar to HP isn't a stretch.

I also accidentally invalidated my own point and replaced it with a new one after the first blank line in my post :smalltongue:

[TuggyNE]

Anyone with a psycrystal or an animal companion or a familiar or a cohort has to keep track of far more numbers than the four suggested in the OP... and there's a complaint about having to do three digit addition and subtraction? (that will never exceed ~160 anyway)

It's actually practical for it to exceed 200, especially with fractional saves: just multiclass/prestige class into a total of four good-X-save classes.

But that brings up another rather unfortunate point that I alluded to briefly earlier: these new gradual defenses are never lower than 100 max, and never higher than the low 200s (barring insane multiclassing that might get up to a theoretical maximum of 350). Most of the time, characters of roughly the same level will have very nearly the same pool size, much more so than HP; 14 Con base wizard/barbarian average HP are, respectively, 6/14 at level 1, 53/93 at level 9 with Con +2 item, and 126/224 at level 19 with Con +6 item — but FP only scales from 100/110 through 115/130 to 130/155. A level 20 Barbarian has only 1.6 times the FP of a level 1 Wizard.

I'm not sure that's exactly wrong, but it's certainly disturbing, and suggests that a close look at the math is necessary to ensure this really works out right.

[Just to Browse]

You completely missed my point. The bread and butter of D&D is basic addition, subtraction, and keeping track of numbers. Keeping track of four more in a fashion similar to HP isn't a stretch and certainly isn't anathema to "most tables". It's an utterly ridiculous assertion on its face.I can tell by this statement that you're obviously not an average D&D player. You're probably an engineer of some kind (CSE? You have a distinct way of writing lists).

As such, let me enlighten you to this basic fact of mathematical psychology, using some data from D.C. Geary, et al. (1993) in the journal of Psychology and Aging. People with a mean age of ~20 could solve "simple" (1 digit minus 1 digit) subtraction ~1200 ms, but "complex" (2 digit minus 1 digit) subtraction in ~3200 ms. That's increasing the math by just an order of magnitude, and the result is that it moves three times slower. I tried some randomly-generated 3-digit (100-200) minus 2-digit (10-99) and 2-digit (10-99) minus 1-digit (1-9) subtraction and got ~5700 ms v. ~2700 ms, which is two times slower.

So I think it's pretty safe to say that just increasing subtraction problems by an order of magnitude noticeably messes up the time it takes to run an operation. Now couple that with changing the math involved from value comparison to subtraction (an order of magnitude more time) double the number of operations, and you have 40-60 times more play time for every required save operation. And when one of those operations can easily happen 1-2 times per creature's round, you get a lot of problems.

You're replacing a 2d/2d comparison with a 3d/2d addition, 3d/3d comparison, and 3d/2d subtraction... Yeah, that's a terrible idea and you shouldn't be surprised when people tell you it's terrible.

[Belial_the_Leveler]

Leaving aside criticism on the original premise, here are a few points you need to consider;

Against average opposition, HP become a static binary defense - once.
A fighter with a greatsword does 2d6+6 damage on the first level, average 11. When the attack hits, the victim either has enough HP to survive or dies. They don't even get to roll - survival depends on character creation. That is because potential damage is almost the same as HP for any character, or exceeds it. A 6th level fighter will attack twice for 2d6+12, average 38, criticals not included. Half the characters at that level (such as all casters) may not have enough HP to survive if it hits so HP becomes a static binary defense again. A 20th level fighter will attack 5 times for 2d6+32, average 200, criticals not included. Not every character will have 200 HP so if the attack hits, HP are again a not-gradual resource. And then you got the various optimized damage builds that will outdamage the fighter.
Should your system include gradual defenses against mundane damage?


Low-level characters become very tough against supernatural hazards.
Low level spell and ability DCs are around 15-17. This means that a low-level warrior will have 85% chance to resist the first Sleep spell, 70% chance to resist the second, 55% chance to resist the third. Taking his chance to save into account, one would need 5+ sleep spells to stop him. The wizard is not going to have that many offensive spells memorized for the entire day before level 5. Moreover, the wizard is not going to survive 5 rounds since the fighter can kill him in round 1 with his very binary greatsword.
Should low-level characters be much tougher against supernatural dangers than against physical damage?

Your proposed gradual defenses grow weaker as levels get higher.
Levels 1-2 the usual save DC is 15, with the save points being 100. Level 20 the usual save DC is 32-40, with save points being 160. By level 20 casters can pull off more than one spell per round. So against equal opposition the lvl 1 guy survives 5 rounds while the lvl 20 guy will get his points burned off in two. So effectively the "save points" system becomes weaker as the probability for rocket tag becomes higher.
Should another, more gradual system, be used that fixes this?


