Discussion:5e: Repeating Saving Throws

Back to Main Page → Meta Pages → Discussions

Oportet (talk) 19:05, 5 January 2017 (MST)

So I've been noticing two pages I authored have undergone edits. Nothing wrong with that: we're a wiki, we edit things. However, it seems like they're variations of the same edit. They both add this line:

A creature affected with whatever can repeat the saving throw, whatever it may be, at the end of each of its turns, ending the effect on itself on a success.

I was wondering if there was a reason this specific edit was happening on two different pages. I checked the page history and found the edits were by different authors. Presumably, neither of them were in contact with each other. This suggests there's a bit of precedent that I'm missing out on.

While browsing the 5e SRD, I've seen this wording pop up in certain spells that cause conditions, but other spells omit repeating saving throws. After examining various spells, it appears that spells and items designed for use during combat allow you to repeat the saving throws, while those typically used for narrative purposes don't. For instance, dominate person doesn't require a save if the target isn't in combat, but once they start taking damage, saves need to be rolled.

If this is true, I'd like to see if one of the designers explained this decision at some point. I'm a newbie with D&D, and the game design thingymajigies I've seen so far discourage rolling the dice more than needed, so as to reduce downtime to a minimum. This blog puts it so highly as to make it principle 2 out of 4. What's D&D's secret?

Marasmusine (talk) 02:27, 6 January 2017 (MST)

The idea of repeating the saving throw for "combat" conditions is to replace the 3rd-edition style for variable length conditions (e.g. "lasts for 1d4 turns" or "lasts for 6 rounds").

Having to remember to count rounds is worse than having an additional die roll, especially when there are multiple effects with different lengths. Removing round-counting was actually done in 4th edition, which introduced a specific "ongoing" and "save ends" mechanic. You still get a variable length, and that length (on average) also depends on the target's defense against that effect. So a better method all round.

That blog is talking specifically about dice games rather than RPGs!

Oportet (talk) 18:00, 11 January 2017 (MST)

Hmm. I don't see what you mean by counting rounds being "worse" and repeating saving throws being "better all around". I don't dispute it being subjectively worse, but it seems a step too far to assume one method is objectively better. I can see a situation where a table needs to juggle millions of effects at once, but it's hard to quantify how much faster or slower one method would be. I would assume keeping track of effects could be as simple as using tally marks on scrap paper, if a DM's memory needs it. And even then, each DM has a different level of memory.

Additionally, I don't think basing length on a target's defense is always the best option. What if, instead of rolling to end an effect, you had to spend an action? Or if you were under the effects of poison, you had to actively find a cure? What if you could spare yourself from disease by cutting off the infected limb? Repeating saving throws doesn't involve choice; the player is not interacting with the game, and without interaction, there's isn't even a game. Sure, you could justify the lack of interactivity by giving the player something interesting to watch, but then you're losing the advantages of the medium. Doing is a whole different experience than watching. Sure, our stats need to mean something if they are to be valued, but we shouldn't value them too much. Our stats are not the only part of our characters. Our choices and actions define us just as much as our abilities, if not more so. I admit some things warrant a roll, but we should keep in mind the context and purpose of an effect when making players roll, and I mean more than just whether the effect's combat oriented.

A good example is alchemist's fire. A creature hit with alchemist's fire can use their action to try and extinguish the flames, making a check in the process. This means the check is in place of an attack, barring unusual circumstances. It involves a decision, but also factors in a player's stats, without the table needing to keep track of effect lengths. It's not perfect, but considering the circumstances when you'd be throwing bombs around, it fits the duty nicely. My point is there are times where other methods are preferable, regardless of whether you're in combat or not. It's in precedent too.

Off topic discourse: I would imagine, during 4th Edition's creation, the design team would be basing themselves off of playtest results. That's fine, but that can cause the game to swap one subjectively better standard for another subjectively better standard, depending on what their playtesters' preferences are. That's why we need game theory to validate what's "better", or else we exclude those we can't sample from. And if it's not their job to playtest the game, then why would they stick around to voice their displeasure? You can always play a different, "better" game. Playtesting is a great method for checking whether theory works, but it shouldn't replace theory.

P.S.: As long as D&D uses dice, a case can be made to apply dice game theory. Just because you're playing an RPG doesn't mean turn length isn't extended when you roll a die. Just because you're roleplaying doesn't mean people want to sit around waiting on dice rolls. Think of it like this: if a genre's use is as a frame of reference, and if the frame isn't covering the whole picture, the "frame" loses meaning.