Warrior DPS Spreadsheet for Arms and Fury

I've been looking on the internet for this spreadsheet and I don't think it's still out there, at least I have not been able to find it, so I uploaded mine. Hopefully everybody can make good use of this. It was designed by an Elitist Jerks member and posted on Elitist Jerks (http://web.archive.org/web/201308111...et_2_3_beyond/).

http://www.megafileupload.com/2Mm2r/...r_DPS_2_EN.xls

Toy around with it a bit and explore the options, you will find that it lists the DPS cycles for your MH, OH, your actual DPS cycle (BT, WW etc) and during Execute.

You're a saint. Rawr, googledocs and other missleading shyte, now we're talking! :D Thank you!

Quote:

---

Originally Posted by Lulleh

You're a saint. Rawr, googledocs and other missleading shyte, now we're talking! :D Thank you!

---

How is this any different? For Fury this spreadsheet is a joke, talk about "misleading shyte". The only decent Fury spreadsheet for TBC made back in those days was the Landsoul one.

Quote:

---

Originally Posted by SegaRally

How is this any different? For Fury this spreadsheet is a joke, talk about "misleading shyte". The only decent Fury spreadsheet for TBC made back in those days was the Landsoul one.

---

I wrote the reply before actually opening it. I do appreciate the upload but the content was below the expectations. =/

Quote:

---

Originally Posted by Lulleh

I wrote the reply before actually opening it. I do appreciate the upload but the content was below the expectations. =/

---

lmao, it's like upvoting a torrent before even downloading the movie

we're back in the 00's

Quote:

---

Originally Posted by SegaRally

How is this any different? For Fury this spreadsheet is a joke, talk about "misleading shyte". The only decent Fury spreadsheet for TBC made back in those days was the Landsoul one.

---

It's the best there is at the moment. If you have anything better, feel free to upload it.

Using a bad gameplay modelling spreadsheet because "it's the best there is" is a pretty bad way of going about things, it's better to just not use one at all. I have a pretty vivid memory of Landsouls sheet and of the good o'l days to have a good grasp on which pieces are best and why, how the stats interact through scaling and rough stat weights at various gear levels.. I'd rather use my memory and the available theorycrafting than a flawed/incomplete mathematical model based on a different version of the game (it's a 2.3 sheet).

The only thing I'm using at the moment is this, https://docs.google.com/spreadsheets...#gid=115069182 .. Which is merely a gear ranking sheet based on stat weights, a SEP list which seems quite accurate. I do have a version of Landsouls sheet from TBC but it's from the WOTLK pre-patch, so level 70 cap with level 70 gear but with WOTLK talents, sadly... Would love to get a hold of a 2.4.3 version but it seems to have disappeared from the internet, I guess I deleted it years ago as it's not on any of my hard drives.

Guess I just got excited and had too high expectations. I do appreciate the gesture of sharing it but sadly it's of little use. The google doc seems like an upgrade from what I previously had (pre raid and T4 BiS docs). As for Landsoul sheets, was there even a 2.4 version? Like was he even a thing before WotLK? Iirc the WotLK is what initially made him famous. I suppose someone with too much freetime and adequate math / excel knowledge could do a backport of the WotLK sheet. I do remember a Vanilla sheet I used back on Feenix though, doubt it's from 2006 but instead made in more recent times. I'll have a look when I get home. Porting that to work with 2.4 might be easier as the difference of Fury Warriors Vanilla > TBC is smaller than TBC > WotLK.

There was a full spreadsheet in TBC yes, it was on open office and it was widely used during the latter part of the expansion. Maybe his sheet become more famous during WOTLK times but he already had a very fleshed out and accurate sheet during TBC that I spent countless hours playing with as a fellow theorycrafter at the time.

I have the pre-wotlk and 3.3.2 versions of the sheet but neither is really usable for TBC, the talents are different. Warrior stat scaling in WOTLK was very very similar between both expansions but there were massive differences between both the gear (armor penetration % vs armor ignore static, huge crit levels above 40% already in T7) and the class (rampage = 5% crit, titans grip, deep wounds stacking, instant cast slam procs)..

