Quote Originally Posted by Vahnavoi View Post
I think people overestimate both the ease and impact of doing so. Consider: just like in a dice-based game, not every action requires a roll, a game master in a card-based game can just pass and fail trivial actions without allowing spending and refreshing of cards. This means only impactful actions are left for "burning", at minimum costing time on the game clock.

For example, if a player is left with seven worst cards and wants to get a full hand for the next big event, now they have to plot seven moves in a limited time period where failing on purpose won't by itself screw them over. It shouldn't be assumed to be trivial.
That introduces a lot of onus on the GM to prop up the system, take an antagonistic stance towards what the players are trying to accomplish (e.g. it encourages proactively searching for reasons why they can't refresh their hand, when being able to 'take 20' on particular things would be distorting otherwise), and it also creates the potential for a lot of OOC negotiation bogging down play. I think it was the guy behind Arcen Games who said something like "If the optimal course of action in a game is tedious or boring, people will either feel compelled to do it and be bored, or will feel bad about the fact that they could have done better and chose not to". Controlling refresh and in general hand manipulation is a centrally important strategy for most card-based games; something where you refresh your hand when you use it up strongly signals 'if you have only bad cards, you should find trivial stuff to burn them on to get a refresh' as the actual strategic thing to do, so it's going to be that kind of situation for a lot of players. Given that some of the stuff I want to try to design away from is situations like this that already occur with D&D and some other dice-based systems, this isn't the correct design direction to go in.

This isn't saying that card based mechanics are never appropriate. But I'd use them for much more restricted segments of play rather than as a long-term resource, *especially* if basic system functions rely on someone always having a card at hand. Because then you must either deny agency of when to try things that would require a card, or just accept that the player will be able to get the maximum possible result whenever there isn't time pressure, even if there's consequence-of-failure pressure. On the other hand, if cards are an 'extra' then you can just not make it automatic to refresh them when you're out, and that resolves some of the issue. Or even if cards are needed for taking any action at all you can still say 'when you're out you're out and you can't act' and avoid the refresh metagame, but that doesn't work well when things external to the character can force a card to be played (like saving throws, or just the GM calling for checks for whatever reason).

Crusader mechanics are basically a card draw and refresh mechanic like this. They're great in combat, but once you get out of combat they really encourage weird little rituals to get access to infinite healing - start up a spar with your party members to draw a hand, punch a tree to trigger a heal if you drew one, otherwise surrender and then start the spar up again. It's silly and in practice the GM can and should just say 'no, you can't do that', but from the perspective of a designer it would be better to have avoided that in the first place so you're not handing every GM this little conflict to resolve every time they have a new player or group who has the same old idea.

So 'just replace dice with cards and play D&D' wouldn't be my approach.