PDA

View Full Version : Fixing saves and multiclassing chassises (PEACH).



Yitzi
2011-07-06, 01:41 PM
This really addresses three points. The first (brought up in the fighter fix planning thread) is that PC saves are just plain too low.
The second is that if you multiclass heavily then your saves can get absurdly high, because you get to apply that +2 to your good saves once for each class.
The third is that if you multiclass heavily then your BAB gets absurdly low; as an extreme example, a 1 wizard/1 rogue/1 sorcerer/1 cleric/1 druid/1 monk has a BAB of 0, while even a pure wizard would have +3 by his level.

Long version
Therefore, first a rework of saves:

Saves have 3 progressions, shown as follows:

{table=head]Level | Poor Progression | Normal Progression | Good Progression | Normal Progression Bonus | Good Progression Bonus |
1st | +0 | +0 | +0 | +0/+2 | +2 |
2nd | +0 | +1 | +1 | +0/+2 | +2 |
3rd | +1 | +1 | +2 | +0/+2 | +2 |
4th | +1 | +2 | +3 | +0/+2 | +2 |
5th | +1 | +2 | +4 | +1/+2 | +2 |
6th | +2 | +3 | +5 | +1/+2 | +2 |
7th | +2 | +3 | +5 | +1/+2 | +2 |
8th | +2 | +4 | +6 | +1/+2 | +2 |
9th | +3 | +4 | +7 | +1/+2 | +2 |
10th | +3 | +5 | +8 | +1/+2 | +2 |
11th | +3 | +5 | +9 | +2 | +2 |
12th | +4 | +6 | +10 | +2 | +2 |
13th | +4 | +6 | +10 | +2 | +2 |
14th | +4 | +7 | +11 | +2 | +2 |
15th | +5 | +7 | +12 | +2 | +2 |
16th | +5 | +8 | +13 | +2 | +2 |
17th | +5 | +8 | +14 | +2 | +2 |
18th | +6 | +9 | +15 | +2 | +2 |
19th | +6 | +9 | +15 | +2 | +2 |
20th | +6 | +10 | +16 | +2 | +2 |
[/table]

To determine what save each class and creature type gets:
Racial hit dice have normal saves replace good saves, and poor saves replace poor saves.
Classes have good saves replace good saves, and normal saves replace poor saves.
Bonuses: Normal and good saves have additional bonuses. Unlike the base save bonus, this can only be applied once to each save, no matter how many classes you take. Some normal save bonuses have two values; the first is for normal saves from classes, while the second is for normal saves from racial HD.

And then for the rest of the multiclassing issue:
When calculating BAB or any particular save, you may choose to treat any class level (including racial HD) as a level in any other class with an equivalent or inferior progression.

Short version (it's exactly equivalent)

Use the fractional BAB/save system. For racial HD, poor saves are worth 1/3 per level and good saves are worth 1/2 per level with a 2 bonus. For class HD, poor saves are worth 1/2 per level with a 0 bonus before level 5, a 1 bonus from level 5 to level 10, and a 2 bonus from level 11 onward, while good saves are worth 5/6 per level with a 2 bonus. The bonus can be applied only once, no matter how many classes you take.


Thoughts?

Ashtagon
2011-07-06, 04:06 PM
This is easily fixed by the following two common house rules:

* The initial +2 bonus for a good save only applies once, no matter how many times you enter a class with that bonus.
* Use the fractional save/bab bonus rule from Unearthed Arcana.

There is a separate issue of what classes should have what save bonus, but that's a different matter entirely.

I also think you have made the difference between good and poor saves too big. Even at 20th level, the difference is 6 points -- enough that a given threat can still result in a meaningful chance of success or failure no matter what class the target is. With a 10-point difference as in your table, quite often a character will either save or fail depending on his class, no roll needed.

Siosilvar
2011-07-06, 04:18 PM
I also think you have made the difference between good and poor saves too big. Even at 20th level, the difference is 6 points -- enough that a given threat can still result in a meaningful chance of success or failure no matter what class the target is. With a 10-point difference as in your table, quite often a character will either save or fail depending on his class, no roll needed.

That's my knee-jerk reaction too. Discounting a cloak of resistance (because it applies equally to all saves) and without using tomes to increase abilities (only using enhancement items), the difference in saves at level 20 is typically:

Poor: +7 = +6 (base) + 1 (ability)
Good: +14 = +12 (base) + 2 (ability)
Good AND specialized ability (Like Will on a Cleric or Reflex on a Rogue): +20 = +12 (base) + 8 (ability)
[Assuming an ability score of 26: 15 base + 5 level + 6 enhancement]

That's 14 points of difference at the most; the gap doesn't need to be made larger.

While you may say that a character with a poor save will shore up that weakness, mind you that there's 10 points of ability score increase that I didn't use. [18 base + 2 race + 5 level + 6 enhancement + 5 inherent, although prohibitively expensive and cripplingly overspecialized, gives a score of 36]

If you want to add a "medium" progression, the usual one suggested is "+1 plus .4 per level, round down as usual," which results in a nice progression between the current "poor" and "good" saves.

Yitzi
2011-07-06, 04:57 PM
This is easily fixed by the following two common house rules:

* The initial +2 bonus for a good save only applies once, no matter how many times you enter a class with that bonus.
* Use the fractional save/bab bonus rule from Unearthed Arcana.

When you work it out, then (if the fractional save/bab bonus rule works like I think it does) it comes out to exactly the same numbers; it's just a different way of expressing it.


