new death save == leave your buddy on the floor for 3 rounds?

[stripes]

So the new mechanic is "as long as you don't take a crapton of damage, roll 1d20 each round 1-10: bleed, do this 3 times and you die. 9-19: no change. 20: stable, stop rolling".

This looked pretty lethal to me, 20s are rare (5%), 1-10 is common (50%). I figured it would be a lot of deaths.

So I tried it & boy was I wrong.

About 45% of the time a dying character becomes stable on their own (ignoring any ongoing damage, area effects, or monsters screwing with the body).

In fact on average they become stable in just over 3 rounds (like 3.2 rounds). They can never die in less then two rounds, so you can safely leave them there for 2 combat rounds and see if they stabiles slightly faster then average (about 25% of the time they will!)

When death comes it takes just over five rounds on average.

So you are "pretty safe" leaving a buddy on the floor for 3 rounds, and then panicking.

It's kind of intresting how the numbers work out.


Did anyone guess this intuitively?

 

[Syrsuro]

Guess it intuitively? Not exactly.

Run through the statistics for one of my players to calm his fears about its lethality? Certainly.

Carl

 

[Stalker0]

Some people have done the hardcore math on the probabilities, I'm sure the thread is around here somewhere.

 

[keterys]

And the rule for healing people with negative hit points is great encouragement to just do that, too... which is nice.

At least, so far it feels like bringing someone back up with a spell (less than Heal) won't just make them more likely to get actually killed.

 

[FadedC]

On the flip side with multiple combatants, more abilities that restrict movement and less in combat healing magic it's often more difficult to get to your buddy and bring him back up. So you might want to start thinking about healing your buddy well before "panic" time.

 

[Protagonist]

On the flip side with multiple combatants, more abilities that restrict movement and less in combat healing magic it's often more difficult to get to your buddy and bring him back up. So you might want to start thinking about healing your buddy well before "panic" time.

"damage aura" effects that deal damage to an area each round seem to be more common as well. Together with regular fireballs it seems quite likely that your unconcious pal might get blasted further into negatives while you wait

 

[baberg]

Quick question on the rule - it's not "3 failed saves in a row", it's "3 failed saves" right? So if I roll a 2-5-18-3 in four then I'm dead, correct?

And I thought it was 10+ is success, so it's 1-9 (45%) fail, 10-19 (50%) success, 20 (5%) miracle.

 

[Filcher]

On the flip side with multiple combatants, more abilities that restrict movement and less in combat healing magic it's often more difficult to get to your buddy and bring him back up. So you might want to start thinking about healing your buddy well before "panic" time.

I ran into this at D&D Exp. Our paladin got dropped behind enemy lines. He was staring daggers at me, the cleric, thinking "c'mon dude!"

And I was pointing out the carrion crawler and gelatinous cube, thinking "Sorry bro, but no way am I going through THAT to get to you."

Moving bodies BACK to someplace safe is almost as important as getting them healed.

 

[Revinor]

EDIT:

Please disregard this, as it was computed assuming 1-10 bleed, 11-19 nothing, 20 stabilize. Few posts below you will find correct data (1-9 bleed, 10-19 nothing, 20 stabilize).


According to my programming skills, chances are as follows

Round 1,  die 0.000000(0.000000), stabilize 0.050000(0.050000)
Round 2,  die 0.000000(0.000000), stabilize 0.097500(0.047500)
Round 3,  die 0.125000(0.125000), stabilize 0.142625(0.045125)
Round 4,  die 0.293750(0.168750), stabilize 0.179244(0.036619)
Round 5,  die 0.445625(0.151875), stabilize 0.205594(0.026350)
Round 6,  die 0.559531(0.113906), stabilize 0.223033(0.017439)
Round 7,  die 0.636418(0.076887), stabilize 0.233905(0.010872)
Round 8,  die 0.684857(0.048439), stabilize 0.240389(0.006484)
Round 9,  die 0.713920(0.029063), stabilize 0.244126(0.003738)
Round 10, die 0.730735(0.016815), stabilize 0.246224(0.002098)
Round 11, die 0.740193(0.009459), stabilize 0.247376(0.001152)
Round 12, die 0.745396(0.005202), stabilize 0.247998(0.000622)
Round 13, die 0.748205(0.002809), stabilize 0.248328(0.000330)
Round 14, die 0.749699(0.001494), stabilize 0.248501(0.000173)
Round 15, die 0.750483(0.000784), stabilize 0.248591(0.000090)
Round 16, die 0.750890(0.000407), stabilize 0.248638(0.000046)
Round 17, die 0.751100(0.000209), stabilize 0.248661(0.000024)
Round 18, die 0.751207(0.000107), stabilize 0.248673(0.000012)
Round 19, die 0.751261(0.000054), stabilize 0.248679(0.000006)
Round 20, die 0.751288(0.000027), stabilize 0.248682(0.000003)

Where number in front is a cumulative chance of such thing happening till that round, number inside brackets a chance of it happening exactly at given round.

So, rounding things off, there is around 25% of chance for getting stabilized on your own. Whatever is the outcome, chances of getting past 10th round without an answer are very very slim.

Numbers I got are bit different from what your reported. Have you computed them using math, run the random generator many times or computed the probability tree ? (I have done the last, given my rusty math skills)

EDIT:
You probably misinterpreted the rule that 3 failures have to be in row for a person to die. In such case, chances to self stabilize are indeed around 41%.

 

[Dragonblade]

Quick question on the rule - it's not "3 failed saves in a row", it's "3 failed saves" right? So if I roll a 2-5-18-3 in four then I'm dead, correct?

And I thought it was 10+ is success, so it's 1-9 (45%) fail, 10-19 (50%) success, 20 (5%) miracle.

Correct on both counts.