Unchained Rogue: Is Powerful Sneak still a trap option?


Advice


Old Powerful Sneak was mathematically shown to be never worth taking, but what about the new version for the Unchained Rogue? Instead of counting 1s as 2s you instead get to reroll 1s; does this actually make the talent any better? Assuming your attack modifier was high enough to make the -2 to attack irrelevant, how much of a difference damage wise would Powerful Sneak make? I really have no idea how you would do the math on a conditional reroll like this.


Pathfinder Starfinder Maps, Starfinder Roleplaying Game Subscriber

The average would be:

((1+2+3+4+5+6)/6 +2+3+4+5+6)/6 per die.

23.5/6 vs 21/6, so a little more than .4 per die.


Pathfinder Adventure Path, Lost Omens, Rulebook Subscriber

Its still pretty bad, but it increases your average damage per sneak attack die from 3.5 to 3.917. At 10d6 of sneak its 4.17 average extra damage per attack.

So at its peak its a slightly better penalty to damage ratio than power attack with a 1 handed primary weapon.


Well, I'm not a mathemagician either. But it looks to me like BretI got it right.


Sneak attack average damage per die: 3.5 damage.

With Powerful Sneak: 3.72

With Deadly Sneak: 3.9

At level 20, with 10d6 sneak attack, it's 4 extra damage on average for one rogue talent and one advanced talent, and a -2 to accuracy.


BretI wrote:

The average would be:

((1+2+3+4+5+6)/6 +2+3+4+5+6)/6 per die.

23.5/6 vs 21/6, so a little more than .4 per die.

If you wanted to account for the accuracy drop you would multiply by 0.9, so that would be 3.525 average, or barely better than not using the talent in literally the best case scenario short of falling off the RNG so the penalty doesn't matter (it's actually slightly better than best case, but I digress).

I ran some maths on powerful and deadly sneak, and this is what I found.

Assuming NO other sources of damage, Deadly sneak and Powerful Sneak damage equals damage without using those talents when the chance to hit on non-talent attacks is as follows

Powerful: 0.94 - If you miss on a 2 while not using the talent, powerful sneak is decreasing your DPR no matter what. Wonderful
Deadly: 0.625 - If a roll of 8 misses without the talent, Deadly is never worth using.

What this means is that Powerful is useless. Deadly could potentially increase DPR, but considering that rogues hitting on an 8 isn't exactly common on most CR appropriate enemies the two talents spent on this are almost always going to waste or being actively harmful to your rogue. Remember that this assumes that there are no other sources of damage. As soon as you factor in the rogue's dex to damage, weapon dice and weapon enhancements Deadly moves straight back to utterly useless.

So in short, they both are still massive traps, but slightly less so than before. Progress.


For this talent to be the ok the 1's should have been made into 6's. It is still a really bad deal.
Edit: changed good to ok. 5 more to damage on sneeks at level 12, for a rogue talent and -2 to hit is only ok, not good.


1 person marked this as a favorite.
Cap. Darling wrote:
For this talent to be the good the 1's should have been made into 6's. It is still a really bad deal.

For this talent to be good it should have been a flat +1 to each sneak attack damage dice on powerful, and +2 on deadly.


Snowblind wrote:
Cap. Darling wrote:
For this talent to be the good the 1's should have been made into 6's. It is still a really bad deal.
For this talent to be good it should have been a flat +1 to each sneak attack damage dice on powerful, and +2 on deadly.

that would have been the easy fix yes:)


Thanks for the break down. My gut reaction was that it was still a bad choice, but I didn't know enough to prove or disprove that notion. It's good to know for sure!


Overused but still relevant internet meme


Secret Wizard wrote:

Sneak attack average damage per die: 3.5 damage.

With Powerful Sneak: 3.72

With Deadly Sneak: 3.9

At level 20, with 10d6 sneak attack, it's 4 extra damage on average for one rogue talent and one advanced talent, and a -2 to accuracy.

Unfortunately, though, a -2 To Hit translates to -4 Damage per hit over time, so burning 2 Talents that way earns you a net gain of -0.1 effectively.

You're better off using the Talents for things like Weapon Focus (which effectively adds +2 extra Damage because of the +1 To Hit).