I also think you have made the difference between good and poor saves too big. Even at 20th level, the difference is 6 points -- enough that a given threat can still result in a meaningful chance of success or failure no matter what class the target is. With a 10-point difference as in your table, quite often a character will either save or fail depending on his class, no roll needed.

Well, failing due just to his class shouldn't happen unless he's got a lot of racial hit dice without good ability scores (which shouldn't happen); a normal (no racial HD) character should be using at least the normal save. Succeeding due just to his class (barring a natural 1) can happen, but that just forces the casters (who I think we can all agree could use a dose of humility anyway) to target the weaker saves (or, in the case of a monk opponent, to play a purely support role.)


That's my knee-jerk reaction too. Discounting a cloak of resistance (because it applies equally to all saves) and without using tomes to increase abilities (only using enhancement items), the difference in saves at level 20 is typically:

Poor: +7 = +6 (base) + 1 (ability)
Good: +14 = +12 (base) + 2 (ability)
Good AND specialized ability (Like Will on a Cleric or Reflex on a Rogue): +20 = +12 (base) + 8 (ability)
[Assuming an ability score of 26: 15 base + 5 level + 6 enhancement]

That's 14 points of difference at the most; the gap doesn't need to be made larger.

Actually, this won't make the gap substantially larger. It'll fluctuate a bit over the course of 20 levels (1 less at levels 2,4,8,10,14,17,18 and 19, 1 more at levels 3, 9, and 15) and end up at exactly the same at level 20.


While you may say that a character with a poor save will shore up that weakness, mind you that there's 10 points of ability score increase that I didn't use. [18 base + 2 race + 5 level + 6 enhancement + 5 inherent, although prohibitively expensive and cripplingly overspecialized, gives a score of 36]

Those 10 points will be an issue only for classes that have that ability as a key ability score, not those with it as a good save. So when comparing a paladin's will save to a wizard's, for instance, it'll actually narrow the gap.

The real answer, though, isn't that a character with a "poor" save will shore up that weakness as that it'll naturally be shored up to some extent by the ability boosts arising from his taking a monster race (as that's the only way to get a "poor" save), and as for the rest...there's a reason that monster characters tend to be underpowered.


If you want to add a "medium" progression, the usual one suggested is "+1 plus .4 per level, round down as usual," which results in a nice progression between the current "poor" and "good" saves.

Except that I'm not trying to add a "medium" progression in between the current "poor" and "good" saves, I'm trying to raise saves for all character classes up by one level.

Ashtagon
2011-07-06, 11:38 PM
Actually, this won't make the gap substantially larger. It'll fluctuate a bit over the course of 20 levels (1 less at levels 2,4,8,10,14,17,18 and 19, 1 more at levels 3, 9, and 15) and end up at exactly the same at level 20.


Under raw, the gap at level 20 between poor and good is 6 points plus item and ability bonuses. Under your system, the gap is 10 points plus the same bonuses. How can it not be greater under your system?

Reading more fully, I see that your good /normal replaces raw good/poor for classes, and your normal/poor replaces raw good/poor for racial HD. That makes the gap 6 points plus bonuses for a level 20 human -- still bigger than RAW. And it makes racial HD an even bigger nerf than they are already. Seriously -- people refuse to play creatures with racial HD because those are so bad already. Your fix would make that issue even worse.

Seerow
2011-07-07, 12:12 AM
First, your solution to the issue of multiclassing causing funky things with saves is way over complicated. As others have pointed out, the fractional BAB/save systems work a lot more efficiently for that.

Taking your example 1 wizard/1 rogue/1 sorcerer/1 cleric/1 druid/1 monk you would have (.5*2) + (.75 * 4) = +4 BAB.

For saves he would have 5 levels of good will progression, 2 levels of good reflex progression, and 3 levels of good fort progression. So he would have 2+(.3*1)+(.5*5) = +4.8 Will; 2+(.3*4)+(.5*2) = +4.2 reflex; and 2+(.3*3)+(.5*3) = +4.4 fort.

So say he takes another level in say Ranger. His BAB is now +5, his Fort is now +4.9, his Reflex is now +4.7, and his Will is now +5.1. So even though Will was his bad save from this class, it's the only one that increased this level, due to the amounts he had before.



On another note, I made a post over in the Fighter thread recently that kind of touches on this. The gist of what I suggest is saves using this table:

{table=head]Level | Bad Save | Good Save
1 | +1 | +2
2 | +2 | +3
3 | +2 | +4
4 | +3 | +5
5 | +4 | +5
6 | +4 | +6
7 | +5 | +7
8 | +5 | +8
9 | +6 | +8
10| +7 | +9
11| +7 | +10
12| +8 | +11
13| +8 | +11
14| +9 | +12
15| +10 | +13
16| +10 | +14
17| +11 | +14
18| +11 | +15
19| +12 | +16
20| +13 | +17[/table]


And in exchange for having higher base saves across the board, only add half your relevant stat mod to the saving throw, rather than the full thing. This helps to normalize saves to a degree, and lessen the disparity between say a cleric's will save with his primary stat fueling it, and a wizard's will save with a tertiary stat behind it. Because it seems to me that the huge stat disparity is the biggest thing that drives the system out of whack. If you take two characters with a good will save currently, one with a high stat, and one with a meh stat, the one with the high stat will easily pass saves while the one with a meh stat will almost never pass them. Reducing the influence of the stats allows you to tune the saves more easily.