The google sheet is a gear ranking list calculating gear SEP value based on stat weights as you would generate from Landsouls sheet. It's obviously an improvement on the pre-raid list because that was made by someone who has no clue about tbc Warriors, where as this one is using stat weights that line up closely with Landsouls sheet and known theorycrafting. It gives a good approximation of the best pieces but of course a full working model gives you accurate stat weights exact to your gear, gives you a good approximation of dps, allows easy optimisation of gemming, trinket choices, enchant choices and how they benefit you based on your own gear, and that's what the Landsoul TBC sheet gave you.

But most TBC theorycraft is widely known information to anyone who played at a high level at the time, who followed the theorycrafting scene or who can take the time to read through some of the archived theorycrafting, but also to people who take the time to do their own now and to go through the maths, to test things and to re-confirm things using the known formulas, the calculations, to test uptimes and proc mechanics.

To that end, we don't really need Landsouls sheet to figure out the best pieces because that information is largely obvious to anyone with a theorycrafting numbers mindset, and experience with the game. His sheet was just a nice toy to play with, a good way to optimise everything to the last stat point.

While discussing the docs, seems like a solid list (compared to the weirdness seen in the others). I appreciate the explanation of the Dragonstrike procs too (math calc expl). I do see that 1 STR = 2 AP where it should be STR*2.2 taking Kings into consideration (generally giving Plate items a small boost, same for Agi and Leather items (AGI*1.1). Then there's expertise. Something tells me expertise is valued less than its true potential could since the rounding down (0.25%) as each individual item is ranked Vs. others. For example, Shapeshifter's Signet gives 20 Expertise rating which would be 5.07 Exp. The 0.07 is wasted, thus a lower value. Calculating taking into account the total 6.25% Expertise (in my case being Human) the calculation and value of the item would be higher, I'd think.

The only odd item in this (as opposed to the other sheets I've seen) is the Barrel-Blade Longrifle. And some other ranged weapons in general.

Speaking about that known theorycrafting, I'll point out some things about stats, maybe it can help some people get a better grasp.

1. Expertise is stronger than STR per point and due to it being a multiplyer becomes stronger the better the gear gets, but It should not be overvalued, it's about 10-12% stronger than STR per point at T4-T5 levels. But you must account for the fact that it only benefits you in chunks of 0.25% at a time (1 expertise), any rating that doesn't make the next 0.25% is wasted on gear, so you need to count this carefully when building sets around it. Example being the Karazhan trash leather gloves, they have 15 EXP rating which = 3.8 expertise, that .8 expertise (or 3 rating approx) is entirely wasted unless you have another piece to make the difference.

So yes while expertise is very strong per point, items can have a lot of waste on them and merely having expertise does not make an item in itself, the first truly strong expertise item is the Belt of One-Hundred Death's from Vashj, people shouldn't go out of their way grab this stat just because it's the best.

2. Hit - After 95 hit (with 3/3 precision) it's a weaker stat, approx 60-66% the value of strength at T4-T5 level and dropping lower as gear increases and more white hits are turned into Heroic Strike/Cleave hits. It effectively continiously but slowly/steadily gets worse with gear, our weakest stat and often overvalued by players.

3. Crit - At low gear levels (pre-raid to T4 ish) it's roughly equal/better per point than STR due to the synergy of Flurry/Rampage uptime and rage generation interaction, but as gear rises it quickly falls into place with 1 crit being roughly equal to 2AP in terms of SEP. Crit never really falls off, continously scaling alongside AP unless you have an excess of crit and a lack of AP, it's very stable with natural gearing in other words.