TWF works for a Rogue because that -2 per Hit (and thus -4 to Damage) is quickly negated by the additional Sneak Attack damage; it's a total of -8 to Damage, but each Sneak Attack die is adding +3.5 per hit, meaning that by lv3, you're adding +14 Damage, subtracting -8, leaving you with a net gain of +6 damage. From there it grows by +7 Damage every other level, so the total -2 To Hit per attack becomes almost a non-issue.


chbgraphicarts wrote:
Secret Wizard wrote:

Sneak attack average damage per die: 3.5 damage.

With Powerful Sneak: 3.72

With Deadly Sneak: 3.9

At level 20, with 10d6 sneak attack, it's 4 extra damage on average for one rogue talent and one advanced talent, and a -2 to accuracy.

Unfortunately, though, a -2 To Hit translates to -4 Damage per hit over time, so burning 2 Talents that way earns you a net gain of -0.1 effectively.

You're better off using the Talents for things like Weapon Focus (which effectively adds +2 extra Damage because of the +1 To Hit).

TWF works for a Rogue because that -2 per Hit (and thus -4 to Damage) is quickly negated by the additional Sneak Attack damage; it's a total of -8 to Damage, but each Sneak Attack die is adding +3.5 per hit, meaning that by lv3, you're adding +14 Damage, subtracting -8, leaving you with a net gain of +6 damage. From there it grows by +7 Damage every other level, so the total -2 To Hit per attack becomes almost a non-issue.

How much damage +1and - 1to hit is worth is a bit more complex and situational than that.


Cap. Darling wrote:
How much damage +1and - 1to hit is worth is a bit more complex and situational than that.

True but the "+/-1 Attack : +/-2 Damage" ratio is a good quick-and-dirty assessment of the general value of an option.

It's why Power Attack for Two-Handed Fighters is generally a strong/necessary choice (-1 Attack for +3 Damage is considered a total +1 Damage per hit), while it's not a very great choice for Two-Weapon Fighters (+1 Attack, +2 Hit is considered a 0-sum value).

It's also why Weapon Focus and Weapon Specialization are regarded as very-good feats, despite the relatively-mundane effects (+1 to Hit and +2 to damage respectfully, which effectively translates to +4 Damage when all is said and done).


I wonder if paizo even knows how bad this talent is, and they wilfully ignore it?


CWheezy wrote:
I wonder if paizo even knows how bad this talent is, and they wilfully ignore it?

I am sure the read the messageboards here. I just dont know why they did not make it better(as in worth taking).


Maybe they don't think it is bad, because


CWheezy wrote:
I wonder if paizo even knows how bad this talent is, and they wilfully ignore it?

It took them how long to admit that they knew the rogue was bad?


wraithstrike wrote:
CWheezy wrote:
I wonder if paizo even knows how bad this talent is, and they wilfully ignore it?
I am sure the read the messageboards here. I just dont know why they did not make it better(as in worth taking).

I imagine that's what they thought they were doing when they changed it from 1=2 to 1=reroll. They did actually bump up the DPR with that change substantially. Shame nobody bothered to do the math and realize that the talents are still a waste of ink.

Of course, if they had done a public playtest they would have know that the improvement wasn't enough months ago.


1 person marked this as a favorite.
Snowblind wrote:


Of course, if they had done a public playtest they would have know that the improvement wasn't enough months ago.

Or if they thought about it for like, two minutes


CWheezy wrote:
Snowblind wrote:


Of course, if they had done a public playtest they would have know that the improvement wasn't enough months ago.
Or if they thought about it for like, two minutes

It's not two minute's work. It's about half an hour, and it requires a decent understanding of high school level probability to boot.

Since they apparently don't have the time and/or maths skills, I figured shoving the thinking across to the forums where there are definitely people with both seems like the best bet for getting decent well thought out stuff.


I feel like if you are the designer for a fifteen year old game where all the mechanics are resolved randomly, AND have had an internet community that discovered what is powerful and what is not, I think it is ok to not cut them slack for making a bad talent still really really bad


In some ways it's a matter of perception.

The average damage of 2d6, for instance, is only a difference of 7(unaltered), vs 7.11 (1&2 count as 2), vs 7.55 (1, 2, & 3 count as 3).

However, the minimum result is 2, which is pretty awful, even if it's only a 2.777% chance (accordingly, you ALSO only have a 2.777% chance of rolling a 12 for every 2 Sneak Attack Dice).

