When Picking Spells How Many Saving Throws Is Too Many?

[unseenmage]

While I was working on this post about spell combos in items (http://www.giantitp.com/forums/showsinglepost.php?p=21857846&postcount=30) I noticed that adding spells together means allowing for multiple saving throws/skill checks to avoid spell effects either partially or altogether; especially when the second spell's effect hinges on the first.

This got me wondering, how many chances to escape a spell effect is too many? We describe a lot of spells as 'Save or Die' when really there are often several saves or checks between the victim and sudden annihilation. I mean, sure it's way fewer chances to live than vs a mundane swinging something heavy but still.


So, my question is, when choosing spells or when homebrewing your own how many saving throws and/or skill or ability checks does the playground consider to be too many for a spell of a given spell level to be 'viable'?

Makes me wonder if there is any correlation between Spell Level and the number of die rolls a subject gets to avoid the spell's effect(s).

[Flickerdart]

Anything more than none, and you're putting success of your actions into the enemy's hands. Rely on attack rolls, ideally ranged touch attack rolls, for your complex shenanigans.

With one saving throw involved, you still have decent odds that you can target the enemy's weak save. With two saving throws (unless they are the same) you risk bumping into a strong save and fizzling out. At the low levels, when you haven't boosted your DCs yet, every saving throw is pretty much even odds. Two saves - a 75% chance of your spell failing - is not really worth it.

[Telonius]

I'd say it depends on the spell. If the save is for a rider effect to what you're really trying to achieve, you might not care about how many saving throws the target gets. I'm thinking about things like the Orb of X line, where the main thing you're trying to do is direct damage. If the target makes his save, too bad; if he misses it, great; the status effect wasn't really what you cared about.

[Particle_Man]

Spell Resistance is also another line of defence for most spells.

I think the lowest level "save or die" is the level 4 Phantasmal Killer, which has 2 saves and SR to deal with.

[Lazymancer]

Viability of spell is not determined by the reliability (saving throws, spell resistance, or attack rolls) alone. Effect (which includes number of targets) and expenses (action it takes to cast and XP/material components/focus) are also relevant.

So the answer to your question would be: it depends on effect spell has. For example, if effect is something like "-2 to saving throws" it'd better have swift action to cast or area effect. Regardless of spell level it has, I'm not wasting an action in combat on something useless.

[unseenmage]

Viability of spell is not determined by the reliability (saving throws, spell resistance, or attack rolls) alone. Effect (which includes number of targets) and expenses (action it takes to cast and XP/material components/focus) are also relevant.

So the answer to your question would be: it depends on effect spell has. For example, if effect is something like "-2 to saving throws" it'd better have swift action to cast or area effect. Regardless of spell level it has, I'm not wasting an action in combat on something useless.

See, i have to disagree some here. I take it as a given that one would not intentionally choose spell effects one isn't interested in. Spell effects that aren't useful/desirable were never under debate. Spells whose effects are desirable that have a given number of die rolls separating success from failure, however, are the topic of debate. By that, and this is clear by the OP, the viability of a given spell IS determined by it's reliability, if only because the reliability of undesirable spells isn't important to the premise of the thread.

TLDR: Sure spell effects are important, but spell reliability is more important for desirable spell effects.

[Venger]

Spell Resistance is also another line of defence for most spells.

I think the lowest level "save or die" is the level 4 Phantasmal Killer, which has 2 saves and SR to deal with.
yeah, it is.

it's also a mind-affecting fear illusion, so it works a very small fraction of the time.

[Coidzor]

Two-Saves is incredibly risky. Three-Saves is worthless.

A save each round just isn't worth it, unless you only need it for the first round and it'd be nice to have it on the second round.

[remetagross]

yeah, it is.

it's also a mind-affecting fear illusion, so it works a very small fraction of the time.

There is also Touch of Juiblex, level 3. It allows one save SR, requires a close combat touch attack, and takes four rounds to take effect. Also the caster takes 1d6 points of Str damage. But it's a 3.0 leftover from the BoVD so it's not nearly as a staple as Phantasmal Killer!

[Lazymancer]

See, i have to disagree some here. I take it as a given that one would not intentionally choose spell effects one isn't interested in. Spell effects that aren't useful/desirable were never under debate. Spells whose effects are desirable that have a given number of die rolls separating success from failure, however, are the topic of debate. By that, and this is clear by the OP, the viability of a given spell IS determined by it's reliability, if only because the reliability of undesirable spells isn't important to the premise of the thread.

TLDR: Sure spell effects are important, but spell reliability is more important for desirable spell effects.
But there are no uniformly desirable spells.