Actually looking at my table and comparing it to yours, it actually ends up similar to your normal/good progressions, but I'd avoid having a bad progression, personally.

For the purposes of the fractional save system, my good progression is .75(3/4), and the bad is .6(3/5), with the bad save granting a +1 bonus, and the good save granting a +2.

Yitzi
2011-07-07, 07:24 AM
Reading more fully, I see that your good /normal replaces raw good/poor for classes, and your normal/poor replaces raw good/poor for racial HD. That makes the gap 6 points plus bonuses for a level 20 human -- still bigger than RAW.

Wait, how is 6 plus bonuses more than 6 plus bonuses?


And it makes racial HD an even bigger nerf than they are already.

Well, that can't be avoided, for the simple reason that racial HD means your CR is less (usually far less) than your ECL, so for things that are important to both (such as ECL) one of them has to give. Seems to me that because enemies usually do have racial HD, and PCs usually don't, making PCs with racial HD too weak is preferable to making monsters with racial HD too strong.


First, your solution to the issue of multiclassing causing funky things with saves is way over complicated. As others have pointed out, the fractional BAB/save systems work a lot more efficiently for that.

They're apparently more efficient in terms of understandability, so I'll change that; mechanically, however, it's exactly identical.


Taking your example 1 wizard/1 rogue/1 sorcerer/1 cleric/1 druid/1 monk you would have (.5*2) + (.75 * 4) = +4 BAB.

That's +5, you miscalculated.


For saves he would have 5 levels of good will progression, 2 levels of good reflex progression, and 3 levels of good fort progression. So he would have 2+(.3*1)+(.5*5) = +4.8 Will; 2+(.3*4)+(.5*2) = +4.2 reflex; and 2+(.3*3)+(.5*3) = +4.4 fort.

You really should count weak saves as 1/3, rather than just .3; this'll mean Will is 5 (5+1/3 with ranger), reflex will be 2+1/3 (2+5/6 with ranger), and fort will be 5.5 (6 with ranger.)

And under my system, work it out and it comes out exactly the same (to assuming 1/3 for bad saves.) For instance, for BAB he'll count his cleric, druid, and rogue levels as monk and his wizard level as sorcerer for +3+2=+5 (+6 with ranger)


On another note, I made a post over in the Fighter thread recently that kind of touches on this. The gist of what I suggest is saves using this table:

{table=head]Level | Bad Save | Good Save
1 | +1 | +2
2 | +2 | +3
3 | +2 | +4
4 | +3 | +5
5 | +4 | +5
6 | +4 | +6
7 | +5 | +7
8 | +5 | +8
9 | +6 | +8
10| +7 | +9
11| +7 | +10
12| +8 | +11
13| +8 | +11
14| +9 | +12
15| +10 | +13
16| +10 | +14
17| +11 | +14
18| +11 | +15
19| +12 | +16
20| +13 | +17[/table]


And in exchange for having higher base saves across the board, only add half your relevant stat mod to the saving throw, rather than the full thing. This helps to normalize saves to a degree, and lessen the disparity between say a cleric's will save with his primary stat fueling it, and a wizard's will save with a tertiary stat behind it. Because it seems to me that the huge stat disparity is the biggest thing that drives the system out of whack. If you take two characters with a good will save currently, one with a high stat, and one with a meh stat, the one with the high stat will easily pass saves while the one with a meh stat will almost never pass them. Reducing the influence of the stats allows you to tune the saves more easily.

Except I don't think that all good saves should be identical and all poor saves should be identical. Some variation does make sense. (Also, your system will play havoc with monsters; mine is designed to leave that part alone.)


Actually looking at my table and comparing it to yours, it actually ends up similar to your normal/good progressions, but I'd avoid having a bad progression, personally.

If so, all the monsters will have saves that are totally unattainable.

Ashtagon
2011-07-07, 08:36 AM
Wait, how is 6 plus bonuses more than 6 plus bonuses?

ok, a bit of misreading on my part. So the overall effect is to raise saves for both good and bad saves by +4 at 20th level, but severely nerf racial HD. This doesn't strike me as a particularly useful thing to do, unless you really hate high-CR critters as PCs.

Yitzi
2011-07-07, 10:54 AM
ok, a bit of misreading on my part. So the overall effect is to raise saves for both good and bad saves by +4 at 20th level, but severely nerf racial HD. This doesn't strike me as a particularly useful thing to do, unless you really hate high-CR critters as PCs.

More like I really hate having normal PCs have far lower saves than monsters of their CR; if making high-CR critters playable as PCs is desired, it makes more sense to bring ECL closer to CR (subject to differences that actually do make the difference between PCs and encounter monsters) than to give monsters the saves of a PC of their HD (more, actually, as they have better abilities).

Seerow
2011-07-07, 11:29 AM
That's +5, you miscalculated.

2*.5 and 4*.75 gets you 1 + 3, which is 4, not 5.


You really should count weak saves as 1/3, rather than just .3; this'll mean Will is 5 (5+1/3 with ranger), reflex will be 2+1/3 (2+5/6 with ranger), and fort will be 5.5 (6 with ranger.)


You're right on this one. I wasn't really thinking, that should have been represented as fractions, rather than decimals, but with fractions, your math is wrong..

2+(1/3*1)+(1/2*5) = +4 5/6 Will (5 1/3 with ranger); 2+(1/3*4)+(1/2*2) = +4 1/3 reflex (4 5/6th with ranger); and 2+(1/3*3)+(1/2*3) = +4 1/2 fort. (5 with ranger). It is slightly higher than my original numbers, but nowhere near the difference you seemed to get.