2d6 (Unaltered)
2 1/36
3 2/36
4 3/36
5 4/36
6 5/36
7 6/36
8 5/36
9 4/36
10 3/36
11 2/36
12 1/36

Unaltered, the Average Damage is 7, the graph is a Bell Curve, and the most likely result of every 2d6 rolled is 7. However, at only 6 out of 36 possible combinations, rolling a 7 is only a 16.666% chance, while rolling a 2-4 is also a 16.666%. Rolling a 2-6 is a mere 41.666% chance at that.

So while you have a 58.333% chance of rolling a 7 or greater, you have pretty much a 40% chance of rolling low.

It's understandable, then, that only a 60% chance of rolling a 7 or higher seems like pretty sad odds, when you consider that there's a 17% chance of a very-low result (2-4), and only a 23% chance in a moderately-low roll (5-6).

You have a 66.666% chance of rolling anywhere between a 5 and a 9, with the chances of rolling either very-low (2-4) or very-high (10-12) at 16.666% each.

2d6 (1 & 2 count as 2)
4 6/36
5 4/36
6 5/36
7 6/36
8 5/36
9 4/36
10 3/36
11 2/36
12 1/36

With 1 & 2 counting as 2, the Average Damage is increased by a mere .11, yes, but oddly enough the chances of rolling a 4 is equal to that of a 7 (6 possible combinations out of 36 each).

So, you still have a roughly-60% chance of rolling 7 or greater.

You also still have a 66.666% chance of rolling a 5-9, with the chances of rolling very-low or very-high being unaltered (16.666% each still).

HOWEVER, the difference NOW is that "very low" is 4, flat, while "very high" still runs 10-12. So if you have to have 16-17% chance of rolling low, the least you can do is have it be the highest of the low.

2d6 (1, 2, & 3 count as 3)
6 15/36
7 6/36
8 5/36
9 4/36
10 3/36
11 2/36
12 1/36

Here's where it comes down to a real matter of perception:

You now have a 41.666% chance of rolling a 6 for every 2d6 rolled.

This means that, again, while you still have a 60% chance of rolling a 7 or greater, the WORST you can do is a 6, which is frankly fine and dandy.

You can't even ROLL a 5 anymore. Instead, you have an 83.333% chance of rolling between 6 and 9, with that 16.666% chance of rolling very-high at 10-12 being simply icing on the cake.

---

Honestly, I can completely understand the mindset of wanting to assuredly hitting somewhere in the "decent damage" range of 6-9 and still having a 1/6 chance of hitting higher than that - in many ways, it sure beats the hell out of only ever hitting the "sweet spot" 2/3 of the time, and potentially rolling piddling values a whole 1/6 of the time.

While the theoretical, statistical average amount of damage is increased by merely fractional amounts (from 7/2d6 to 7.55/2d6 with 2 Talents), the hard fact is that assuredly dealing a minimum 6 damage per every 2 Sneak Attack dice is quite useful in the field in practice.

Knowing that, for instance, if you're level 10, you're attacking something with high DR, it's immensely more comforting to know that when you Sneak Attack you're going to do a minimum 9 damage + Weapon + Dex with a 60% chance of rolling 10 or greater, versus realizing with terror that you have a 4-out-of-10 chance of rolling only 5-9 and potentially barely denting your enemy.

---

So, yes, it's not a very good option when given a look through a statistical lens concerning DPR, and over the course of your career it may prove not terribly useful/relevant, but it may also prove useful a practicality standpoint regardless, due to increasing the minimum damage from 1/die to 3.

I'm still siding, by the way, that's it's NOT a great option, but I am saying that I can understand the allure and reasoning behind taking it: "Prepare for the Worst" vs "Hope for the Best" in this case.

The killer for me is that fact that it costs 2 Talents to pull off; if it were simply 1 Advanced Talent (Deadly Sneak), then it'd be an entirely different story, and I'd totally side with the "play to the minimums, not the average" mindset.

---

A simple fix, however, to help both statistics and practicality would probably have been to simply increase the Die type, from 1d6 to 1d8 with Powerful Sneak, and 1d8 to 1d10 with Deadly Sneak.


3 people marked this as a favorite.

I wonder if a simple fix for this is to simply remove the attack penalty. I don't think its going to all of a sudden make rogues into War Gods if you do that.


2 people marked this as a favorite.
Ventnor wrote:
I wonder if a simple fix for this is to simply remove the attack penalty. I don't think its going to all of a sudden make rogues into War Gods if you do that.

If they do that, Powerful Sneak becomes roughly equivalent to Weapon Spec for rogues (2.08 average damage at level 10), but only when sneak attacking.

