Sunday, September 21, 2014

Why I Use the Classic Saving Throw System


Blackleaf didn't get a Saving Throw, and we know
how that ended
Saving Throws are an iconic element of table top  roleplaying games, that likely has its roots in the First Edition of Dungeons and Dragons, those Little Brown Books (well before that really) .  Saving Throws are still a part of Dungeons & Dragons 5th edition, but frankly I think they’ve lost something.  Don’t get me wrong, I like 5th edition a lot, and have enjoyed the flexibility of the character generation, the careful balancing of armor class (a real problem area if one is trying to limit power creep) and how despite its heroic elements 5e has maintained much of the feeling of character peril one might get from Basic/Expert style D&D.

Yet 5e does something strange with Saving Throws, something I think is a holdover from newer editions of D&D, in that it links them to character statistics.  This is a huge departure from the LBB’s and the editions that followed them.  In early editions saving throws are static based on level (with a bonus for a high Wisdom in some editions).  I like this system; I also like the eclectic names of the classic saving throws “Death Ray or Poison, All Wands Including Polymorph and Paralization, Stone, Dragon Breath and Staves & Spells”.  I like the way Saving Throws are managed in the LBB’s (and similar systems) because they are related to class and level, without consideration for ability scores.  Likewise the variety of saving throws are bizarre, but clearly they all relate to terrible, likely deadly effects and seem so specific that they encourage adventure designers and GMs to expand their use into other areas/against other dangers.


This classic system of Saving Throws makes them something of dire consequence, and indeed the name implies this.  The player is offered a single roll with a (fairly small to begin with) chance to ‘save’ the character.  I always saw this in the manner of Conan or a similar swords & sorcery hero resisting the vile magic of some evil sorcerer, much to the surprise of the sorcerer, and largely because they were the hero in the story, able to withstand  magics and trials beyond those of normal men to face. Yes, the Saving Throw is a weak form of plot immunity, based on how accomplished and long played a character is. This is just the diegetic understanding of the saving throw, but mechanically it has a similar purpose.

Saving Throws become terribly important in Mid-Level games, especially in systems with higher hit-point and Armor Class Progressions.  This isn’t to say physical Hit Point damage isn’t a danger in games with mid-level characters, but it’s not as much an immediate danger as in a game where one has 5D6+1 HP (A 5th level LBB fighter – and the LBBs are stingy with HP compared to later editions) and normal attacks do 1D6 damage, including such attacks as a dragon’s bite.  In games like this melee may be dangerous, but it’s a predictable danger at higher levels with damage slowly wearing down HP (unlike first level when fighter have 2-7 HP) – another resource to track.  Attacks like dragon breath (apparently inflicting the full HP of the dragon), lightning bolt and fire ball are some of the few attacks that do greater than 1D6 damage, and can be quite catastrophic in this system.  Likewise monster poison, polymorph and petrification (wand, stave or otherwise) result in immediate death or removal from play.  At mid-levels however these sorts of effects become more common, as monstrous opposition increases in potency and right around the time that normal melee become less immediately dangerous.  The Saving Throw effectively creates a great deal of the character risk in a mid-level game. 

Additionally, if one is running the LBBs by the book, it’s at sixth level that the raise dead spell becomes available, and hence poison and death by assorted spell become less permanent. Remove Curse and Neutralize Poison are available to Clerics at 5th level, so poison and polymorph are less dangerous to a well prepared party as well.  Only petrification requires the high level spell “Stone to Flesh” which a wizard can learn at 12th level, but a character reduced to statuary is less permanently dead and more a logistical and patronage problem for the party to revive.  


In a recent post on another blog I saw someone bemoaning the existence of a separate “stone” Saving Throw, apparently because beside the singular Medusa there aren’t any/many mythical creatures of legend that turn people to stone.  While this is the sort of gripe that I find absurd when discussing a table top game about fighting strange creatures in inexplicable labyrinths, it’s true that the classic saving throws are a bit bizarre.  Yet this isn’t a complaint worth addressing at length, any more than complaints about the need for variable weapon damage or other ‘realism’ based rules modification are – it’s a game (largely about exploring strange labyrinths, recovering enormous amounts of gold and fighting strange mythical creatures) and its mechanics are rules, not simulation.

Poison/Death (though I use a separate death save – but that’s a different story), Wand, Stone, Breath and Spell are silly names, but they aren’t required to be exacting categories, what they really model are different character classes ability to withstand different types of danger.  Magic-Users are susceptible to Poison and Wands while Clerics are more resistant, but Magic-Users have the best ability to withstand being turned to Stone and are good at mitigating spell damage.  Fighters on the other hand are best against Breath and worst against Spells.  The Saving Throw then is a means of creating class difference in a system with very few other distinctions.  If one doesn’t like the ‘silly’ names, perhaps rename them, “Stone” can become “Wyrd” and “Poison” become “Natural Physical” or whatever else feels properly serious.  The distinction here isn’t really between type of risk, it’s between the character class taking that risk and so a Saving Throw table can become an excellent additional mean of distinguishing a class (great saving throws are one of the halfling’s best abilities in Basic and similar systems for example). 

