Altering items is perfectly safe, as is inserting additional rows to add new items. The only place you need to be careful about inserting stuff is weapons, as there are some custom ranges there that you need to be careful to handle correctly.

My current weapons are [Librarian's Paper Cutter] and [Anarchy]. All this time I have been main-handing LPC with Instant Poison and off-handing Anarchy with Deadly Poison. So far, so standard.

However, I have recently been playing around with this spreadsheet and it seems to indicate that if I were to switch my weapons around, that is, to main-hand Anarchy with DP and off-hand LPC with IP, I would gain an additional 0.21% DPS.

This seems to be contradictory both to accepted wisdom which is to MH the faster weapon with IP and OH the slower weapon with DP and Vulajin's Roguecraft Spreadsheet which indicates that doing so is a 0.65% DPS loss. On the other hand, I understand that this spreadsheet is supposed to model Mutilate more accurately.

Can anyone shed any light on this or are the differences small enough to be mathematically negligible?

So, fundamentally what's going on is this: while at equal DPS, it is clearly superior to keep the faster weapon in the MH. However, when there is a large DPS discrepancy between the weapons - such as the 26 DPS difference between LPC and Anarchy - the loss of physical damage from both Mutilates and autoattacks can be sufficient to offset the extra IP procs on finishers. So that's why it's showing up with Anarchy as better in the MH.

Note that the conventional wisdom presupposed weapons of comparable DPS. And even if they're only 1 tier apart - such as WD and SR - it's close enough to still be true. But when they get far enough apart - such as in your case - the conventional wisdom may break down.

So, the remaining question is: why the disagreement with Vulajin's sheet? Well, among other differences, one of the major distinctions between this sheet and Vulajin's is that it has a more sophisticated poison stacking model. I know Vulajin's sheet makes some allowance for the fact that you may not always have 5 stacks of DP while envenoming, but I believe the model for such is somewhat primitive (owing in part to the fact that the sheet has sophistication in other areas that this sheet lacks, which makes it harder to include sophistication in this). And what this has meant in practice is that this sheet tends to like a fast weapon to apply DP more than Vulajin's. Both sheets agree that you want the weapon attacking more to have IP and the one attacking less to have DP; where they disagree is how well damage scales with the speed of the DP weapon.

So: when applying DP with a slow weapon like Anarchy (or Sinister Revenge), you don't always have full stacks for Enevenoms, particularly when you are only doing 1 Mutilate between finishers. Thus, getting the extra procs from finishers has some DPS value for DP. Note also that this is not just any additional proc; the proc gained from an Envenom is a proc at the one time when you are absolutely guaranteed to have low DP stacks. Hence, it provides more DPS than a randomly timed proc of equal frequency.

So the fundamental difference between the two sheets in this matter is simply this: this sheet places more value on that proc, and believes that, coupled with the higher physical damage from the higher DPS dagger, it is sufficient to offset the loss of IP procs from those finishers. Vulajin's sheet assigns this proc a lower value, and thus finds LPC MH to still be superior - hence the difference between the two sheets. As to which is actually correct in the matter - that's hard to say. The real answer lies somewhere in between, and thus lies in a grey area where it's not immediately clear which is truly better.

Is it known for sure what order thing occur when Envenoming? You seem to be assuming that if the MH procs DP, this occurs after the Envenom clears the stack rather than beforehand. Has this been tested? If not, I'll try and find time to do it over the weekend. It's simple enough to test:

1) Build 5 CPs, make sure DP stack is at 5, and /stopattack
2) Envenom and /stopattack

If DP procs after the Envenom clears the stack, you should see a single DP stack on the dummy ~40% of the time (or 55% of the time if the Envenom buff applies before the poison procs). To test the converse, you need to build 4 DP stacks and 4 CPS in step 1. If the Envenom in step (2) matches the damage range for a full 5 point Envenom, then DP must have procced off the Envenom before the stack was cleared.

This is purely speculation, but I'd imagine the proc occurs after the stack is removed.

You gain the envenom buff even if envenom is dodged or doesn't hit, so this would lead me to believe the application of the buff happens before the chance to proc occurs.

From there, I'm just assuming the DP stack is removed when the envenom buff applies, but I am not certain. This would be easy to test by using gear with 0 expertise and envenoming in front of the heroic dummy until you get a dodge or parry and check if the envenom stack was removed.