That seems fair.


Ventnor wrote:
I wonder if a simple fix for this is to simply remove the attack penalty. I don't think its going to all of a sudden make rogues into War Gods if you do that.

That'd probably also do it, honestly.

Most Rogues are already going to be swinging at a -1 or -2 to Hit (depending on if they take Weapon Focus, which they more than likely will), because they'll be Two-Weapon Fighting.

Giving them an additional -2 to hit on TOP of that really bites into them in a way that screws up the usefulness of the Talent.


Snowblind wrote:
Cap. Darling wrote:
For this talent to be the good the 1's should have been made into 6's. It is still a really bad deal.
For this talent to be good it should have been a flat +1 to each sneak attack damage dice on powerful, and +2 on deadly.

That's how I've used these talents in my game all along.

The Exchange

Pathfinder Adventure Path, Lost Omens, Maps, Rulebook, Starfinder Roleplaying Game Subscriber

An alternative solution would be to have the talent act a bit more like Power Attack. Specifically the ability to activate the talent on the second attack. With the Debilitating Injury(Bewildered) the Rogue is atleast a bit more capable of hitting enemies after the 1st sneak attack (-4 AC against the rogues attacks at level 4), so being able to activate Powerful Sneak or Deadly Sneak after the 1st sneak attack hits would also make the ability more effective.


Pathfinder Adventure Path Subscriber

I like the flat +1 Powerful, +2 Deadly and the remove attack penalty ideas. Either would be acceptable.

The only reason why they didn't I think is maybe because of corner case Sap Master builds? Which make rogues almost look good. Dunno.


xevious573 wrote:
An alternative solution would be to have the talent act a bit more like Power Attack. Specifically the ability to activate the talent on the second attack. With the Debilitating Injury(Bewildered) the Rogue is atleast a bit more capable of hitting enemies after the 1st sneak attack (-4 AC against the rogues attacks at level 4), so being able to activate Powerful Sneak or Deadly Sneak after the 1st sneak attack hits would also make the ability more effective.

Are you missing on a 2 and hitting on a 19? Would your attacks do any damage through the target's DR/Hardness if you didn't have sneak attack. If the answer to both of the above is yes, then powerful sneak is never worth using.

Hence why it sucks horribly.

Deadly sneak is a little better, but it is still a minor DPR boost that only works on targets with very high DR or very low AC. For 2 talents this is terrible.

The Exchange

Pathfinder Adventure Path, Lost Omens, Maps, Rulebook, Starfinder Roleplaying Game Subscriber

For the record, I don't actually think the rogue talent is good either. I'd be perfectly fine with removing the penalty to hit and just going with minimum sneak attack damage die results of 2 for Powerful Sneak and 3 for Deadly Sneak (and there are much better options to take if one is a mythic Trickster in case someone decides to point out they have this exact same option). But I was more commenting that the penalty to hit is a little less crippling then it was before.


Both are not exactly overwhelming feats, but I guess in some situations they can be useful:

1) Opponent is flat-footed (e.g. coup de grace) and loses a serious amount of AC because of it
2) Opponent CR is unusually low (e.g. when many enemies), so their AC is also low
3) You have an excellent AB (e.g. by being a slayer and focusing your build around AB)

Personally, I'd modify the feats in my games, if the players agree:

Powerful Sneak: 4 damage per sneak die (saves a lot of rolling), -2 AB
Deadly Sneak: 4 damage per sneak die, no AB penalty


Still a clear trap option. The -2 to hit needs to be erased.

Also when discussing the DPR-effects you need to include the non-sneakattac dice based damage, because this damage gets hit by the -2, too. And you need to compare "more realistic" game situations, as flatly basing every calculation on a level 20 rogue with 10d6 damage scewers the bonus effect of the Talents massively into the positive. More D6s to influence = Talent seems more usefull. Almost all rogues will have less sneak D6s when fighting, yet all get the -2.


I'd consider it with Knife Master, but that -2 on a 3/4 BAB class...it is still a trap.


I'd auto-pick for every single Rogue I used it if it was +2 attack and made me reroll 6s on my sneak attacks.


Havoq wrote:

I'd consider it with Knife Master, but that -2 on a 3/4 BAB class...is a trap.

It's actually even worse on a knife master unless you are using something other than a knife.

The chance of getting a reroll are lower, but the accuracy loss is equal.