It’s tempting when examining the classic Saving Throw system to create a single save (as in Swords and Wizardry) with modifications by race or class, or to tie Saving Throws into character statistics (after all is resisting poison is a measure of health and endurance, shouldn’t it be effected by Constitution?), but in doing so one removes a useful tool. Saving Throws, as broken out in the traditional categories provide an important rules mechanic that is different from and independent of other ‘survival mechanics’, such as the combat mechanic of Hit Points and Armor Class, or the general mechanic of Ability Score checks.  By providing another metric entirely based around level and class, reserved for dread dangers and dependent on a single roll, the Saving Throw adds something to the game by providing a GM or adventure design tool that differentiates challenges and creates a different kind of player reaction.  There is a reason that players often look aghast when the GM says “I need a save vs. poison” and it’s because the default use for the poison Saving Throw is for the character to avoid death for guessing wrong about a risk, or otherwise engaging in some dangerous task (like fighting a giant spider) and performing less than optimally. 

Apparently while I was writing this up the fellow over at B/X Blackrazor has been thinking about similar things, and come to an alternate and perhaps entirely opposite conclusion.  I will keep the classic Saving Throws in my games, and have done so, re-purposing them to cover more areas (Wands is also the standard save in HMS Apollyon against small bombs, shotgun and automatic weapon fire – including the instantly deadly heavy machine gun, while Breath is reserved for large explosions and artillery bursts), though I have been considering a rewrite of the Saving Throw tables for some of the classes to create more variation, especially amongst demi-humans.  The key for me in deciding to stick with the classic category based Saving Throw is that not only does it provide a different metric based on class and level, but that the progression can be different.  Stat based saves will average out, as do 3D6 statistics, and rarely increase (unless you use 5E’s stat increase system), while ones based on class and level can be modified both by level progression and modifier depending on the needs of setting or GM.  For example, Flying Monkeys aboard the Apollyon are very potent characters, with almost the full abilities of fighters, several thief skills and the ability to fly – yet being part of a species designed and conditioned as magical thralls has it’s downsides and the Flying Monkey always remains susceptible to spells, especially mid effecting ones thanks to the design of their saving throw table.


  1. Thanks for writing this. I'd been little down on saving throws after reading JB's B/X Blackrazor posts. I do like his reasoning for his own spinoff game, but you're absolutely right... the save is iconic for D&D, and it doesn't need to be realistic.

  2. Absolutely right on. The classic saving throws helped to create the overall fantastical vibe of OD&D. They weren't just part of the mechanic. They were part of the aesthetic. It's much more fun to Save vs. Dragon Breath than to, say, make a dodge roll based on your dexterity score.

  3. Flying monkeys don't do well against whole body eating devils either...

  4. I'm already sold on old school saving throws. I really want to love them ...but there's something holding me back: they're a "dissociated mechanic". I have no idea what they represent or even if they're meant to represent anything in the game world when you get right down to it. Is saving vs a fireball really leaping out of the blast? Then shouldn't wizards be terrible at it? Should the character be moved to edge of the blast? Are demi-humans good at saves because they're small? etc, etc.

    I feel like to use them properly I'd have to go through the game and collect evidence about what they're supposed to represent by looking at who's good at what and everything that modifies them.

    1. It's pretty simple really
      1) 'poison' (easiest) "physical resilience" Clerics: good Fighters: ok Mu: bad - presumably clerics have divine protection and fighters are tough
      2) 'wands/paralysis' (moderate) "magical directed" Clerics: good, Fighters: ok, Mu: bad - again divine protection for clerics and physical ability for fighters
      3)'Petrification' (moderate) "magical natural/divine" Clerics: moderate, Fighter: Moderate Mu: Good - magic users know how to short circuit these sorts of undirected magical attacks
      5) 'Breath' (hard) "nasty physical area effect attacks Cleric: Bad, Fighter: Moderate, Mu: Bad - it takes an athletic individual to tough out a bomb blast with only a shield.
      6) 'Spells' (hard) "nasty magical area effect" Cleric: Moderate, Fighter: bad, Mu: Moderate - you need some sort of magical or divine protection to survive phantasmal cosmic fires.

      Sure they are dissociated, but most everything is in D&D - HP? Armor Class? HP represents a simplification of various factors allowing survival - you decide in your game if it's luck, brawn, magical warding or whatever. AC, same thing - is it inability to hit?, ability to suffer minor wounds? - lots of things grant AC in various ways. Same with Saving Throws - they are a rule mechanic for modelling the survival chance form terribly largely lethal effects. A table top game where everything 'represented something' would be really complex.

    2. Thanks man that was a big help!

  5. I've played D&D since the beginning, while the saving throws category names are cool sounding for the game, quite frankly, the tables themselves are not. I agree that as a fantasy game, the class should factor into the saving throws. But leave the old school tables in the past. The less tables that are needed for the game, the better.
    I can assure you that many times errors in reading the table or forgetting to update characters from the tables only ended in heart ache or if the GM was a softie, then rewriting the story, which is just as bad.

    1. The real point isn't the cool names - as mentioned above, Saving Throws unrelated to stats are a plus because they devalue Stats as a primary source of mechanical 'power' and differentiation for characters. Like low stat bonuses they then discourage stat inflation and allow more variety in playable character statistics, thus making stat checks a useful mechanic.

      I agree on the minimization of tables, and really one could set up old style saves without a table (i.e. your fighter's base poison save is 16 minus Level/3 or something). Still I am unconcerned with the idea that either character death will cause heartache (at least I should hope not in the sort of games I run) due to transcription errors or table reading errors will be common enough to damage gameplay. The difficulty of looking at a saving throw table to update a character sheet every three levels or so is minimal compared to the advantage that another success metric provides.