Except I don't think that all good saves should be identical and all poor saves should be identical. Some variation does make sense. (Also, your system will play havoc with monsters; mine is designed to leave that part alone.)


It doesn't make all good saves identical, it just brings them closer together. Instead of having a good save varying anywhere from +16 (+5 cloak of resistance -1 stat) to +30 (+5 cloak of resistance +13 stat), you narrow the range down to between +22 (+5 cloak, -1 stat divided by 2 becomes 0) and +28 (+5 cloak +13 divided by 2 becomes +6). So the range narrows from a difference of 14 down to a difference of 6, which is far more manageable.

Given your average save DC at level 20 is ~DC32, this makes the low end of a good save capable of saving on roughly a 10, with the high end of a good save saving easily on a 4. The poor save is 4 points behind that, so has to make a roll of a 14 on the low end of the stat distribution, or a roll of a 8 at the high end of the distribution. It is a meaningful difference between a good save and a bad save, but not an unsurmountable one.


As to monsters saves being too high, I'd argue that's a problem with monster design, where having double the hit dice of PCs is considered normal and fine for a monster of equal CR. Remember, the flip side of what you propose, as others have pointed out, is that a player who wants to play a monstrous character gets gimped even further than normal.

edit: Then there's also the negative effects against monsters who DON'T have huge hit dice, such as outsiders. I mean, look at the Balor, his saves are currently on par with average PC saves. You'd be dropping its relative effectiveness a fair bit.

Yitzi
2011-07-07, 02:30 PM
2*.5 and 4*.75 gets you 1 + 3, which is 4, not 5.

Whoops...I miscalculated. :smallredface:


You're right on this one. I wasn't really thinking, that should have been represented as fractions, rather than decimals, but with fractions, your math is wrong..

2+(1/3*1)+(1/2*5) = +4 5/6 Will (5 1/3 with ranger); 2+(1/3*4)+(1/2*2) = +4 1/3 reflex (4 5/6th with ranger); and 2+(1/3*3)+(1/2*3) = +4 1/2 fort. (5 with ranger). It is slightly higher than my original numbers, but nowhere near the difference you seemed to get.

I was speaking purely of the two approaches to multiclassing; the big difference came from the change in save progressions, which is a different matter entirely.

So, to go through it:
Will: The sorcerer level is counted as a wizard level, and the druid level as a cleric level for 1+1 plus 2 bonus. Once he takes ranger, he can count the ranger and monk levels as rogue for another 1. So 4, 5 with ranger.
Reflex: The sorcerer and wizard levels are counted as druid levels, and the monk level as a rogue level, for 1+1 plus 2 bonus. Again, 4, just like you got.
Fort: The sorcerer and wizard levels are counted as rogue levels, and the druid level as a cleric level, for 2 plus 2 bonus; once he takes ranger, he can count that as a monk level for another 1. So again, the same results you got.


It doesn't make all good saves identical, it just brings them closer together. Instead of having a good save varying anywhere from +16 (+5 cloak of resistance -1 stat) to +30 (+5 cloak of resistance +13 stat), you narrow the range down to between +22 (+5 cloak, -1 stat divided by 2 becomes 0) and +28 (+5 cloak +13 divided by 2 becomes +6). So the range narrows from a difference of 14 down to a difference of 6, which is far more manageable.

True; the problem is that firstly it plays havoc with the monster stats, and secondly makes the key ability less important than I'm comfortable with. It's also worth noting that even a difference of 14 isn't all that much when the lower value succeeds on a roll of 5.


Given your average save DC at level 20 is ~DC32, this makes the low end of a good save capable of saving on roughly a 10, with the high end of a good save saving easily on a 4. The poor save is 4 points behind that, so has to make a roll of a 14 on the low end of the stat distribution, or a roll of a 8 at the high end of the distribution. It is a meaningful difference between a good save and a bad save, but not an unsurmountable one.

And the way I see it, a poor save should average around needing a roll of 10 to succeed, and a typical good save (i.e. based off a secondary ability) should succeed on a roll somewhere in the 2-5 area.


As to monsters saves being too high, I'd argue that's a problem with monster design, where having double the hit dice of PCs is considered normal and fine for a monster of equal CR.

And yet the resulting numbers are quite playable when compared to the stats of PCs; the problems are only in the PCs' saves (which this is meant to address) and in the rare case (which is actually sort of a variant rule) of monsters as PCs. Since monsters are normally encounters, and rarely PCs, and the MM gives guidelines that fit with the high hit dice for monsters, I'd say it's monsters-as-PCs that needs fixing, probably on a case-by-case basis as requested (also, underpowered features are easier to handle than overpowered features, because it means a powergaming player will push for them to be fixed rather than trying to sneak them past.)


edit: Then there's also the negative effects against monsters who DON'T have huge hit dice, such as outsiders. I mean, look at the Balor, his saves are currently on par with average PC saves. You'd be dropping its relative effectiveness a fair bit.

Yes; instead of being far more dangerous than an equivalent-CR PC, he'd be more even. Remember, it's still something like a 50% chance to get a spell to stick on him if you bypass his SR, so (assuming minor if any extra penetration capability) that's something like a 30% success chance, and with UMD he can protect against some of the nastiest options. Oh, but he's also got unholy aura, so it's actually a 30% chance once you bypass SR, and 18% success chance. Compare that to a PC, who's got maybe a 5% success chance if you target a good save but a 50% chance if you target a poor save (note that all the balor's saves are roughly the same.) The balor's also got an impressive offense. In short, don't worry about the outsiders, they can handle themselves.