4. Haste - It's a decent stat that starts out fairly weak at a similar place to Hit (say pre raid level) and by endgame Sunwell is fairly close to Crit in value, and can even surpass Crit in especially Crit heavy gear setups. So we have the stat at approximately 0.7-0.8 of STR at T5 levels but between 0.8-0.9 of STR at Sunwell levels. The only issue with Haste is that pre-Sunwell we don't really have many well itemised pieces with Haste on, save for Dragonstrike and Dragonspine Trophy, even in Sunwell it's an afterthought when compared to Ignore Armor.

The main point about Haste is that while Hit gets worse with more Heroic Strike use, Haste stays constant and scales strongly as other stats rise, especially with Crit. Haste does by default lower flurry uptime and increase how much rage we can burn.

5. Armor Ignore, this starts out as a "decent" and "ok" stat and very quickly ramps up to being outrageously strong when stacked. However the right amount of armor ignore depends on a boss to boss basis and based on buffs (Sunder Armor, Faerie Fire and Curse of Recklessness primarily).. Armor Ignore is probably the most misunderstood stat.

1. It's more effective on low armor bosses, rather than high armor bosses... The closer you get to the boss reaching 0 armor the more it increases your damage per point, the opposite and equal effect of armor itself, it' has increasing returns with each point worth more than the last. But optimising this on a per-boss basis requires you to know the armor values of each boss, and tailoring your gear setup to that boss (so maybe you have gear switches depending on bosses, at the highest level of min-max).

As an approximaton, at T4/T5 level 600 armor ignore is worth approximately 200 ap, which is also why Icon of Unyielding Courage (30 hit + 600 Ignore Armor for 20s on 2min cooldown) is a really lacklustre trinket. Of course at a T6 level in more armor ignore gear the value of that on-use becomes exponentially more valuable, turning a trinket that was once getting outshined by a Hellfire Peninsula quest item (Bladefist's Breadth) to one that is far stronger, albiet it's still not worth it by comparison to other choices at that level.


Fast vs Slow offhands - People often over estimate how much different slow offhands will make, in theory they increase flurry uptime (by the same logic that haste reduces flurry uptime) and also increase offhand whirlwind damage, but that offhand whirlwind damage is only base damage, where as AP normalisation remains static so the bonus is minor. Faster offhands on the other hand provide smoother rage generation allowing often for more consistent and steady rage management which can lead to less wasted rage, and can sometimes give stronger execute phases (by allowing every GCD to be filled more reliabily with Execute).

The point being that while slow offhands are very slightly preferred, you should always go for the best stat breakdown and dps, fast offhands are fine.. Daggers on the other hand, that's a different matter.

Quote:

---

Originally Posted by Lulleh

While discussing the docs, seems like a solid list (compared to the weirdness seen in the others). I appreciate the explanation of the Dragonstrike procs too (math calc expl). I do see that 1 STR = 2 AP where it should be STR*2.2 taking Kings into consideration (generally giving Plate items a small boost, same for Agi and Leather items (AGI*1.1). Then there's expertise. Something tells me expertise is valued less than its true potential could since the rounding down (0.25%) as each individual item is ranked Vs. others. For example, Shapeshifter's Signet gives 20 Expertise rating which would be 5.07 Exp. The 0.07 is wasted, thus a lower value. Calculating taking into account the total 6.25% Expertise (in my case being Human) the calculation and value of the item would be higher, I'd think.

The only odd item in this (as opposed to the other sheets I've seen) is the Barrel-Blade Longrifle. And some other ranged weapons in general.

---

Actually the combat ratings/notes tab is just for reference/information. If you look at the sheet stat weights you will see that STR is calculated at 1 while AP is calculated at 0.4545 which is correct. It's already accounting for Kings. Agility is also calculated with kings in mind in the actual sheet calculations.

The wasted Expertise rating is not calculated out on items because depending on your specific builds the waste can sometimes be reversed, however there are notes in the information on each expertise item pointing to the combat ratings/notes tab where the waste is explained and fair warning is given. Racial expertise isn't taken into account on the sheet as it's just a generic, it could be put into the sheet though. The axes do have notes to account for orc-racial though.

Edit : Stat weights at each gear level are written in the top very right hand corner of the sheet, out of the way.

