In my experience there's a LOT of really weird crap going on in both Rawr and that spreadsheet. I've got my own spreadsheet to figure things out, and have kept it really simple. It just takes imputing raw stats and it outputs dps, no complicated equations to generate graphs or select gear/buffs. It's more work, but I'm a lot more comfortable with the answers. Also, building my own spreadsheet taught me a LOT about the mechanics, since I had to know every single equation involved in the process.
The short version: If you want it done right, do it yourself.

I don't know if I already said this before, if so I apologize, but after thinking a bit about it, the "obscene" Jow procs rate we're seeing might have nothing to do with Vindication.
Remember that bug on Libram of Avengement? Whites hits procced it, making its user happy, with a permanent 53 crit (or was it CS? I just got one, so I couldn't test it before 2.4).
Considering how Seals works with spell reflection, I think we can "blame" those extra procs on the game mechanics. When you hit a spell reflecting mob who has a judgement on, your white hits will cause you to be judged. So, each white hit would "cast" it's own Judgement spell (as long as your target is under a Judgement spell). And since spells have a chance of proccing JoW, that would explain a lot.

Of course, this is pure theory (or call it fantasy if you want) I don't have much to back that up, but if you consider those earliers wws reports at 80% JoW procs without Vindication, that's pretty much the only answer I can come with.
Juste food for the tought still, I have absolutely no clue about how I could back that up.


edit for weird typos.

edit again, cause I didn't said it earlier: That's one badass looking avatar you got Avitus. Worth every SoB in the world xP

Ignoring graphs (which shouldn't have anything to do with results of the calculations), isn't the selection of "gear/buffs" and debuffs what make Rawr/Bellator's Spreedsheet accurate?

Does your spreadsheet consider how all the party/raid buffs and debuffs boost your damage?

I do agree with your principle however, one shouldn't completely rely on spreadsheets, manual calculations to confirm are always useful (I make pretty extensive use of pen and paper myself, or rather notepad and keyboard).

In other news, why not post your spreadsheet here? Maybe we can figure out where the discrepancies lie for the benefit of all three spreadsheets.


Yea this is exactly my conclusion. Salty posted 87% proc rate without Vindication. I posted 83% and 89% with Vindication. Looks pretty equal to me, regardless of Vindication.

Given that info, Vindication doesn't become completely useless (it still should proc on trash), but I guess it looks very likely that it's at least useless for boss fights. I might throw that 1 point back into Benediction myself, still not sure, it's a negligible difference either way.

Ultimately this needs testing. If someone is up for running a log autoattacking mobs for a few hours with JoW on them and see what it says, it should shed some light on things.

If it averages out at 50%, then it's something else we're missing. Possibly try 2 tests, one with 2 paladins where another paladin puts up JoW.

Additionally I intend to do a try on Brut where I'm the one judging JoW and see if things are different.

Hmm thanks Wish Blizzard would think the same.

The paladin dps modeling is relatively simple.Since I started from scratch when developing the Rawr model, I'll be surprised if there's anything which both Bellator and I happened to miss, unless it has to do with new findings about game mechanics(the SoC-Windfury stuff in the last page).

So my buddy ran WWS last night for our Hyjal clear and I found some pretty decent results for myself. I'd like any input on places for improvement if possible.

Here's the link for the whole raid just be aware I stepped out after Kaz'rogal (was supposed to get back in for archimonde but raid leader forgot).

Full Report

My dps got screwed on Kaz due to the mana drain crap but I managed to get in and do some damage at some points. So I'm focusing on the rage and anetheron kills to check how I really did.

Rage Winterchill Kill
-ignore the 1 parry, the tank screwed up and turned him towards the melee

Anetheron Kill
-2 parries were probably when he turned around to cast inferno

Considering this is the first time I'm posting any WWS for my raiding I'd appreciate any and all feedback you guys could give me.

The biggest clue to me that something is off is from having really strange gear suggestions like S1 shoulders being better than S2 shoulders. There wasn't a whole lot of it, but I really have to wonder when I see it.

My spreadsheet just takes raw stats, so if you want to include buffs, you've got to calculate them into your numbers before inputing them.

It was posted multiple times much earlier in the thread, but it didn't get a lot of interest largely because it isn't as easy to use as the other options. I'll definitely make sure it's up to date and post it again though if people are interested.

if that were true, wouldn't we be able to show this simply by judging wisdom and white swinging until we get a double proc?

I know it would be obscenely complicated, but is there any way you could set an option to ignore hit and/or expertise (seperately would be preferred) when comparing items? I say this because I often have to juggle gear around to satiate hit caps.

