And with "Yes" he means "average of the high/low values". As seen in the weapon's tooltip.

Maybe I've misunderstood something, but there are a couple of things I'm unsure of regarding this spreadsheet. First, how come the projected dps of sub is about 3k higher than combat? This basically leads me to the next question; how do I actually use this spreadsheet? I'm kind of new to this whole theorycrafting area but I'd love to learn more.
I've filled in the green boxes, but.. what now? I read something about adding 40 agility etc, but in which boxes? Sorry for a total newbie question, but I'd really like to learn more.

The first question is one I've been trying to answer for several days actually. It's not implausible that the specs aren't particularly balanced at the moment but the spreadsheet projection has been pretty extreme. Some live server testing last night led me to find a couple of bugs that were inflating sub's overall damage without affecting the value of the stats much and one bug that was inflating the value of agility in particular. (New version of the spreadsheet momentarily). Stat priorities for sub aren't going to change, but matching socket bonuses is going to seem like a more worthwhile endeavor and the projected difference between the two specs is more reasonable. Also, as I said in the original post, the combat sheet doesn't implement Bandit's Guile at all, which is more or less a 10% buff to combat's damage on a single target fight.

Note that the bug with HAT's combo points has made the spec virtually unplayable in ICC fights that aren't festergut and BQL, at least for the time being. Seems like the kind of thing that's server side (and thus could be hotfixed without a client patch) but at the same time maybe not blizzard's highest priority bug at the moment.

As for using the spreadsheet, I fear getting the modelling into a first order approximation of right has been more pressing than user friendliness. After you've entered your gear/spec, etc into the green boxes you can do one of a couple of things. First, you can change some of the talent points or glyphs around to see what yields the highest predicted DPS. Second, you can get a personalized value for each of the stats. Start by copying the listed DPS number in cell H26 (the value not the forumula) into the green cell F28. That's your baseline number. Now go to the agility number in the upper left and change it to be 50 lower, and then copy the new DPS number into F29. Then put agility back to what it was, and take 40 away from crit, copy that into F30, and so on. When you're done with all of those (remembering to add 40 to mastery rather than subtract), the sheet will tell you in column H what the value of the stats is relative to agility.

Final Note: Because of some changes I made to the sheet when fixing a bug, users of v0.11 or later wishing to get appropriate stat lines for subtlety need to do -50 agi and -40 other stats instead of -40 across the board. (40 * 1.25 = 50). It's not pretty, but it'll have to do for the moment. I changed the description so that if you're using the subtlety calcs it reminds you.

So I'm really trying to get into spreadsheets and theorycrafting on my own. I took Calc 3 and Differential Equations in college but my Statistics knowledge is quite sparse. Any tips and sites I could read to a get a grasp on interpreting data? Anywho I'm not sure if this will even help as my first test wasn't very long, but my second test was roughly 4000 melee hits and 1500 sinister strikes. I was trying to see if I could get a somewhat static percentage of time that the Combat insight buff was up. The first 2 stages of the buff seemed to have a roughly 20% uptime while the Deep Insight buff's uptime was ~30% in the first test and ~37% on the second test.

World of Logs - Real Time Raid Analysis
Deep Insight: 29.6%
Shallow Insight: 17.6%
Moderate Insight: 17.6%

World of Logs - Real Time Raid Analysis
Deep Insight: 37.8%
Moderate Insight: 21.3%
Shallow Insight: 21.1%

I took off my trinkets and ICC ring so that any buffs I gained were only the Insight buffs. This was only me Sinister Striking repeatedly for a long time. If you guys want me to try again with a longer test then I'd be happy to. Just looking for a direction to start in and hoping to help.

EDIT: Idk if this is a factor but the first test was using 2 x Hand Axe: Hand Axe - Item - World of Warcraft
The durability on these 2 weapons withered down so fast that I had to stop the test short due to the OH becoming red. I stopped the test as soon as it became red.

The second test was using MH Razorscale Talon: Razorscale Talon - Item - World of Warcraft and OH Kinetic Ripper: Kinetic Ripper - Item - World of Warcraft which have much higher durabilities and no enchants.

The trinket lists at the bottom are merely an approximation of the trinket's value using the EP numbers that the sheet computes. So, for instance, HDV is essentially 288 * AP_VALUE + 510/3 * AGIL_VALUE. The sheet does no further modeling of these trinkets more than this, but they're the best approximation for their value I have.

The trinket toggles above in green are really just so you can add the procs in without having to go look up their uptimes. (This is why HWT feeds in 340) The reason to do the procs seperately is that any stats/ratings on the trinket are already reflected in your character sheet, and the values at the top are really meant to just be read straight off an unbuffed character sheet for simplicity. Adding the rest of the trinkets in is on the to do list, but honestly pretty far down, especially as the backend of the sheet is undergoing some structural changes this week.

Edit: My bad didn't update my page since last night and didn't see your last post

Hi, I've been playing around with your sheet alot. Love the design love how simple it is. It's just been missing one thing for my personnal sheet and since I'm probably not the only top geared rogue wearing that, I was curious as if you were gonna implement the H WFS and regular version. I've been going through your Cells E7 through E20 and I was wondering if you just didn't do the maths for them or if you were somehow supposed to use the trinket valuations you added in the cells E37-46. Thanks a lot, I appreciate that someone took the time to make a spreadsheet for 2 months of raiding some of us are still working on Heroic LK 25 and recently I felt like I was the only one who cared about that in the PTR. Thank you!

Thank you for the clarification.

I found that I wasn't able to cap expertise at 200 (0% Doge/Parry on a level 83 Boss). I had to go into the 400s. Did I mistake something? I'm still spell hit capped (converted mostly crit from items into expertise), adjusting all gems to what was called for in the spreadsheet. Dungeons really aren't going to be a good way to test out straight DPS, so I'll have to pug my way into a raid to see what it's really capable of.

Man, you guys are killing me with the fundamentals! That's what I get for being out of the game for so long. Will have to undo some of those expertise reforges and put it into haste. Thanks for all the facepalms, the criticism is appreciated.

I see some people are say HC BB is only slightly ahead of HC Bloodvenom, but with my gear set in your spread sheet it seams to add 500dps..... Is this accurate?

Thanks and Good Work

Are you correctly adjusting the stats? (lack of agi/etc on BB)

Adding the AP as "AP from enchants" rather than as agi is a little bit better (you get 5% extra agi from the new leather armor mastery plus 5% from kings), but that sounds about right.