For example, Orb spells are not good damage dealers, their effect is mediocre. It is their lack of SR/saving throw that makes them important. On the other hand, Sleep is great because it has great effect: area SoD. It is good enough to tolerate both SR, and saving throw, and "mind-affecting", and low HD cap.

[zergling.exe]

But there are no uniformly desirable spells.

For example, Orb spells are not good damage dealers, their effect is mediocre. It is their lack of SR/saving throw that makes them important. On the other hand, Sleep is great because it has great effect: area SoD. It is good enough to tolerate both SR, and saving throw, and "mind-affecting", and low HD cap.

That last bit is the most important, because for the level of enemies you use it on (4 and under) SR is nearly non-existent, saves are too low to counter a high ability mod, and mind-affecting immunity is basically skeletons, zombies, and vermin.

[Ashtagon]

Anything more than none, and you're putting success of your actions into the enemy's hands. Rely on attack rolls, ideally ranged touch attack rolls, for your complex shenanigans.

Why do you distinguish between spells that need to get past a save to succeed, and spells that need to get past an attack roll (whether normal or touch) to succeed? Both can be gamed to increase your attack bonus/save DC, and both can be countered by various means by the defender. I'm not sure how these are qualitatively different from each other.

[Telok]

Since 3.0 first came out I've seen Phantasmal Killer cast about five or six times. It worked once, on a warblade, who rolled two ones in a row and still almost made it.

[ZamielVanWeber]

I use phantasmal killer when I want someone to know that I am willing to kill them but am not actually trying to kill them. It has worked out amazingly on that front.

[Flickerdart]

Why do you distinguish between spells that need to get past a save to succeed, and spells that need to get past an attack roll (whether normal or touch) to succeed? Both can be gamed to increase your attack bonus/save DC, and both can be countered by various means by the defender. I'm not sure how these are qualitatively different from each other.

It's fairly trivial to load yourself up with Evasion, save rerolls, and general "can't touch this" abilities when it comes to saves, but it's much harder to do the same thing to AC. Meanwhile, boosts for attackers (starting from the humble true strike) are very easy to come by.

[Venger]

It's fairly trivial to load yourself up with Evasion, save rerolls, and general "can't touch this" abilities when it comes to saves, but it's much harder to do the same thing to AC. Meanwhile, boosts for attackers (starting from the humble true strike) are very easy to come by.

yeah, which is why traditional wisdom says you're probably better off with something that provides miss chance or similar.

[Zanos]

It's fairly trivial to load yourself up with Evasion, save rerolls, and general "can't touch this" abilities when it comes to saves, but it's much harder to do the same thing to AC. Meanwhile, boosts for attackers (starting from the humble true strike) are very easy to come by.
Yeah, DCs are hard to increase, while saving throw bonuses are common. The inverse is true with attack bonuses and touch AC.

In reference to how many saving throws I'm willing to stomach before a spell does something, I put my limit on 1, except for outliers. 1 save is fine for a single target. 2 is okay if your spell hits a whole bunch of stuff. The issue is that more and more rolls stack multiplicatively. If I have a 50% chance to get through SR, and my target has a 50% chance to fail his will save, then a 50% chance to fail his fortitude save, my phantasmal killer has a 1 in 8 chance of actually doing something, and if it doesn't I wasted my entire turn and a spell slot. Now if you're combining spells where the first step makes the enemy a lot more likely to fail the second step it isn't such a big deal, and a 1 in 8 chance of accomplishing something matters a lot less if there's a whole bunch of targets you can drop it on at once. Also, if you can stack your success rates really high this doesn't matter as much. If those chances above were 90/90/90, we've got a 72.9% chance of doing something, which isn't bad. The problem is that it's hard to stack your DCs high enough to hit those kind of ratios.

[Flickerdart]

yeah, which is why traditional wisdom says you're probably better off with something that provides miss chance or similar.

True strike takes care of most miss chances, too.

[unseenmage]

But there are no uniformly desirable spells.

For example, Orb spells are not good damage dealers, their effect is mediocre. It is their lack of SR/saving throw that makes them important. On the other hand, Sleep is great because it has great effect: area SoD. It is good enough to tolerate both SR, and saving throw, and "mind-affecting", and low HD cap.

Uniformly a given individual will desire a given effect; that effect being whatever effect it is they're after at the time.

My question is, how many die rolls does that effect have to hinge on before the spell becomes untenable as a means of achieving that effect?
To that question, the effect doesn't matter.

[Coidzor]

Uniformly a given individual will desire a given effect; that effect being whatever effect it is they're after at the time.