No biggie though, just curious

I do get weird results occasionally, but never fail it has to do with hit rating or my meta -

if it's the former, I just eyeball it - just from playing with it, sometimes it values the full amount of +hit on an item even though I only need a small bit of it. (i.e. I need 2 HR to cap, the item as 30, and it greatly over values the other 28 hit)

If there isn't hit on either item, or if that doesn't line up, then I'll go check the meta requirement and uncheck it, and it seems to fix itself.

Note the last issue never happened in the beta version.

Neither of these things are really a big deal, just use some discretion - the tool shouldn't be making the choices for you, anyway.

Personally I trust the spreadsheet more than Rawr. I know Rawr is new and cool to look at but here are my gripes:
-It tends to screw up meta gem requirements
-Unequipping and reequipping items sometimes gives a different item ranking
-Optimizing always gives me gear with no gems
-DPS out seems to be a lot higher than what I can achieve in a raid
-No Haste Potions in consumables

Im not calling Rawr bad, because its a good tool for getting relative value of different gear sets. I dont think the spreadsheet is 100% accurate either, but Rawr still needs some tuning.

I think Rawr has some bugs. I tried to active the leader of the pack buff and saw no rise of my dps...

The estimated DPS output shouldn't vary that much from the spreadsheet. Of course, practically your CS may be 6.5s apart on average instead of 6.0, your judgement frequency may be lower than 9s etc,

As far as rawr is concerned you get as much out as you put into it. I find equipping an item in a slot I'm not looking at that will push my hit well past cap helps me get more honest ratings of other pieces. You have to remember rawr's main concern is upgrades to your currently selected gear set. It can't assume that you you will properly balance you +hit when you get that new item without hit on it, so you have to help it by making it believe you have no need for more hit. The same is true for meta gem requirements. often a particular item will rank lower then it really should because losing your meta gem is a significant loss of dps, just keep it in mind and work around it.

As a side note it would be cool if I could turn on and off hit rating and weapon expertise. I would use this to judge items as potential upgrades that would go along with other best in slot hit pieces later.

As far as JoW is concerned I might go out and do some initial tests. I'm curious to see if judging jow yourself and having your white hits update it will proc. more. If it doesn't then I think 2 trials where I have another pally judge wisdom while I hit without a seal and with a seal judged should shed some light on the topic.

Doesn't show that for me. Are you sure you had everything set up correctly? Are they showing equal gems?

I've been pretty much hounding Anarkii and Bellator before him for any and every discrepancy I could find and I think they're pretty accurate now.


I do tend to use the spreadsheet on ocassion since it does have a few more buffs/debuffs than Rawr currently has, however they should be coming soon according to Anarkii, at which point it will be a full replacement for the spreadsheet.

The main advantage of Rawr remains being able to very rapidly plug things in and have them displayed, which is somewhat cumbersome in a spreadsheet.

It lists me as ~1930 DPS on Brut. I've done ~2k, but we did have a full drums rotation which isn't implemented yet in Rawr Ret. Seems fairly accurate to me.

At least, until blizzard messes up mechanics again and we have to start from scratch...

Why have a ret pally in the raid if they're not in the melee group?

Apparently I was too slow to post

The dps benefit a ret pally gains from the melee group is far more than that of a rogue. Sure a rogue will probably do more total dps... but lets look at an example (feel free to have your raid leader look this over). I'm going to use fairly arbitrary numbers for now, but for Avitus's main post in a few days I hope to write up a detailed benefit analaysis with a lot of math (which shows the same thing, just more accurately and less chance of someone arguing). The numbers are preliminary from what I've gotten so far with the post I'm writing, so they aren't far off what the final post will have.

Outside of melee group, with average raid buffs on brutalus, you can expect something like this using haste pots and cooldowns:

Ret: 1600
Rogue: 2000

The best possible personal compensation for NOT having windfury:

Ret: Righteous weapon coating or sharp/weight stone - <30 dps
Rogue: Deadly Poison - ~70 dps

"Best" possible out of melee group DPS:

Ret: 1630
Rogue: 2070

--------------------------------------------------------
What they get from a melee group:

Strength of earth totem (improved):

Ret: 200 ap
Rogue: 100 ap

Unleashed rage:

Ret: 320 AP
Rogue: 270 AP

Note here: Ret paladins typically have significantly higher AP than rogues

Windfury:

Ret: All melee attacks affected
Rogue: Main hand only... about 40% of all melee attacks
-------------------------------------------------------------------------------
I forget the other math I used, but it also used Battle shout and LotP, and showed that a ret paladin gains about 400 personal dps from the baseline to end result after group buffs, and a rogue gets soemthing on the order of 310 or 320.

Given this knowledge, you have two options:

Rogue gets melee group, pally doesn't...

Ret: 1630
Rogue: 2320
Total: 3950

Pally gets melee group:

Ret: 2000
Rogue: 2070
Total: 4070

Net raid result: 120 dps, and thats not including faster vegeance stacks, or mana gained from the enh shaman's mana spring totem.

