Ilbranteloth
Explorer
Would the first example be considered differently if the character had taken a significant amount of poison damage?
As a pure example, a trap in one system has a 50% chance to kill a character. A trap in another deals 50% of their HP.
In one,either the character either knows of the existence of the threat and can start taking measures against that knowing that the ext one could kill them, or they're dead.
In the other, they know of the threat and can start taking measures, knowing that the next one will probably kill them.
There is suspense either way.
Yes, there is. But the stakes are higher with save or die effects.
Again, it's not for everybody. And I can really only speak to my experiences, but we still have save or die effects because that's what my players prefer. Along with no resurrection.
We've tried a lot of different approaches over the years, and for us it comes down to the same thing. The save or die approach works better for us.
Also, the thing to recognize about the pit example is that it is possible for a PC to die. It's just that the way it's designed the chance is much lower than the players think. They think they got lucky. Why would that be?
I think it's because of one of the things that is usually used as an example as to why OD&D and AD&D is bad game design. There are three different mechanics, all with different math, and two of which are non-standard. That is, not only do the players not really know what the rules are, but in the heat of the moment it's not that easy to calculate the odds. Layered checks has that effect as well, and was leveraged in the 4e skill challenge design. Even if it's not save or die, this approach will work. But I think with higher stakes, it works even better.
And I guess that's kind of why we keep ending up really liking the approach. Because from the players' point of view it consistently builds suspense. The two false entrances were designed, in my opinion, for similar purposes. One does a huge amount of damage, no save. It can be detected, but the players have to be smart. There's also a good chance that it will be hirelings/henchmen that fall victim to it. But it tells the players right up front that the adventure doesn't pull any punches. Except, as I pointed out with the pits, it's somewhat of an illusion. After that trap, there are a handful of truly deadly traps (not including the poison that is largely at the start of the adventure), but most of them seem that way. It repeatedly feels like you barely escaped death.
The other false entrance uses a different mechanic, but one that comes up several times in the adventure. The DM starts counting backwards. Yes, it's a metagame approach, the players know something is up, and they run. There are two (maybe three) other locations where this occurs, and oddly, the 5e version only retained the last one (replacing the others with Ability checks). So the final version doesn't really work either, because they haven't experienced it earlier in the dungeon.
There are lots of ways to approach these sort of things. In our case, I make a note of what has worked well for us so I can try to learn from it. And it was particularly evident when we tried to figure out why ToH in 5e didn't work for us, where the original did. Our initial theory was that it had to do with how fast healing is in 5e. But the reality is, even though it takes much longer, PCs could do the same thing in AD&D. Stop to rest, have the cleric heal everybody, do the same thing for a second day if needed, and then start again the next day after everybody is fresh.
The players never did that, mostly because hit point loss wasn't the biggest drain in the adventure. It plays like an Indiana Jones adventure - either you make it past a given trap, or you don't. Which is exactly the way we think it should be. If this is a tomb that has killed everybody who has ever tried to plunder it, for hundreds if not thousands of years, then it should be very deadly.
So back to your example - I think that a trap that has a 10% chance of killing a PC outright still feels more deadly, and will be treated with more respect than one that can deal 50% of their hit points. Even better if they think the chance they will die is higher. It's kind of a balance, if they think it's too likely they will die, they won't try it at all. But the point is, unless there is some other immediate danger, being reduced even to 0 hits points is just a speed bump in 5e. Especially outside of combat or some other circumstance that makes recovery less likely.
Here's an alternative mechanic for you, and let me know how you think your players would react to this mechanic.
You cannot recover by making a successful death save while you suffer the poisoned condition.
So you still have to roll three actual failures. But the "successes" don't count toward your three successes to recover until the poison has run its course, or an antidote or neutralize poison is used. They just postpone the failures.
It's no longer save or die, but it would make players try to avoid being dropped to 0 while fighting venomous creatures.