Seerow
2011-07-07, 02:50 PM
I was speaking purely of the two approaches to multiclassing; the big difference came from the change in save progressions, which is a different matter entirely.


Okay, so you were basing it off your rules rather than trying to correct my math with fractions. That makes more sense.

That said, your explanation did nothing but confuse me further, and makes me further believe that fractional saves are the better answer.


True; the problem is that firstly it plays havoc with the monster stats, and secondly makes the key ability less important than I'm comfortable with. It's also worth noting that even a difference of 14 isn't all that much when the lower value succeeds on a roll of 5.


And the way I see it, a poor save should average around needing a roll of 10 to succeed, and a typical good save (i.e. based off a secondary ability) should succeed on a roll somewhere in the 2-5 area.


These two statements are contradictory. First you say a difference is 14 is fine, then you give your expected values as only 5-8 apart. In fact, the expected results you want actually fall far better in line with my suggestion than yours. Unless you expect the good save to just have a lot of overkill, so you're only failing on a 1 because a natural 1 always fails.

The associated stat still matters with my method, it just isn't as huge a deal, which is important because you have 3 different stats associated with saves, and you are as likely to see people with a +0 as you are to see a +13, and having a 65% advantage based on your primary stat is way too huge. You're talking about a high save good stat succeeding on a 2, with a low save low stat succeeding only on a 20.

Hell, in the fighter thread someone was talking about expected saves being 1.5x level for a good save. A person with a good save + good primary stat -already- meets or exceeds that at all levels. The problem is the low end of saves is far too low. Where the good save good primary stat ends at +30, the bad save bad stat ends at +10-+12, that is way too wide of a variation. Reducing the difference from stats is the best way to bring that in line.

Yitzi
2011-07-07, 04:45 PM
That said, your explanation did nothing but confuse me further, and makes me further believe that fractional saves are the better answer.

Seeing how much my replacement system seems to confuse people, expressing it as fractional saves (which was the basic idea anyway; I just wanted to base it off the tables for some reason) does make more sense.


These two statements are contradictory. First you say a difference is 14 is fine, then you give your expected values as only 5-8 apart.

Not at all...If one save bonus is 10 below the DC, while the other is 4 above, then the bonuses will be 14 apart but the to-save numbers will be only 8 apart.


Unless you expect the good save to just have a lot of overkill, so you're only failing on a 1 because a natural 1 always fails.

Not all of the good saves will...but the ones that are also based off of primary classes will.


You're talking about a high save good stat succeeding on a 2, with a low save low stat succeeding only on a 20.

The only way a low save low stat (and he'd probably be advised to see if he can boost that stat a bit; at 8k, an ioun stone costs less than the last point of his cloak of resistance) could succeed only on a 20 is if he's either got a substantial penalty or is facing an enemy with a CR something like 5 higher than his ECL...in which case he probably should be running away anyway.

An 18 spread looks big, but it isn't really if only half of it overlaps with the portion that makes a difference.


Hell, in the fighter thread someone was talking about expected saves being 1.5x level for a good save. A person with a good save + good primary stat -already- meets or exceeds that at all levels.

Actually, it doesn't at high levels (if you look at the post you're talking about, I work out the calculations); in addition, that "1.5Xlevel for a good save" isn't assuming good primary stat.


The problem is the low end of saves is far too low.

The medium end (good save, secondary/tertiary ability) is also too low.


Where the good save good primary stat ends at +30, the bad save bad stat ends at +10-+12, that is way too wide of a variation. Reducing the difference from stats is the best way to bring that in line.

I still don't see why the difference between "even chance to make the save" and "fails only on a 20 even with a substantial penalty" is too big.

Seerow
2011-07-07, 05:30 PM
Not at all...If one save bonus is 10 below the DC, while the other is 4 above, then the bonuses will be 14 apart but the to-save numbers will be only 8 apart.


So you're assuming the good save DC is actually going to be 4 above the DC, so that you'd save on a 2 even with a -5 penalty to your save? Because that's the only way what you say here makes sense.

If you go with bonuses that are 14 apart and saving on a 2 exactly with a good save, you have this:

DC: 32
Good save: +30
Bad Save: +16

Good save saves on a 2, bad save saves on a 16. Yes, that is a HUGE discrepancy.

What you are saying is boost them both up another 4, for the good save still on a 2, but the bad save saves on a 12, which is more reasonable... but now you're just making the good save arbitrarily high rather than actually doing any good at balancing it.


Actually, it doesn't at high levels (if you look at the post you're talking about, I work out the calculations); in addition, that "1.5Xlevel for a good save" isn't assuming good primary stat.

Frankly, I don't buy into 1.5xlevel is the right number to shoot for. A +30 at level 20 is too high. Your save DCs are 32-34, so +27-28 is more in line with what I'd expect, for a save on around a 5 or 6 (and honestly, not all save DCs are even that high. Refering to the Balor again, all of his save DCs are DC23-27. The 32-34 are more along the lines of an optimized wizard, which should be harder to save against).

If you actually go for what you want, you will have players with +40 saves if they have a good stat and high save. Please show me anything that's near CR20 that needs saves anywhere near that high.

Saves are frankly pretty much good enough as they are when you have a decent stat feeding into them. The only thing that needs fixed is the gap between good and bad being reduced, which is what my solution does.