For reference, average damage per dice is...
Normal:4.5
Powerful:4.445
Deadly:4.725
Which is worse than on a d6


Snowblind wrote:
Havoq wrote:

I'd consider it with Knife Master, but that -2 on a 3/4 BAB class...is a trap.

It's actually even worse on a knife master unless you are using something other than a knife.

The chance of getting a reroll are lower, but the accuracy loss is equal.

For reference, average damage per dice is...
Normal:4.5
Powerful:4.445
Deadly:4.725
Which is worse than on a d6

d8 with rerolling 1s:

(4.5 + 2 + 3 + 4 + 5 + 6 + 7 + 8) / 8 = 4.9375

d8 with rerolling 1s and 2s:

(4.5 + 4.5 + 3 + 4 + 5 + 6 + 7 + 8) / 8 = 5.25

Seems better than with d6s, actually.


Makhno wrote:
Snowblind wrote:
Havoq wrote:

I'd consider it with Knife Master, but that -2 on a 3/4 BAB class...is a trap.

It's actually even worse on a knife master unless you are using something other than a knife.

The chance of getting a reroll are lower, but the accuracy loss is equal.

For reference, average damage per dice is...
Normal:4.5
Powerful:4.445
Deadly:4.725
Which is worse than on a d6

d8 with rerolling 1s:

(4.5 + 2 + 3 + 4 + 5 + 6 + 7 + 8) / 8 * (1-0.1) = 4.9375 * (1-0.1) = 4.445

d8 with rerolling 1s and 2s:

(4.5 + 4.5 + 3 + 4 + 5 + 6 + 7 + 8) / 8 * (1-0.1) = 5.25 * (1-0.1) = 4.725

Seems better than with d6s, actually.

You forgot to factor in the accuracy loss. See the bolded.

Sovereign Court

Snowblind wrote:
BretI wrote:

The average would be:

((1+2+3+4+5+6)/6 +2+3+4+5+6)/6 per die.

23.5/6 vs 21/6, so a little more than .4 per die.

If you wanted to account for the accuracy drop you would multiply by 0.9, so that would be 3.525 average, or barely better than not using the talent in literally the best case scenario short of falling off the RNG so the penalty doesn't matter (it's actually slightly better than best case, but I digress).

-2 to hit is almost always considerably more than a 10% drop in accuracy. Even if you would normally hit a target on a 3+, taking a -2 to hit would be a 11.11% accuracy drop. If you normally hit on a 11+, it'd be a 20% drop. The less accurate you are, the less beneficial it becomes. (similar to Power Attack - only not nearly as good) And, considering rogues still have a bit of an issue with accuracy...


to be worthwhile it should be either no negative modifier to attack, or trade every 1 in the sneak die for a 6. That would mean (6+2+3+4+5+6)/6, or 26/6=4.3 per die, which means 0.8 per die, or 0.4 per level (as rogues get +1d6 sneak each 2 levels, aprox)

As written, it's a trap


Charon's Little Helper wrote:
Snowblind wrote:
BretI wrote:

The average would be:

((1+2+3+4+5+6)/6 +2+3+4+5+6)/6 per die.

23.5/6 vs 21/6, so a little more than .4 per die.

If you wanted to account for the accuracy drop you would multiply by 0.9, so that would be 3.525 average, or barely better than not using the talent in literally the best case scenario short of falling off the RNG so the penalty doesn't matter (it's actually slightly better than best case, but I digress).
-2 to hit is almost always considerably more than a 10% drop in accuracy. Even if you would normally hit a target on a 3+, taking a -2 to hit would be a 11.11% accuracy drop. If you normally hit on a 11+, it'd be a 20% drop. The less accurate you are, the less beneficial it becomes. (similar to Power Attack - only not nearly as good) And, considering rogues still have a bit of an issue with accuracy...

It is the damage if you have a 100% chance to hit without the talent, and 90% with the talent.

This is slightly better than the best possible case that doesn't involve falling off the RNG (95%/85%). The calculation is nice because it is quick and easy and gives an upper bound to the damage dealt. If the damage with the talent gets worse under this system, then you know that the talent is never worth using unless faced with extremely high DR or AC that puts your to-hit roll off the RNG.

Community / Forums / Pathfinder / Pathfinder First Edition / Advice / Unchained Rogue: Is Powerful Sneak still a trap option? All Messageboards

Want to post a reply? Sign in.
Recent threads in Advice