My question is, how many die rolls does that effect have to hinge on before the spell becomes untenable as a means of achieving that effect?
To that question, the effect doesn't matter.

I think effect matters, but more in terms of whether it's a major or minor effect relative to the level.

I tend to favor more minor effects (Save or Suck) with no or fewer rolls and more chance of success or guaranteed effects over more major effects (Save or Lose) with more rolls and less chance of success, and I favor both of those over encounter-ending/creature-removing effects (Save or Die) which have a very low chance of success.


Single Target:

An attack roll, rolling to overcome SR, and the target getting a saving throw, and if any of those fail, then nothing happens is too many.

The target getting a saving throw, then another saving throw if they fail that saving throw, and passing either means that nothing happens is too many.

The target getting a saving throw, then taking some minor but level appropriate negative effect and having another saving throw to see if they get an more major but still level appropriate effect on top of that is OK to me.

Multi-Target:

Two Saves is fine, especially if one of them is only a save for a reduced effect.

Two saves and rolling to overcome SR and missing any one of those leads to complete negation is too much.

Two Saves which are individually partial negation of different negative effects of a spell and SR once is more palatable.

[Quertus]

An I remembering correctly, that if one of your multiple targets makes its SR, the whole spell fails?

[ZamielVanWeber]

No. SR only protects the individual. If you fireball 5 drow and only one get its SR to work only that drow will not take any damage. The rest reflex for half.

[Telok]

I use phantasmal killer when I want someone to know that I am willing to kill them but am not actually trying to kill them. It has worked out amazingly on that front.

Hilariously that's exactly what I was using it for. I think the DC was around 14 or 15 too.

Another thing that affects this is if your DM uses a single save per monster group. It's a way to speed thigs up but it also really harshes on AoE spells that have multiple failure points as a single roll can negate five or seven targets worth of spell.

[Venger]

Hilariously that's exactly what I was using it for. I think the DC was around 14 or 15 too.

Another thing that affects this is if your DM uses a single save per monster group. It's a way to speed thigs up but it also really harshes on AoE spells that have multiple failure points as a single roll can negate five or seven targets worth of spell.

Wow that is... a terrible idea. Does he take the average or what?

[MesiDoomstalker]

Wow that is... a terrible idea. Does he take the average or what?
I believe Telok means that a single die is rolled and the result applied to all monsters of the same variety. Fireball that mob of 40, level 2 Rogue Goblins. Oops, they rolled a 20. No damage for you. It is indeed faster, since rolling 40 d20s is nuts (But so is an encounter with 40 opponents, regardless). Even in a saner number of opponents, it's a tempting time saving option for a DM.

[Venger]

I believe Telok means that a single die is rolled and the result applied to all monsters of the same variety. Fireball that mob of 40, level 2 Rogue Goblins. Oops, they rolled a 20. No damage for you. It is indeed faster, since rolling 40 d20s is nuts (But so is an encounter with 40 opponents, regardless). Even in a saner number of opponents, it's a tempting time saving option for a DM.

That's still a bad idea. Use an electronic roller if you're using such unwieldy lumps of enemies (also why are you using 40 monsters?) If as a gm you like to throw unworkably large numbers of enemies at your pcs at once, like if you have an ubercharger in the party or something, you may be better off prerolling a huge chain of numbers in advance

[unseenmage]

This thread got me wondering what a metamagic feat that removed all but the first die roll to negate or mitigate a spell would look like.


A friend suggested it'd be a +2 metamagic increase to have Disintegrate only require 1 save vs all of its effects.

[Coidzor]

This thread got me wondering what a metamagic feat that removed all but the first die roll to negate or mitigate a spell would look like.


A friend suggested it'd be a +2 metamagic increase to have Disintegrate only require 1 save vs all of its effects.

+1 per obviated saving throw/layer of defense?

[MesiDoomstalker]

That's still a bad idea. Use an electronic roller if you're using such unwieldy lumps of enemies (also why are you using 40 monsters?) If as a gm you like to throw unworkably large numbers of enemies at your pcs at once, like if you have an ubercharger in the party or something, you may be better off prerolling a huge chain of numbers in advance

Oh I agree. An encounter with more than twice the enemies as there are PC's is way too unwieldy and slow (PCx1.5-PCx2 is still difficult). But there are GMs who do it and then employ a "One Die to Roll them All" approach. I, at least, use 1 die roll for Initiative (for groups with the same bonus).

[Fouredged Sword]

I find it is easiest to have a electronic roller and subtract the DC from the save ahead of rolling. Every negative number is a failed save, every positive is a pass. It makes it easy to roll 20-30 saves in bulk with just 2-3 sets based on save totals. Excel is your friend.