Well, according the the combat log, the order is:

1) Deadly Poison stacks removed.
2) Envenom buff activates.
3) Any poison procs generated by the finisher occur.

Furthermore, by standing sidewise to the target dummy (so as to disable autoattacks) with Deadly Poison on, I was able to stack DP 5 times on the dummy, perform a 5-pt envenom, and wind up with a single stack ticking afterwards, which seems to me to be fairly strong evidence that the proc is checked after stacks are removed. Thus, I am assuming that the order in the combat log is correct.

It's not so much that it's no longer optimal be capped, as that the loss of other stats to become capped is no longer worth it. By using items that are independantly better, you can get to 209 expertise; when it was though that swapping out for the gear that gives 214 expertise was giving a full .25% dodge reduction, doing so proved to be worth it; now that we know that it only gives about .13% dodge reduction, the other stats pull ahead.

Based on the Wowhead spell details for Find Weakness, it looks like they fixed the bug with it giving extra damage to Rupture (possibly in 3.0.8?).

Quick in-game test confirms that the bug is indeed fixed.

New version in the first post. Updated for 3.0.9, as well as the rupture change noted in the preceding two posts, and a minor bugfix that was causing the sheet to slightly overestimate Envenom damage.

I'm showing about a 300 DPS increase from the changes, with the only itemizion change being Strong-Handed Ring nosing ahead of Ring of Invincibility in the currently-optimal gear setup.

Honestly i would have assumed the Mutilate nerf would have hit harder.

It's a 20% nerf to 20% of our total damage, of course it's not going to hit very hard - 4% or so of the total. The S'n'D buff is a ~7.5% buff to ~50% of our damage (white damage + IP damage, even neglecting second order effects on Focused Attacks etc.), so it almost exactly cancels it out. HfB buff on top is what makes the difference. Overall I wouldn't be at all surprised to see dual IP pushing significantly ahead of IP/DP now, since an IP/IP setup double-dips on the S'n'D buff while ignoring the Mutilate nerf.

Can someone with PTR access check what's going on with HfB? As it stands from the patch notes, it looks like a simple buff to it, with no change in number of stacks and no bleed requirement. It may be that got rolled back, or it may be planned to go in with 3.1 and this is a temporary change.

I'm guessing this is a temporary change, and they're still planning the bleed version. This patch has the feel of a short-term fix - tweak a couple constants here and there to reduce the potency of rogue burst in PvP and give a slight PvE buff. I'm sure we'll see more significant changes when 3.1 rolls around.

So WD/Murder > WD/SR now, (which is very logical), and WD/WD still good value ahead both other.

It's actually a 30% reduction. 50% -> 20% = 30%

This is a very interesting point.

If IP/IP becomes the way to go, then perhaps Sinister Revenge's days as a best-in-slot OH are over?

But 1.5/1.2 = .8 = 20% reduction, which is the more relevant part.

You're wrong Blakkheim. You haave to look at it this way:

Mutilate prenerf: damage_mutilate*1.5 (50% more damage)
Mutilate nerfed: damage_mutilate*1.2 (20% more damage)

What you need to do is divide the two to get the correct number.

(damage_mutilate*1.2) / (damage_mutilate*1.5)

Since damage_mutilate is in both factors we can remove them (x/x = 1).

1.2/1.5 = 0.8, or a 20% nerf.

EDIT: So much for trying to be educational.

I ran the numbers on double-IP. Provided I didn't make any mistakes (which I'm not going to swear to, as it is late and I am about to go to sleep), swapping Turn the Tables for Improved Eviscerate and running 2xIP with eviscerate does damage almost exactly identical to the IP/DP/Envenom setup. I'll double check my numbers in the morning, but for the moment: it is not better. It is not worse. It is as close to exactly equal as you will ever see.

Did you take the glyph of Evis into account in the above calcs Aldriana?

I was wondering about that too, but AFAIK glyph of evis is kinda standard for mutilate builds, so I think it's safe to assume he did take it into account .

How is it standard? it sometimes gets left there as I reglyph from a pvp or HaT setup, but it provides no benefit to a PvE mut setup, so that slot is just as likely to end up being Glyph of Sinister Strike or Glyph of Feint.

i think that using evi as finisher instead of envenom, it would be pretty standard.