Yitzi
2011-07-07, 08:14 PM
So you're assuming the good save DC is actually going to be 4 above the DC, so that you'd save on a 2 even with a -5 penalty to your save?

The normal good save won't. The good save with a primary ability key will. (Of course, you're also more likely to have that penalty if the key ability is your primary.) But yes, I am assuming that the very top will have that level...of course, that just means the enemy will have to target a different save.


What you are saying is boost them both up another 4, for the good save still on a 2, but the bad save saves on a 12, which is more reasonable... but now you're just making the good save arbitrarily high rather than actually doing any good at balancing it.

Not really; remember, this isn't just the good save, it's the best of the good.


Frankly, I don't buy into 1.5xlevel is the right number to shoot for.

Well, that's too bad, because when it comes to CR it's right there in the MM.


If you actually go for what you want, you will have players with +40 saves if they have a good stat and high save. Please show me anything that's near CR20 that needs saves anywhere near that high.

It doesn't. But it's a natural side effect of "secondary stat and high save" producing +30 saves and stats being sufficiently important for the low-save people. And since it's a natural side effect, it needs a reason to ban it, not a reason to keep it.


Saves are frankly pretty much good enough as they are when you have a decent stat feeding into them. The only thing that needs fixed is the gap between good and bad being reduced, which is what my solution does.

And I think that it should be possible to boost your save by boosting the key ability modifier on a 1-for-1 basis. Why should it be any different from any other property, which gets a 1-for-1 basis?

Seerow
2011-07-07, 08:27 PM
Well, that's too bad, because when it comes to CR it's right there in the MM.


That's for NPC saves. While you can argue NPC humanoids with class levels are a little behind there, they generally either have other defenses, like the excuse you made for the balor, plus stronger offensive power (casters), or suck regardless and need other fixes (martial characters).

NPC saves need to be higher because player characters tend to have higher save DCs than enemies. Again, the optimized PC save DCs are in the 32-34 range by level 20, most monsters are in the upper 20s. This makes a drastic difference in what is needed to defend successfully.


And I think that it should be possible to boost your save by boosting the key ability modifier on a 1-for-1 basis. Why should it be any different from any other property, which gets a 1-for-1 basis?


Because everything else that increases on a 1 for 1 basis is typically based on your class or abilities, and you chose your stats to make sure those things are going up optimally. Because of this, those abilities will typically increase at a predictable rate, and thus you can balance around that.

The problem with saving throws is that they go based off of set stats that you aren't guaranteed to have at any set level, so they vary far more wildly than any other mechanic in the game. Imagine for example if any spell the Wizard used that granted a Fort save had a DC based off one ability, any spell granting a Ref save went off a different ability, and any spell granting a Will save went off yet another ability. Now imagine spells per day went off an ability separate from any of those 3.

That is the kind of MAD that saving throws as they currently exist represent. It makes it impossible to predict what the saves are going to be at any given level, because whether the stat guiding that save is the primary driving force. If you gave a Cleric a bad Will Save, he'd still save on a 1 against Will because his wisdom is going to be so high. Yet anyone who doesn't have wisdom as a primary stat suffers.

Reducing the variation from that slightly isn't total homogenization, but it brings the variation within a range that is more acceptable within game parameters, and makes it FAR easier to balance abilities.

Yitzi
2011-07-07, 08:54 PM
That's for NPC saves. While you can argue NPC humanoids with class levels are a little behind there, they generally either have other defenses, like the excuse you made for the balor, plus stronger offensive power (casters), or suck regardless and need other fixes (martial characters).

The martial characters might need other fixes, but they also need this one.


NPC saves need to be higher because player characters tend to have higher save DCs than enemies. Again, the optimized PC save DCs are in the 32-34 range by level 20, most monsters are in the upper 20s.

On the flip side, the monsters with those DCs in the upper 20s tend to have them at-will, while the PCs' top DC can really only be used once, maybe twice, per encounter (assuming 1/5 resources per encounter). Also, if you notice, even with the "good" progression I gave, the typical good save (i.e. not based on a primary or enhanced ability) will be a few points below that 30, which largely compensates for those few points between the PC's DC and the monsters'


Because everything else that increases on a 1 for 1 basis is typically based on your class or abilities, and you chose your stats to make sure those things are going up optimally.

So how is DC different than, say, AC (which depends in part on DEX)?


Imagine for example if any spell the Wizard used that granted a Fort save had a DC based off one ability, any spell granting a Ref save went off a different ability, and any spell granting a Will save went off yet another ability. Now imagine spells per day went off an ability separate from any of those 3.

The difference is that offense is supposed to be highly focused, while defense is supposed to be spread-out.


That is the kind of MAD that saving throws as they currently exist represent. It makes it impossible to predict what the saves are going to be at any given level, because whether the stat guiding that save is the primary driving force.

Except it isn't; barring extreme cases, it'll be less important than the progression.

You're contrasting cleric and wizard, but that's the extreme case. A better contrast would be barbarian and ranger for Fort, or cleric and monk (or monk and wizard) for Will.


and makes it FAR easier to balance abilities.

Maybe we have different definitions of "balance"; as far as I'm concerned, balance is primarily determined by one's strongest attack and weakest defense.

Seerow
2011-07-07, 09:07 PM
The martial characters might need other fixes, but they also need this one.


Why? If monsters can get by because they have defenses besides just saves, why do PCs need such high saves that they can basically never be hit by monster abilities?


On the flip side, the monsters with those DCs in the upper 20s tend to have them at-will, while the PCs' top DC can really only be used once, maybe twice, per encounter (assuming 1/5 resources per encounter). Also, if you notice, even with the "good" progression I gave, the typical good save (i.e. not based on a primary or enhanced ability) will be a few points below that 30, which largely compensates for those few points between the PC's DC and the monsters'


So look instead at the Solar, who has casting just like the spellcasters, and his DCs top out at 26 (17+spell level).

Nevermind the fact that at will doesn't mean as much when your life expectancy is one encounter.

Consider this, with what you suggest, even a moderate stat (say a 22 at level 20), you have a +12+6+5=+23, and will save against those abilities mentioned (in the mid 20s) on below a 5.


So how is DC different than, say, AC (which depends in part on DEX)?


Heavy Armor can limit sharply how much dex you need to max out. Your typical character will max it out with just a +stat item.

If you'd like to homebrew some saving throw armor that puts a cap on stat contribution to saves in exchange for a saving throw boost, I'd admit that as a viable alternative solution, which would help significantly narrow the range of saves without reducing the contribution from stats directly. My solution is faster and easier to implement.



Except it isn't; barring extreme cases, it'll be less important than the progression.

You're contrasting cleric and wizard, but that's the extreme case. A better contrast would be barbarian and ranger for Fort, or cleric and monk (or monk and wizard) for Will.

The point is you are ignoring the extreme case. I am arguing balance the extreme cases, and everything in between falls into place naturally.


Maybe we have different definitions of "balance"; as far as I'm concerned, balance is primarily determined by one's strongest attack and weakest defense.


This statement doesn't even make any sense.

Yitzi
2011-07-08, 07:29 AM
Why? If monsters can get by because they have defenses besides just saves, why do PCs need such high saves that they can basically never be hit by monster abilities?

They don't...they just need high enough weak saves to make the save half the time; their strong saves should naturally be quite a bit stronger. At lower levels, their strong saves can be hit.


So look instead at the Solar, who has casting just like the spellcasters, and his DCs top out at 26 (17+spell level).

He has a lot of SLAs in addition to those spells, though. Plus SR, regeneration, and no weak saves.


Nevermind the fact that at will doesn't mean as much when your life expectancy is one encounter.

That makes all your abilities closer to at-will, not closer to the restrictions a PC has.


Consider this, with what you suggest, even a moderate stat (say a 22 at level 20), you have a +12+6+5=+23, and will save against those abilities mentioned (in the mid 20s) on below a 5.

And when he targets your poor save?


Heavy Armor can limit sharply how much dex you need to max out. Your typical character will max it out with just a +stat item.

True. So there is that.
It still seems to me that the extra difference in your strong save (which most monsters won't be targetting if they can help it) isn't such a big deal, and the extra difference in your weak save is desirable.


The point is you are ignoring the extreme case. I am arguing balance the extreme cases, and everything in between falls into place naturally.

And I'm arguing that because there are 2 independent variables, you can only balance at most 3 of them, and the highest one is the least important (both because you fail on a 1 no matter what and because nobody's going to target a cleric's Will save if they have a choice.)


This statement doesn't even make any sense.

Nobody cares about the balance of a wizard's melee attack, because he's not going to be using it. Likewise, the balance of a cleric's Will save isn't so important, as anyone with a choice will pick a different save to target.

Seerow
2011-07-08, 10:23 AM
He has a lot of SLAs in addition to those spells, though. Plus SR, regeneration, and no weak saves.


And the SLA DCs top out at DC25.

Having all the SLAs and spells in the world means nothing if you can't even hit someone with a bad save with them. And remember this is a CR+3 encounter. And while it has no weak saves, all of its saves are between +18 and +20, which is the same or weaker than a PCs bad save in your system.

I'm not even sure why you're bringing up the SR and regeneration, since the point I was making is that monster save DCs aren't high enough to require the huge saves you are aiming for.


That makes all your abilities closer to at-will, not closer to the restrictions a PC has.


That statement was made in context of the solar's save DCs being as weak as the at will abilities.

The ONLY thing that has the sort of DCs you're aiming for is a optimized NPC full caster, or a monster with a full NPC wbl using that wealth on ability score boosting. (spending 172,000 on a tome + enhancement item brings the archon's DCs up to where the PCs are... too bad the Archon doesn't actually get WBL.


And when he targets your poor save?


That IS him targeting the poor save. With your system, the poorest base save for PCs is +12. The poor +6 progression only applies to monsters from my understanding. That example was a poor save with a moderate stat. A good save with a moderate stat would save on a 1.


True. So there is that.
It still seems to me that the extra difference in your strong save (which most monsters won't be targetting if they can help it) isn't such a big deal, and the extra difference in your weak save is desirable.


A strong save shouldn't be untargetable. It should be an easy save, but you shouldn't always save a strong save on a 2. Similarly, you shouldn't be saving with a bad save on a 5.


And I'm arguing that because there are 2 independent variables, you can only balance at most 3 of them, and the highest one is the least important (both because you fail on a 1 no matter what and because nobody's going to target a cleric's Will save if they have a choice.)


Whereas I have shown it is possible to balance all 3, you just don't like the solution because it's different, and might actually result in a failed saving throw once in a while.

Yitzi
2011-07-08, 11:50 AM
Having all the SLAs and spells in the world means nothing if you can't even hit someone with a bad save with them.

Except that the bad saves wouldn't be more than +17 plus ability score, which is quite achievable in most cases.


