Ah, bugger. This bug was in the last revision but I forgot to check if it was still around - guess it is.

The basic problem is that OO doesn't like the "IF" statement in the Allowed Values Paramater.... as a simple fix, go to Data -> Validity -> Criteria -> Source and change the contents of that cell to "MetaGems"

As an update on the AP-summing issue - I looked through the sheet and it seems to be largely working for me, with the exception of the fact that the AP value of Malchazeen is entered incorrectly (oops!). Those of you that use daggers might want to go to the weapons page and update it's AP value from 30 to the correct 50. I'll fix it in the next major revision either way.

It's certainly true that daggers don't have *higher* DPS potential than swords - they're at best comperable. Whether or not it's actually true that swords absolutely crush daggers in terms of damage (i.e. Vindicator's + Latro's vs 100 dps daggers) again depends on some of the subtleties of how sword spec works, which no one actually knows.

The other caveat I'd throw in is that the dagger cycles model isn't as highly optomized as the sword cycles model, so it's possible that a bit more damage can be wrung from it. It's probably not going to make 50 dps different, but it might be a 10 or 20 dps upgrade.

Fists (and, with the advent of Rod of the Sun King, Maces) are on the agenda... but, I confess, fairly far down on the list. It'd take some fairly major overhauls to include them, so while I would like to add them I'm going to grab other, easier-to-add features first. Once the core functionality of swords/daggers is working well, I'll look at adding fists, maces, and mutilate; until then, it's not really practical.

I'm not sure I understand what you're saying. Miss chance is floored at zero - you shouldn't be gaining additional benefit from hit once your miss chance is at zero. So the calculation needs to be done the way it is to correctly value hit.

So, in order to get gear to show up in the drop down menu, you also need to add it to the named ranges. There are two ways to do this; either a) manually update the corresponding named range, or b) add it in such a way that it's already in there.

I'm not going to explain the first one; if you already know how to do it, feel free - but as I personally don't remember off the top of my head how to do it, and there's a perfectly good workaround....

For the 2nd one, the easiest possible thing to do is just overwrite an existing item - take some of the Black Temple items and replace them with the name and stats of the item that you want to add.

In terms of actually adding gear, it generally works if you insert the row in the *middle* of the block of statistics, copy an adjacent row onto the newly-created empty row, and then update the name and stats.

An important warning about modifying the armor lists: inserting in this way tends to work, but if you try to reorder the existing list (via sort or whatever) it will frequently break the sheet. The references in column BF frequently don't update correctly which tends to break things very very quickly.

I may be having a sudden lapse of sight, but how are the values for offense and defense calculated?

Great sheet, keep up the good work. A bit disheartening to see the tremendous superiority of swords considering the poor itemization, no offhand sword drop until BT is ridiculous.

I noticed weapon skill rating seemed to be valued quite low, only about 1.3 AP for me. What values are used for weapon skills in the sheet? I assume 0.1% hit since we have some data on that one but what about crit and dodge? Blizz claims skill is supposed to be the best stat and yet refuse to give us the real numbers, very frustrating. Also, have there been any testing to see if weapon skill is indeed rounded down or if it's just a char sheet thing?

Btw shoulderpads of the stranger have 16 crit rating and not hit rating (unfortunately).

I'd be pretty surprised if it isn't being floored, but I haven't seen any formal testing.

Sheet uses .04% crit, .1% hit, and .04% dodge per weapon skill. Should probably up it to .08% dodge based on the findings in the weapon skill thread, but was waiting for better numbers.

Ah, indeed. I'll fix that in the next release. Good catch.

I was lucky enough to gather up both a spiteblade/gladiator quickblade and a malchazeen/shiv

I also have 308 hit rating in both setups (spymistress/maliciousness) with just a bracer swap. I have the haste metagem, and dragonspine trophy. Is there any combatlog recording that I could do outside of a raid that would be remotely useful in comparison of the two setups? Or any information that would be valuable here.

