These are questions that the spreadsheet can answer better than we can.

Regarding the 50% crit buff, the only way to accurately answer that would be with a WWS log so we could properly compare the damage from an Eviscerate crit versus a Rupture. My anecdotal experience, though, would likely suggest to keep using Rupture.

As for white crit capped, you most likely were, so hit gear would be actually the same as crit except the conversion would be 32.77 for 1% crit (actually a hit, but works like a crit if crit capped) instead of 45.9 (would only apply to white attacks, though).

My napkin math says your Rupture should do 10000~ damage (assuming from my average Rupture tick and the fact that our gear is similar and you had Blood Spatter), which is probably more than your average Eviscerate (edit your post with a WWS log and I'll edit mine with the correct analysis).

Sorry but I didn't save log. Our guild used WWS in the past. But I don't know if anyone from yesterday have done it.

When crit capped for white:
- crit rating converts hit to crit for yellow/poison, but does nothing for white
- hit rating converts miss to hit+crit for yellow/poison (if below cap), and converts miss to crit for white
If I assume 30% damage is from white:
EP for crit rating should be lower (1.5 x 0.7 + 0.0 x 0.3 = 1.0?)
EP for hit rating should be higher (1.6 x 0.7 + 1.6 x 0.3 x 2.4 = 2.1?)
No?

This is a somewhat long-ish question, but it may end up with simple answers.

I was comparing a WWS of my Patchwerk (10-man) performance last night to a 0.4.3 spreadsheet model of my gear and buffs I believe I should have experienced. From looking at the WWS stats, I was able to eliminate Faerie Fire since it was rarely applied to the boss. I need to look more closely, but the other buffs/debuffs appear to have been present.

The spreadsheet shows my maximum potential as Combat (15/51/5) to be about 2740 dps in this fight, while my actual dps for the kill was 2382. I have identified some mistakes and possible mistakes (staying too close to front of boss after KSpree and getting parries, delays between Rupture falling and being reapplied, and Deadly Poison dropping once), which should net me about 99 dps increase, putting my adjusted dps at 2481. This is still 259 dps below the spreadsheet.

Does the spreadsheet take into account target spell/poison resistances? He did resist some of the damage and would have accounted for another 15 dps.


Accounting for the adjustments, which assumes I can make Deadly Poison never drop, Rupture immediately when I need, and that there is 0% resistance, I'm seeing the following dps variances:

White Damage: 890 actual, 1006 expected (116 difference, 11.5%)
KSpree Damage: 115 actual, 139 expected (24 difference, 17%)
Sinister Strike Damage: 727 adjusted, 754 expected (27 difference, 3.6%)
Rupture Damage: 282 adjusted, 328 expected (46 difference, 14%)
Poison Damage: 425 adjusted, 458 expected (33 difference, 7.2%)
Garrote Damage: 15.6
Eviscerate Damage: 10.3

I was able to open with Garrote shortly after the tanks started smacking him, and I had enough energy and combo points in my rotation (3s5r) to Eviscerate once. When cooldowns were available, I used KSpree by itself, AR + BF once, and then BF by itself. SnD didn't drop.


Is the number of total hits too small to be concerned about these differences from the spreadsheet, as I may have experienced a dry period? What would be an appropriate sample size to have in order to say that the differences are not due to the RNG?

How much of an effect can latency have on actual dps vs spreadsheet dps? I am on the East Coast USA and play on a West Coast USA server, typical latency of 150-200 ms.

What else should I look for, in order to find either my mistakes or other unaccounted-for reasons for such a large variance from the expected values?

(Edit: I forgot Sinister Strike, though that seems to be fairly close in this fight)

I can't remember which thread it was but someone calculated the dps loss as 2% from the 15 energy from tricks, not counting focused attacks. as combat's energy regen is higher I would imagine that the loss is a little bit smaller even, however if you're boosting someone's dps by 3% that is at the bottom of the dps and you're at the top of the dps, it's likely not worth doing it on that person, it seems like it's very highly relative on your dps and your raid's dps. personally I've kinda given up trying to use it every cooldown and go with using it at critical points, low threat on a mob that needs to die fast (sarth+3 for example), during heroism, or if I know "my" rogue just popped their cooldowns.

You're very unlikely to lose DPS by using Tricks on every cooldown. I mean, the 15 energy every 30s is shown as a 1.7% DPS loss for me, in my gear. Using Tricks every cooldown is, averaged, 20% uptime, or 3% increase in overall damage. So, I lose around 60 DPS to use it, so my target would need to do more than 2000 DPS for Tricks to be worth using, this is assuming I do 4000 DPS. Overall, it should always be worth using it on someone doing DPS at least equal or close to yours.

perfect, thanks for the replies, that's just what i needed to know.

Well, the point is that Master Poisoner, as a raid buff, contributes a large amount of group DPS... assuming no one else is doing it. So if you almost always have a ret pally or ele shaman around to apply the +crit debuff, you'll do better with TtT; if you don't, Master Poisoner is clearly superior. So, look at your group comp - if you almost always have another source of the crit debuff (or at least, almost always on DPS-limited encounters), TtT will be the better choice; but if you find yourself without it with significant regularity, you should get Master Poisoner instead.

This advice is really best expanded to ANY paladin.

Heart of the Crusader is a standard talent for Protection paladins who want to get up the tree to Pursuit of Justice.

It's an even more standard talent for Holy Paladins who want to get up the tree for all that juicy crit.

Paladins of any spec will be judging every 30 seconds.

So long as the paladin is on the primary DPS target (which is the main advantage of a RET paladin specifically), of any spec, TTT is better.

Ok one what glyphs are you running, if you're running snd glyph then a 3s/5r/5e rotation with imp snd talented is completely doable but it may not be wise as you may end up drooping off snd before you get back to three combo points for snd. I on the other hand run Ar glyph, Rupture glyph, and Sinister Strike, i find it quite easy on moving fights to keep up a 4s/5r/4-5e rotation without any send downtime. The real pickle you get into is when your snd glyph procs going into your 3rd combo point. Obviously this means you're just running a 5snd for the next rotation but it can still throw you off a bit.

In heroics, ruptures might a waste, if stuff dies too fast.

As for gear, no meta gem helm, no enchant on shoulders, no enchant on chest, no reason for blue gem on chest w/o meta, no enchant on bracers, no gem on belt, no enchant on boots, no enchant on sword, no enchant on dagger, slightly overcapped on expertise (1 too many), talent wise 5/5 CQC and 4/5 Sword Spec is slightly better.

Using wound poison over instant poison?