Players, DMs and Save or Die

[Grog]

I think that is more a problem of save or die monsters being under-CRed.

But how can you possibly even pick a CR for a save-or-die monster? How can you even balance something like that?

The only thing you can balance it against is the PCs access to Raise Dead and similar spells. Basically, you say that save-or-die is okay when used against PCs who have Raise Dead, and not okay when it isn't. But all that does is turn death into just another status effect, and contributes to the "early to dead, early to rise" problem that many of us have with 3.X D&D.

I think an important point that's gotten lost in this discussion is that save-or-die is an all-or-nothing effect, and D&D is not a game that's built for all-or-nothing effects. Never has been.

 

[FireLance]

But how can you possibly even pick a CR for a save-or-die monster? How can you even balance something like that?

The only thing you can balance it against is the PCs access to Raise Dead and similar spells. Basically, you say that save-or-die is okay when used against PCs who have Raise Dead, and not okay when it isn't. But all that does is turn death into just another status effect, and contributes to the "early to dead, early to rise" problem that many of us have with 3.X D&D.

Yes, in theory, save or die is balanced by its specific counter death ward. Raise dead balances it only imperfectly because of the level loss. Plus, as you pointed out, not everyone likes the flavor of easy death and easy raising.

I think another important aspect in balancing save or die is ensuring that the PCs will get sufficient warning and are able to prepare specific counters or avoid the encounter if the players do reasonably well. Unfortunately, this point is not emphasized in the rules or DM advice.

I think an important point that's gotten lost in this discussion is that save-or-die is an all-or-nothing effect, and D&D is not a game that's built for all-or-nothing effects. Never has been.

I would say that all-or-nothing effects aren't suitable for some gaming styles. Some players do enjoy the additional risk, though.

 

[Cadfan]

Yes, in theory, save or die is balanced by its specific counter death ward. Raise dead balances it only imperfectly because of the level loss. Plus, as you pointed out, not everyone likes the flavor of easy death and easy raising.

Just to toss this in here...

Raise dead doesn't work on death effects.

The ONLY counter for a death effect is Death Ward, until you can cast Resurrection at level 13.

So to recap...

If Death Ward works, Raise Dead doesn't. And vice versa.

 

[Hussar]

You need to go back and learn how to determine odds, my friend. So long as there is a chance to avoid death on each save, the chance of PC death is never 100%. In the case of, say, a medusa, you can also do things that lower the chance of death via the SoD effect....or even remove it altogether.

RC

RC, reread what I said. 20 saving throws. That's statistically going to result in a 1 at least once. That's 100% chance of fatality since a 1 always fails. There is no chance of avoiding death on a 1 and your chance of rolling a 1 on 20 rolls is close enough to 100%. Thus, that SoD encounter is lethal 100% of the time. (or close enough for this anyway)

Grog makes a good point. How do you peg the CR of a SoD effect? An EL par encounter should not result in fatality. It might, granted, since the baddie might get surprise, roll a crit, win initiative, roll another crit and obliterate a PC. OTOH, the party might kill the baddie before it acts as well. However, both situations are anomalies, not what you expect from an EL=par encounter. You expect the fight to last 3-5 rounds, the PC's to get smacked around a bit and carry on.

The whole CR system is predicated on that.

SoD breaks that completely. An EL par encounter with a SoD creature should result in a PC death almost 100% of the time. 3-5 rounds of everyone making saving throws is going to turn up that dreaded "1" very often. Never mind a bigger party as well. In a party of 6, SoD actually becomes far more deadly.

Straight up CR says that I should have to go about EL+4 or +5 before I should expect fatalities. But, if we whack up the CR on a bodak up to 13, it still doesn't work. It's far too weak at that level, but, it still has a small chance of PC fatality which a creature without a SoD effect would not have. If I jacked an unmodified gnoll to CR6, would it have a chance of killing a 6th level PC? I'm not sure, but, I'm pretty sure that his chances are far less than 4 tries at 5%.

SoD adds a layer of lethality to the game that is cannot be accounted for by the CR system. Anything that adds lethality to the game only hurts the PC's.

 

[Lonely Tylenol]

RC, reread what I said. 20 saving throws. That's statistically going to result in a 1 at least once.

Actually, it only results in at least one 1 64% of the time. The chance of rolling 1 on ndn is always ~63% after the die size gets above about a d10.

The chance of rolling 1 on ndn is specifically 1-((n-1/n)^n). It approaches a limit around .63 as n approaches infinity. Of course, if you roll more than n times, the chances go up significantly, and approach 100% as the number of rolls approaches infinity.

That's 100% chance of fatality since a 1 always fails. There is no chance of avoiding death on a 1 and your chance of rolling a 1 on 20 rolls is close enough to 100%. Thus, that SoD encounter is lethal 100% of the time. (or close enough for this anyway)

Well, 63% of the time, which is still not great.

 

[RFisher]

I think an important point that's gotten lost in this discussion is that save-or-die is an all-or-nothing effect, and D&D is not a game that's built for all-or-nothing effects. Never has been.

But it was. Men & Magic p. 20...

Failure to make the total indicated above results in the weapon having full effect, i.e. you are turned to stone, take full damage from dragon's breath, etc. Scoring the total indicated above (or something higher) means the weapon has no effect (death ray, polymorph, paralization, stone, or spell) or one-half effect (poison scoring one-half of the total possible hit damage and dragon's breath scoring one-half of its full damage.

 

[RFisher]

The ONLY counter for a death effect is Death Ward, until you can cast Resurrection at level 13.

IMHO, the counter to a death effect is not doing things that can kill you. If anything is "badly designed", it's having death effects that come out of the blue.

Although what qualifies as "out of the blue" can vary from group to group. The DM should adjust to the tastes of his group. (Yes, even to the point of eliminating or modifying death effects.) & it's OK if it isn't always perfect. It's just a game.

 

[RFisher]

The whole CR system is predicated on that.

The CR system was never meant to be that perfect. Even if it had been, you didn't have to play with it long to discover that it wasn't--whether "save or die" is there or not. If you keep expecting CR to be even close to perfect, the eliminating "save or die" isn't going to stop your disappointment.

 

[hong]

But it was. Men & Magic p. 20...

Yeah, it wasn't built for instakills even then.

 

[RFisher]

Yeah, it wasn't built for instakills even then.

Men & Magic, p. 31

Death Spell: An incantation which kills from 2–16 creatures with fewer than seven hit dice. The creatures must be within an area of 6" x 6" to come under the spell. Range: 24".

(For anyone unfamiliar with the convention, inches (") meant 10s of feet indoors & 10s of yards outdoors. Yeah, seems strange, but the logic or illogic behind it is a topic for another thread.)