Sneak Attack Damage

[Gorr_the_Gastly]

Am I the only one who thinks that rogues damage should not be based on extra dice rolled. I understand that it is suppose to be that maybe if you roll bad you only made a glancing blow at a near vital area. But it doesn't feel like the sneak attack dice should be the way a rogue deals its extra damage.

I was wondering if maybe an additional 1d6 is fine, but maybe the sneak attack damage doubles, triples, quadruples, etc in place of additional damage. So at 20th level you roll a 1d6 and multiply its results by say 9 times. Is this a worthwhile idea or not?

[torrasque666]

well, if you're using a d6 weapon for it it already does that, at least in the manner of criticals.(my understanding is that crits multiply the dice, not the static things from strength or enhancements)

[Anxe]

So that the Sneak Attack just has a higher variance in damage it deals? 9d6 will pretty much always be around 30 extra damage. 9x1d6 will have static steps from 9-45. (9-54)

I'm a little confused. Your method actually favors the glancing blow idea, but the language of your post makes me think you were trying to avoid that. Am I missing something? Could you clarify?

EDIT: Math mistake.

[Mountain]

The only difference between 1d6x9 and 9d6 is that the 1d6x9 will be much more variable. 9d6 would generally be pretty close to the average (31.5), while 1d6x9 would be just as likely to give you the minimum (9) as the maximum (54).

[Anxe]

well, if you're using a d6 weapon for it it already does that, at least in the manner of criticals.(my understanding is that crits multiply the dice, not the static things from strength or enhancements)

Not how it works by RAW.

"A critical hit means that you roll your damage more than once, with all your usual bonuses, and add the rolls together. Unless otherwise specified, the threat range for a critical hit on an attack roll is 20, and the multiplier is ×2.

Exception: Extra damage dice over and above a weapon’s normal damage is not multiplied when you score a critical hit."

Sneak Attack damage is not multiplied by crits because of that extra damage exception. The +2 on a weapon would be multiplied, but the flaming would not. Strength bonus is multiplied as well.

[torrasque666]

Huh, you learn something new. Now I need to have a talk with one of my DMs. He has a triple crit rule where you do max if you land 3 20s and his interpretation of someone who does 1d10+1 with a x2 mult is that it would be 21. Not much changes at that level, but you see how that could be a problem with higher damage.

[Anxe]

It's not like that's a bad way to play. It's just a houserule instead of RAW.

[Gorr_the_Gastly]

Your medium is based off 3 right?
Well you have a higher chance of rolling a one out of nine times then you do with one right?

Also there exist a feat to always roll medium damage on the dice.

[Brookshw]

I was wondering if maybe an additional 1d6 is fine, but maybe the sneak attack damage doubles, triples, quadruples, etc in place of additional damage. So at 20th level you roll a 1d6 and multiply its results by say 9 times. Is this a worthwhile idea or not?

So, 2e then :smallconfused: fine by me.

[Shining Wrath]

Your medium is based off 3 right?
Well you have a higher chance of rolling a one out of nine times then you do with one right?

Also there exist a feat to always roll medium damage on the dice.

The average roll on a D6 is 3.5. 1+2+3+4+5+6=21/6=3.5.

I think Rogue sneak attack damage ought to depend on size. A small gnome should not be able to do extra damage to a colossal dragon just by attacking their tail.

[Afgncaap5]

Technically the medium is more like 3.5 points of damage per 1d6. Or to use whole numbers, the medium (or average (by which I mean mean)) damage would be 7 points per 2d6.

Edit: Swordsaged.

[dascarletm]

The average roll on a D6 is 3.5. 1+2+3+4+5+6=21/6=3.5.

I think Rogue sneak attack damage ought to depend on size. A small gnome should not be able to do extra damage to a colossal dragon just by attacking their tail.

Small characters can't have nice things.

[grarrrg]

Well you have a higher chance of rolling a one out of nine times then you do with one right?

...Yes...but you also have a higher chance of rolling a 6. What's your point?

I know what you're trying to say, but rolling many dice it is very unlikely that you will botch more than a couple rolls, and just as likely that you'll get a couple good rolls. So it balances out, and you wind up 'in the middle' the vast majority of the time.

If you roll 1 die and multiply it, then you are at the full mercy of the dice gods, and your chances of botch, massive damage, and average damage are all identical.

Let's look at a simple example.
1d6 multiply result by 2
vs.
2d6

In the first example, your have a 1/6 chance of getting a 2 (1x2), or a 1/6 chance of getting a 4 (2x2), or a 1/6...etc... All results are equally likely.
But with 2d6, you have a 1/36 chance of 2, 1/18 chance of 3, and 1/12 chance of a 4. Add that up and you have a 1/6 chance of getting a 4 or less.
Compare that with the first example, in the first one, your odds of getting a 4 or less are 1/3. Much more likely* that you'll get a poor roll with Multiplying. You are equally likely to get a super good roll with Multiplying.


If you really want to change something, then forget dice and start adding static modifiers. +3.5 damage per "sneak dice" instead of rolling. Or maybe keep have the dice, and replace every other with a +3.5 bonus. Then all your Sneaks with do similar damage, and there is little worry of getting a lot of 1's.

*Yes, there is some minor math fudging, as 5's should be considered in there somewhere, so if you take half the 5's and consider them as 4's then you have an 2/9 chance of a "4" or less.

[Hangwind]

The average roll on a D6 is 3.5. 1+2+3+4+5+6=21/6=3.5.

I think Rogue sneak attack damage ought to depend on size. A small gnome should not be able to do extra damage to a colossal dragon just by attacking their tail.

That, um...wasn't his tail.:smallyuk:

[Mato]

Well you have a higher chance of rolling a one out of nine times then you do with one right?You have a 0.000009% chance of rolling a total of '9' using 9d6. And it's the same odds of rolling a total of '54'.

The more dice you add the more is trends towards the average. For example, in a 2d6 roll the total of '7' can be made out of six possible combinations (1+6, 2+5, 3+4, 4+3, 5+2, 6+1) whereas the total of '1' only occurs when both of the dice are a certain value (1+1).