Is the system open to exploits?
People start multiclassing to get the best saving throw bonuses possible.
People start using effects that allow them saving throws against melee attacks.
People start using low-level multiattacks. "Oh look! A BBEG! Since he'll totally save-point against everything direct I got, I'll summon/call 20 imps with one spell (CR 2, 3hd monsters_ and have them suggestion for DC 15, forcing him to lose 150 points no matter what. And then I'll hit him with a quickened lose-no-save now that his points are gone.
People start using effects that force saves on abilities that don't normally have them. For example, a saving throw to negate the damage of or outright prevent melee attacks.

[Maginomicon]

Against average opposition, HP become a static binary defense - once.
A fighter with a greatsword does 2d6+6 damage on the first level, average 11. When the attack hits, the victim either has enough HP to survive or dies. They don't even get to roll - survival depends on character creation. That is because potential damage is almost the same as HP for any character, or exceeds it. A 6th level fighter will attack twice for 2d6+12, average 38, criticals not included. Half the characters at that level (such as all casters) may not have enough HP to survive if it hits so HP becomes a static binary defense again. A 20th level fighter will attack 5 times for 2d6+32, average 200, criticals not included. Not every character will have 200 HP so if the attack hits, HP are again a not-gradual resource. And then you got the various optimized damage builds that will outdamage the fighter.
Should your system include gradual defenses against mundane damage?

No.

First, something I should make clear: Note that there's a difference in the save points terminology between "failed save" and "failed saving throw". The saving throw happens first and involves your d20+modifiers. This determines whether your save points drop by a lot or by a little. Only after you determine how much your save points drop do you determine whether it's a failed save by rolling a d%. Also note that it takes days to recover save points to full strength, so for any given encounter you're highly unlikely to have your full complement of save points. You're likely to have "(3d6*5)-2+(5*base_save_bonus)" save points (or "50+(5*base_save_bonus)" save points if you use the simple average for NPCs and monsters).

Further, I should reiterate that strife specifically applies only to effects that have a saving throw line that says "saving throw: none" and directly affect a creature. Melee damage does not have a saving throw "line". Strife is thus irrelevant to melee damage.

Now, it's important to note that there are different kinds of gradual defenses:

True Gradual Defenses: These gradual defenses are not just a number that acts as a buffer, but actually decrease your overall effectiveness in some analog way or ideally decreases the effectiveness of the defense itself in some fashion as the gradual defense decreases. Save Points as-written are "true gradual defenses" because as your number of save points goes down, the chances of you succeeding on the save (regardless of the save DC) also goes down in analog.
Binary-Effective Gradual Defenses: These gradual defenses are essentially a binary defense that has a simple buffer that determines whether some condition is on or off. Sometimes a binary-effective gradual defense has multiple states along the way as it drains. HP as-written and nonlethal damage as-written are "binary-effective gradual defenses". HP protects you against the dying and dead conditions, and nonlethal damage protects you against the knocked-out (unconscious) and staggered conditions. Strife as-written is a binary-effective gradual defense in the same way as nonlethal damage. However, Strife does not have multiple states, it either wholly protects you against no-save effects... or it doesn't.


Just because you can have a gradual defense doesn't mean you should have a gradual defense. There has to be a justifiable game-balance reason for it to exist. The crux of why gradual defenses are important is so that you have warning that you're in danger. I don't know where you got your numbers for melee damage, but there are a number of factors in-play that make the necessity of a gradual defense specifically targeting mundane damage of all things... well, nonexistent: DR, AC, range, speed, mobility, etc., to name a few. The dragon in the air isn't sweating the fighter with the greatsword.

Saving throws however? You have no warning against those by RAW. Unlike the synergy between AC and HP, saving throws and "no-save" effects aren't partnered with a gradual defense that warns you that you're in danger.


Low-level characters become very tough against supernatural hazards.
Low level spell and ability DCs are around 15-17. This means that a low-level warrior will have 85% chance to resist the first Sleep spell, 70% chance to resist the second, 55% chance to resist the third. Taking his chance to save into account, one would need 5+ sleep spells to stop him. The wizard is not going to have that many offensive spells memorized for the entire day before level 5. Moreover, the wizard is not going to survive 5 rounds since the fighter can kill him in round 1 with his very binary greatsword.
Should low-level characters be much tougher against supernatural dangers than against physical damage?

Again, keep in mind that you're far more likely for any given encounter to have "(3d6*5)-2+(5*base_save_bonus)" save points (or "50+(5*base_save_bonus)" save points if you use the simple average), not maximum save points. At the start of an adventure, sure, you'll be at maximum, but as the day goes on, you'll start to drop. Even once the day is over and you rest, you don't go back to full, you just regain 20+base_save_bonus save points of each type as appropriate. No effect can speed this up really, and even a Hewards's Fortifying Bedroll only speeds this up in the short term. This is intentional. Much like ability burn, nothing short of natural healing through rest recovers save points or ever should recover save points. Save Points are deliberately designed so that the only way to recover them is time.