That being said, depending on how stubborn your raid leader is, 120 dps may not be enough to sway him

Again, I will have LOTs of hard math from rogues and other dpsers in my guild, all who read these forums fairly religiously and know their classes very well.

Well I did a test with JoW judged for the entire 20 min fight. Here are the results:

Wow Web Stats

Hits landed: 642 norm + 61 crit = 703
JoW ticks: 354

354/703 = 0.5036 or approximately the 50% - as expected from previous trials.

This could indicate that a pally refreshing their own JoW will only proc off their hits/spells. (refreshing JoW can not trigger JoW) In order to determine if refreshing other seals on a JoW'd target will proc JoW I'll need to try and recruit another pally later today. It would also be interesting to see if CS updating several judgments can proc JoW from each refresh. At the moment I am holy , but I have a friend with an alt pally who might lend me some of their time.

Nice post Zurm, I think some values would need a bit adjusting, but that's something that the OP will definitely need, since it's a pretty frequent argument.

Also I believe you're slightly overestimating the rogue DPS outside melee group. 2k on brut outside melee group would be very impressive (assuming no glaive combo).


In addition to that argument, a usual "outside melee group buffer" are feral druids and BM hunters. They give nice incremental increases to rogues, where as for ret it's all about windfury that either makes or breaks it. As well as the higher gain from SoE/UR as you mentioned.

If you can manage this, it would be very valuable. It would be enough to conclusively rule out or confirm there being a difference in jow procs if the judgement is put up by you or cast by someone else.

This test backs up the experiences I have had doing tests on blasted lands Servants of Alistarj (sp) with another paladin. For each ~8 minute test we would alternate judging Wisdom and Crusader. When I was designated JoW bitch, my proc rate was very close to the expected rate: ~50%. When I was designated Crusader judger, my JoW proc rate skyrocketed. The other paladin experienced exactly the same phenomenon. Since then I've been sure that Wisdom is always judged by another paladin if at all possible. Unfortunately this was some time ago and we didn't bother to WWS the tests, so it's just personal experience that backs up some of the numbers earlier posted.

Edit: At that time, both of us had at least 1/3 Vindication in our builds, so we attributed it to that. Now I'm not so sure.

It is worth pointing out that until/unless Rwar is ported to Macs it will not replace the spreadsheet for a fair number of people (myself included), so if we do start a new thread it still should be important to list the spreadsheet.

As for JoW, my guess is that it is simply the old Judgement refresh bug behaving slightly differently.

Lets take a random WWS of mine with a large JoW proc chance:
Naj'entus Kill

192 attacks
164 JoW procs
85.5% proc chance

Obviously a much higher proc chance than we should assume. Ideally we should be seeing 96 procs for a perfect 50% proc rate.

Now the "old" refresh bug basically made it so that whenever you refreshed JoW with an autoswing you would have a chance to proc JoW from the white swing and a 50% chance to proc JoW from actually refreshing JoW. As Morindor shows, this isn't the case anymore and it was fixed.

However, assume for a moment that refreshing a different judgement with an autoswing could still proc JoW. That would mean each of my autoswings refreshing JotC could have possibly triggered two JoW procs. So, when calculating proc chances, we would double the number of autoswings. Our numbers come out like so with this assumption:

279 attacks
164 JoW procs
58.7% proc chance

Now we're getting back into the realm of statistical probability.

We could even take it further, and assume that CS refreshing Judgements other than JoW can also cause double procs. Our holydins were being lazy so there was no JoL in the WWS above, but this would be worth another 34 chances for a JoW proc from refreshing JotC. Now we're looking at this:

313 attacks
164 JoW procs
52.4% proc chance

This is all guesswork, and it needs to be tested. How? Get two pallys and head to Blasted Lands. Have the other pally judge Wisdom and you judge Crusader. Now just autoattack. If the above is true you should see some double procs from your autosing refreshing JotC.

If not, I'm fresh out of ideas. I'll try it when later if no one else has time to do so.

Incidentally, if this is the case it shows some very lazy coders at Blizzard. Instead of flagging all judgement refreshes as unable to proc JoW they only altered JoW itself.

EDIT: No I don't/didn't have Vindication either.

Avitus --

Like I said, it was a very early and preliminary post, with lots of guesses. I have a lot of work to do, and I figure when I'm done ill just post it here or PM it to you, and you can include it with thread 2.0. If there are any other variables you think are necessary for me to include, please do (and that goes to anyone, I read my PMs often and don't ignore anyone). The goal is to essentially have a mathematical proof for a ret pally's place in a melee group, as a replacement to a rogue or feral druid. I expect it to be fairly lengthy, so prepare for a wall of math.

BTW Fiola, I may need you to dust off that statistics book again, as I sold mine back to the school store after I took Statistics and Data Mining