Quote:

---

Originally Posted by SegaRally

Using a bad gameplay modelling spreadsheet because "it's the best there is" is a pretty bad way of going about things, it's better to just not use one at all. I have a pretty vivid memory of Landsouls sheet and of the good o'l days to have a good grasp on which pieces are best and why, how the stats interact through scaling and rough stat weights at various gear levels.. I'd rather use my memory and the available theorycrafting than a flawed/incomplete mathematical model based on a different version of the game (it's a 2.3 sheet).

---

What is so bad about it? If you unprotect the sheets you can take a look at the maths behind the calculations and they're accurate afaik. One problem I have found indeed is that it does not count offhand crits for flurry (2.4 change) but that is easily added if you'd want to use it for Fury. Other than that the maths are solid, personally I've been using it for Arms. The biggest downside of this sheet is that you can't just insert pieces of gear but you have to calculate the stat totals and insert those instead, so you won't get a nice list with stat weights but instead you have to insert every piece of gear manually while testing upgrades. It's a lot more work on your part and it makes testing a lot harder but that doesn't mean the results are wrong.

Quote:

---

Originally Posted by poepinjehoofd

What is so bad about it? If you unprotect the sheets you can take a look at the maths behind the calculations and they're accurate afaik. One problem I have found indeed is that it does not count offhand crits for flurry (2.4 change) but that is easily added if you'd want to use it for Fury. Other than that the maths are solid, personally I've been using it for Arms. The biggest downside of this sheet is that you can't just insert pieces of gear but you have to calculate the stat totals and insert those instead, so you won't get a nice list with stat weights but instead you have to insert every piece of gear manually while testing upgrades. It's a lot more work on your part and it makes testing a lot harder but that doesn't mean the results are wrong.

---

Just straight out of the box issues like it missing game mechanics and proc mechanics, the fact that it overvalues Crit/Haste as a result, the fact that it's showing a Poisen Vial trinket being worth effectively no more than an empty trinket slot at 1500 dps, showing Tsunami Talisman as only a 3-4 dps gain over Hourglass, showing Madness as a dps loss vs Hourglass. The fact that it values the Rod of the Sun King/Syphon procs at 1 dps gain up at 1500 dps total, 0 dps on the offhand. Hell the whole trinket and proc situation going on is nonsensical. It's just like Landsouls sheet in that calculations have to be run multiple times to arrive at a clear result, but unlike Landsouls sheet those results do not add up with expectations and reality.

The fact is that even Landsouls sheet was primitive compared to what we have today with Simulationcraft (and worked largely based on averages, rather than the complex and more accurate proc calculations people use these days), but this sheet here is primitive compared to Landsouls sheet, largely because it's unfinished and is missing many nuanced calculations/aspects (that are far beyond my capability to fix) of the gameplay modelling that is resulting in it having skewed results. Look I'm not an expert on spreadsheets, whether it be Open Office, Excell, Google Sheets or otherwise, I've barely touched them since I was in school (and that was a long, long time ago) and so I cannot fix what is wrong with this sheet.

The google docs sheet I use is very different, it's very simple calculations wise and simply uses stat weight snapshots at specific gear levels closely matching Landsouls sheet SEP results to calculate gear strength in a simple fashion, it's something you'd normally use to accompany a quality gameplay model, to give a good approximation of the gear hierarchy on a slot to slot basis. But along with that all procs have been tested where possible and data has been collected to ensure accuracy, so while it's a work in progress the list is actually very accurate to the results Landsouls TBC sheet was suggesting.

Fixing (or at least greatly improving) this sheet is simply not within my capabilities, I'm a napkin math/ingame testing theorycrafter and not a post-graduate level mathematician like Landsoul.

Use the dev of the gdocs list this forum?
I have some suggestions to add and some questions (BT trash hammer best in offhand slot after glaive and s3 for t6 post za?)
Anyway, the lists are almost the same what the most players declare as best in slot for fury.