Unless I'm interpreting you wrong, you got your percentage chances for an NPC 1st-level warrior wrong. The correct numbers for a first level warrior are not 85%, 70%, 55%, etc. Assuming that the 1st-level warrior fails the saving throw for each sleep spell, the correct numbers for his chance of success (assuming DC 15) are 35%, 20%, 5%, etc. because his NPC average Will Points (much like how average HP is calculated) are equal to "50+(5*0)" and you deduct from save points before the d% roll.

As for the wizard surviving, you're assuming it's one-on-one wizard-vs-warrior. That's not actually a realistic scenario. Odds are the wizard is going to be hiding behind someone tougher in the party, and they should, because 1st-level wizards are squishy and should be squishy. You seem to be trying to argue that rocket tag should exist (at the very least at low-level).

Further, a significant amount of the benefit of the save points system is to favor the PCs and give the GM a reasonable way to fine-tune encounter balance versus caster-archetype PCs as they see fit. The PCs can over time (through rest) recover their save points so that they're more effective against monsters that have SLAs such as a cockatrice (CR 3 by the way, which with a DC 12 Fort save or be turned to stone is a fairly tall order for an early-game effective-save-or-die). NPCs on-average don't get that option because they use the average of "50+(5*base_save_bonus)", although if the GM feels that caster-archetypes in the party are being ineffective or super-effective he can lower or raise the save points on NPCs and monsters as he sees fit (meaning he can fine-tune encounter balance vs caster-archetype PCs very easily just like he can give NPCs more or less HP (within the confines of their hit dice) if he feels that mundanes are being more or less effective).

Again, the crux of why save points are necessary is because there's no warning that they're in danger. Generally-speaking, creatures in-fluff can get a sense that their save points are getting low by feeling ill (Fort points), that their muscles are aching (Ref points), or they have a headache (Will points). NPCs, thus, start off the encounter on average feeling on the cusp between sickly and raring. (a save point value of 50 represents a "typical" day feeling "about-average", and as an NPC's hit dice go up they feel more and more virulent on a "typical" day as-represented by their average save point value going up). As for PCs, they can see their stats, so they know that they're in danger in the meta-game (although in-character this is represented in-game the same way as an NPC). PCs need that warning much more than NPCs because the GM is in control of the NPCs.

And, yes, they should be tougher against supernatural dangers than against mundane dangers... at least, tougher against them than they already are (which is to say, by RAW they're not tough against them at all).


Your proposed gradual defenses grow weaker as levels get higher.
Levels 1-2 the usual save DC is 15, with the save points being 100. Level 20 the usual save DC is 32-40, with save points being 160. By level 20 casters can pull off more than one spell per round. So against equal opposition the lvl 1 guy survives 5 rounds while the lvl 20 guy will get his points burned off in two. So effectively the "save points" system becomes weaker as the probability for rocket tag becomes higher.
Should another, more gradual system, be used that fixes this?

This is intentional. As your Encounter Level gets higher, it's okay that your Save Points become slightly less effective because by then you have other methods of defense against those effects to compensate such as magic items and class abilities. In truth, save points act more as a way to reduce the likelihood of rocket tag rather than to utterly eliminate it (although, when you're dealing with multiple enemies, essentially rocket tag is eliminated, especially in mid-game and late-game since they'll have defenses against supernatural effects too).


Is the system open to exploits?
People start multiclassing to get the best saving throw bonuses possible.
People start using effects that allow them saving throws against melee attacks.
People start using low-level multiattacks. "Oh look! A BBEG! Since he'll totally save-point against everything direct I got, I'll summon/call 20 imps with one spell (CR 2, 3hd monsters_ and have them suggestion for DC 15, forcing him to lose 150 points no matter what. And then I'll hit him with a quickened lose-no-save now that his points are gone.
People start using effects that force saves on abilities that don't normally have them. For example, a saving throw to negate the damage of or outright prevent melee attacks.

Honestly, multiclassing needs more love.

There... there aren't any saving throws vs melee effects, even with save points and strife implemented, so I'm not sure where you're getting that assertion.

As for the imp scenario... that really isn't that overpowered or broken. You're having to wait to use your no-save effect. Also I think you're mixing save points with strife. Suggestion offers a will save, and if the BBEG would make the saving throw against each suggestion (which I think is what you're implying), yeah he'd lose 7 WP each time, but he wouldn't accrue strife from that. You accrue strife when you fail a save-allowed effect.

As for the force save on abilities that didn't before... again... strife only applies to effects that have a "saving throw" line that says "none" (and directly affect a creature). Melee attacks don't have a saving throw line.

There's one exploit that I can think of, and that involves buying multiple Heward's Fortifying Bedrolls and using them throughout the day to regain save points faster than normal. But... by then you're buying multiple Heward's Fortifying Bedrolls. :smallamused: