Favorable and unfavorable saves

Swords & Wizardry collapsed all the saving throws into a single number, with some class-based modifiers. For example, clerics get +2 when saving against paralysis or poison, and magic-users get +2 when saving against spells. The traditional saving throw categories do provide atmosphere (death ray, dragon breath), but are somewhat cumbersome and nonintuitive.

One thing I’ve been doing recently is using “most favorable” or “least favorable” save numbers for cases where the choice of what save category to use is not immediately clear. If it seems like something the class in question would have some competence with, the character gets to use the most favorable. Where this has come up most is for the saving throw involved in carousing-like activities.

I have previously discussed simplifying saving throws by deriving them directly from character level. Here is another approach. Rather than have one column of saves per class, as S&W does, or use the more complicated multi-save system as did the original TSR editions, why not have two progressions by level: favorable and unfavorable. All classes would reference the same values, but would differ as to which number was used by situation. This method would require two numbers, but would avoid needing any class-based or situational modifiers.

Putting this idea into Third Edition terms, combat oriented classes would use the favorable numbers for direct physical situations, whereas magic-users would use the favorable numbers for resisting sorcery or mental effects.

Saving Throw Competency by Class
Class Favorable Unfavorable
fighter fortitude reflex, will
magic-user will fortitude, reflex
thief reflex fortitude, will

The cleric does not fit quite so neatly into the 3E classification. A cleric should probably have favorable saves when dealing with demonic or undead threats, for example, but not necessarily for situations that require general toughness. Classes like the cleric would be easily handled by this proposed dual number system, without needing to spell out the types of threats beforehand.

Bifurcated Saving Throws
Level Favorable Unfavorable
1 – 3
4 – 6
7 – 9
10 – 12

These numbers are derived from the OD&D fighter best and worst saves. I chose the fighter because A) the fighter is the most fundamental class and B) fighter saves improve every third level rather than the less frequent schedules of the other classes. The resulting pattern is also quite nice and easy to remember, as the two numbers always differ by 4 (20%) and always improve by 2 every tier (once a player has written down 12/16 they never even have to every consult the table again as long as that basic rule is remembered).

All classes would use the favorable number for the 0 HP death save.

I am aware that some people think that saving throws should be collapsed into ability checks, but I do not think that is the best approach for a level-based game, as improving saving throws should be a reward for longterm successful play, not a trick of the dice at first level. See here for more on my philosophy of saving throws.

10 thoughts on “Favorable and unfavorable saves

  1. Hedgehobbit

    I came up with a similar system. However, rather than make the distinction between favorable and unfavorable simple on Class, I make the decision based on the character’s concept. For example, a northborn barbarian fighter would have favorable saves versus cold and exposure whereas the city-born swashbuckler fighter would not. However, the swashbuckler would have favorable saves when it comes to balancing or deftly leaping away from traps. And the noble knight would have favorable saves versus things like Charms and Fear.

    An abbreviated version is here:

    1. Brendan


      Very nice work; thanks for the link. The five different difficulty levels are probably more than I would need, but I can definitely see your system working well.

      I also like the connection to character concept (class is really just a special case of character concept, in some ways at least). If my two number save system makes it into my Pahvelorn Player’s Guide, I will probably amend the language somewhat to reflect that individual character concept should be considered. Class is still a good fallback though.

    2. Hedgehobbit

      My main goal was to encourage players to develop their characters by giving them a significant mechanical advantage for doing so. I think I view classes differently than you. I don’t see classes as a list of choices for the players to pick from. Instead, the player specifies what type of character he wishes to play and the DM then either picks a class he has already developed or creates a new one. So the classes written down are more like examples than options. In an ideal world, the players wouldn’t even know what class their characters where using.

    3. Brendan

      Hedgehobbit wrote: the player specifies what type of character he wishes to play and the DM then either picks a class he has already developed or creates a new one

      I actually do exactly this as well. I am in the process of creating a “cultist of the elder animal gods” class (sort of a replacement for the druid) for one of the players in my current game. I do think that the main four classes can handle most character concepts pretty well though.

      Fundamentally, I share Talysman’s belief that a class is basically a method of solving problems. A new class is warranted if the character in question requires new mechanics for problem solving appropriate to the new concept.

  2. Brendan

    Minor note: classes whose “thing” is more resilient saves in general (dwarves, halflings) should use the numbers for the next higher tier when making saves. Thus, a first level dwarf would save as a fourth level character.

    Maybe thieves should also be classified as characters with better saves; not sure yet.

  3. frijoles junior

    Given the spread between the good and bad save tiers, it seems like there is some implied room for a middle tier, making it step by two both horizontally and vertically.

    I’m guessing as you intend it the lower tier is more of a baseline than a “bad” tier. But Maybe the unsophisticated holy blacksmith is good versus fire and demon magic but bad against charms. If his poison save is the same as his save versus charm, then we haven’t really captured his lack of sophistication.

    1. Brendan


      Yes, exactly. As Hedgehobbit was saying, class is really a crude proxy for concept. So a holy blacksmith could have saves distinguished as you describe.

      I also think it’s important to realize that a saving throw is really a form of second chance, so there doesn’t really need to be an “easy” tier. Just a hard and harder one.

    2. Keith Davies

      It could reasonably be the converse, though: normal people (of my level) can expect this degree of ass-saving, but I’m _good at it_ (higher save).

      The barbarian can withstand poison and other systemic damage because he’s _that tough_. A paladin can stand resolute against mental corruption because he is _that pure_. The rogue can trust his speed and balance to a phenomenal degree because he’s _that agile_.

      While saves do represent a ‘second chance’, I think there’s room for ‘that chance’ to be really quite good.

      I don’t mind there being a four-point difference between favorable and unfavorable. The extreme effectiveness comes in at high levels (as it should, and I wouldn’t mind seeing one more step at 20th level, even).

      As far as differentiating the mid-grade between favorable and unfavorable, applying ability score modifiers can probably handle that. The holy blacksmith might be expected to be hardy (good Con), which makes him better at poison saves, but relatively dull-witted (which dings his charm saves).

      Unless the class actively penalizes you somewhere, I don’t know that you need a step worse than unfavorable. Ability scores already provide some character differentiation


Leave a Reply