At first glance, the swords just have a hugely optimized skill cycle with 2pc netherblade. I was able to run 1pnt snd's, and easily fit in 5point rupture/5point envenom (although I don't know the value of envenom vs just rupturing more.

I'm very interested in getting definitive comparison of potential between these two specs.

It's certainly possible, although personally I'm somewhat inclined to doubt it, for the following reasons:

1) We showed pretty conclusively in the Rogue DPS Spreadsheet thread that against same-level mobs, +skill gives +.1% hit. Now, the value *could* be different against higher-level opponents, but if it was going to increase, +.14% seems like sort of an odd amount to increase to.

2) The data I posted in the Hit Caps thread seems to me to be a fairly compeling argument in favor of the hit cap being 24.5% + whatever you get from 10 points in weapon skill. So if weapon skill gave +.14%, the hit cap would need to be 25.9%, and on the whole that strikes me as less probable than .1% and 25.5%.

I did make some assorted model changes, which... may or may not drop the value of hit. I don't honestly know. I know the value of hit assuming you're using Thundering Skyfire Diamond is down somewhat, but other than that I don't have a good feeling for how the value of hit would have changed.

Regarding weapon skill:

I was looking over the data that was taken on the Weapon Skill thread ([Rogue/Warrior] Weapon Skill Adjustment Discussion post 64), and it occurred to me that we're making sort of an unwaranted assumption, namely that weapon skill while dual-wielding works the same way that weapon skill while single-wielding. And maybe it does - but given that some other mechanics seem not to carry over between the two cases, it's conceivable that they don't.

So, first things first. Beaglej had no +hit and no +weapon skill, and his miss chance on the boss was roughly 8%. So, first thing we know: while single wielding, one loses more than half a point of hit per mob level - even though evidence is fairly strong that that's what you lose for dual wielding. Grinding some statistics, we find that the base miss rate probably lies between 7.4 and 8.54 percent. So lets assume, for the moment, that the base miss chance is 8% - that is, each extra mob level increases the amount of hit required by 1%. There seems to be a certain logic to this, at least to me - single wielders lose 1% hit per mob level, dual wielders lose half a point on each weapon, for a total of 1% (sort of).

Now, Therasiv also had no hit, but 365 weapon skill. Grinding some statistics on his miss rate, we find that it most likely lies between 4.79% and 5.66%. Assuming that the base 8% number is correct, this means the hit granted per point of weapon skill is between .156 and .214%. Now, given that range, .2% looks like a pretty plausible value to me.

Now, the initial issue with this is that .2% has been fairly thoroughly disproven in the Rogue DPS thread. The data in that thread points fairly strongly to the answer being .1% per skill. However, since mob level reduces hit chance by twice as much for single-wield as it does for dual-wield, it seems plausible to me that +weapon skill gives twice as much hit for single wield as it does double wield (.2% versus .1%).

Now, how this extends to crit and dodge, I have absolutely no idea whatsoever - I'm not aware of any large-scale testing of these with dual-wielding. But it's the best explanation I can think of for the miss-chance data we've seen so far.

As an additional note: in addition to raising miss chance by 3%, the mob seemed to be lowering enemy crit rate by 3% as well (tooltip crit rates of 5.97 and 5.25 became actual crit rates of 3.278 and 2.258) which is significantly larger than the oft-used .2% per level. More data on this effect would probably be worth seeing.

I've been thinking about whether or not finding/using a Wowwebstats log with rogue dps on a single level 73 target the whole time would be good for analyzing this type of thing. We always go off and test on random 70 mobs for weapon skill in the past and it hasnt been helpful.

Problem is that we have to know the rogues gear and spec at the time of the fight. Also he has to be on the boss the whole time, no adds.

Ald, what would be the confidence level for weapon skill effects on a ten minute fight with normal cycles? (assume like 5% margin of error...)

If it's significant in demonstrating effects I will arrange for it on our next morogrim kill with myself and another rogue.

Selecting Spymistress's Wristguards on the v0.5 sheet makes everything change to "#N/A"

That's downloading the MS Office version of the sheet and changing nothing other than selecting Spymistress's from the wrist dropdown.