And remember this is a CR+3 encounter. And while it has no weak saves, all of its saves are between +18 and +20, which is the same or weaker than a PCs bad save in your system.

Of course, that doesn't count the +4 to saves vs. evil from the aura of protection. Still, looking it over it would appear that the 1XCR/1.5XCR rule is for "brute"-sort monsters (for instance, the Tarrasque fits it, except for its insanely high Fort save); PCs are generally assumed to not be of that sort, so the system does need changing. I'll give the matter some thought...


I'm not even sure why you're bringing up the SR and regeneration, since the point I was making is that monster save DCs aren't high enough to require the huge saves you are aiming for.

Oh, about monster save DCs. Sorry, I misunderstood you.


Whereas I have shown it is possible to balance all 3, you just don't like the solution because it's different, and might actually result in a failed saving throw once in a while.

Actually, I don't like the "apply only half key ability score" approach not only because it is fundamentally different from everything else in D&D, but because it'll play havoc with the monsters and also because there aren't 3 points to balance, but 4, and not allowing characters to boost their weak saves by boosting the key ability is, IMO, a bad idea.

Yitzi
2011-07-08, 11:54 AM
I've thought about it a bit more, and it looks like at level 20 the goal is approximately 17 poor save, 22 good save, and 27 DC. Now to think about how to achieve that...

Yitzi
2011-07-08, 12:36 PM
Ok, here's my idea (completely replacing the old idea on changing save slots):

-Amulet of health is moved to the Belt slot, so that someone can take gloves of DEX, periapt of WIS, and belt of health to boost their saves further. At 12k, the first point isn't really all that much more than the 9k for the 5th point of resistance (but of course those who are already maxing out the ability get no further benefit.)
-Belt of giant strength is moved to the Bracers slot, so that the fighter can get both the STR boost and the CON boost.
-Inherent bonuses are capped at +1. This'll keep DCs under control and keep primary ability scores from pulling too far ahead (and will help the MAD classes, which could definitely use it.)
-Character classes' poor progression (but not those of racial HD) is upgraded to 40%Xlevel.

The result: At level 20, poor saves have +8+3+5=16 plus the starting key ability modifier, good saves have 20 plus the starting key ability modifier, and DCs for 8th level spells (as a rough average of those likely to be used over the course of an encounter) are 10+8+3+3=24 plus the starting key ability modifier (generally 4 or 5). So casters still have a somewhat high "hit" chance, but that's compensated for by SR and squishiness (if they're not squishy, that's a separate issue.)

Thoughts?

Epsilon Rose
2011-07-08, 01:20 PM
What if someone doesn't want all of there slots filled with stat boosting items? There are a lot of interesting things in the various books and requiring people to take +x to stat items seems kind of sad.

Also how does capping inherent bonuses help mad characters?

Also also, what's the advantage of you're system over the fractional system? You've said a number of times that they're mechanically the same, but the fractional one's simpler and semi-official, so why use yours?

Yitzi
2011-07-08, 03:08 PM
What if someone doesn't want all of there slots filled with stat boosting items? There are a lot of interesting things in the various books and requiring people to take +x to stat items seems kind of sad.

They don't have to; they just won't have as high saves if they don't. Everything has its price.


Also how does capping inherent bonuses help mad characters?

It's more like it hurts SAD characters, who are the ones who will gain the most use from +5 inherent; MAD characters generally can't afford to boost more than one of their attributes that way.


Also also, what's the advantage of you're system over the fractional system? You've said a number of times that they're mechanically the same, but the fractional one's simpler and semi-official, so why use yours?

Because I didn't know when I posted it that the fractional one was semi-official; that's why the new version just uses the fractional one.

PairO'Dice Lost
2011-07-11, 12:13 PM
What if someone doesn't want all of there slots filled with stat boosting items? There are a lot of interesting things in the various books and requiring people to take +x to stat items seems kind of sad.

As of the MIC, you can add the functions of a stat booster to any other item in that slot for no markup (e.g. you can add +4 Dex to your gloves of arrow snaring for the price of gloves of arrow snaring plus the price of gloves of dexterity +4), so you don't have to choose between stat boosters and interesting items.

Yitzi
2011-07-11, 07:56 PM
It's occurred to me...rather than capping inherent bonuses, better to change them so they actually increase your base ability score, but your total base ability score is capped at 18+HD/4 (rounded down.)

That way, they can't be used to boost an already bloated primary ability, but do help with secondary ability scores (helping both saves and MAD classes overall).

Ashtagon
2011-07-12, 02:51 PM
This got me interested in 1e/2e saves...

First up, except for what appears to be a transposition error on the thief/rogue level 21+ row, the saving throw tables are identical between 1e and 2e.

Second, fighters start with the worst saving throws at level 1, but end with the best saving throws at level 20. Their saving throws improve almost twice as fast as any other class.

For my methodology, I compared each class at 1st and 20th levels with the average, best, and worst saves for all four archetypes at levels 1 and 20. I took "poison, paralysis and death" and petrify/polymorph" to both be Fort saves, "rod/staff/wand" and "spells" to be Will saves, and "breath" to be a Reflex save. My results are as follows.


{table=head] Class | Fort | Reflex | Will
Fighter | Average | Average | Average
Rogue | Average | Poor | Average
Cleric | Good | Average | Average
Wizard | Poor | Average | Good
[/table]


The big surprise here was that rogues in 1e/2e had poor Reflex saves. Who knew?