Soil Gorget & WSFollow

I use Dancing Edge for almost everything (except for Absolute Virtue, where Pyrrhic Kleos does better for some reason), but do occasionally use Pyrrhic Kleos -> Evisceration for self Darkness.

Soil gorget will definitely do better than kubira beads.

I always default to Pyrrhic Kleos unless I know that everyone should have capped accuracy against the mob I'm fighting. In Abyssea parties, you're gonna have a lot of people with poor hit rates, so Pyrrhic Kleos will add a lot more damage overall than Dancing Edge ever could. Also, on more evasive targets, not only is it a good idea to use PK due to the evasion down, but it has one less hit than Dancing Edge, meaning you're more likely to connect with all hits. On things like frogs in La Theine, I can't imagine using any other WS - those are evasive suckers, and the melees need every bit of help they can to maintain a respectable hit rate. Also against higher end NMs, the melees won't be capped in accuracy, and normally neither will a DNC, so it's a good idea to use Pyrrhic Kleos.

Dancing Edge isn't that much stronger than Pyrrhic Kleos, and certainly not by a large enough margin to favor it over a WS that will buff the accuracy of everyone fighting that mob. If you're in an alliance situation, you're adding 10 Accuracy to every single person fighting that mob - that's quite significant.

I'm always surprised to see that most Dancers simply don't use Pyrrhic Kleos, when our job is to buff first, with damage as a secondary function. (If we're in a healing role, then we normally won't be WSing, so that's another story) So why choose a slightly more potent WS over one that will buff everyone fighting the mob? Pyrrhic Kleos will not peak as high as Dancing Edge in pure damage, but against anything you'd use it on (something where people aren't capped in accuracy), I've always gotten more consistent numbers with it. I like to stack STR where I can once my accuracy is otherwise capped, using items like AF Body +1 and Relic Head. Back when I used my Dancing Edge set, I always got lower returns on PK than DE, but I spent a lot of time tweaking it and trying out different gear and I'm now very happy with my PK damage.

And yes, I'd recommend using the gorget for all three of those WS.

That came out wrong, I deserve to be lol'd at for not explaining that statement.

It has one less hit, but is able to put up nearly similar damage due to the higher fTP modifier, and the STR modifier, though small, allows for increased gains by stacking STR when accuracy is capped. If a 4-hit WS and a 5-hit WS had similar damage overall, I would choose the 4-hit one, because you have less chances to whiff. Since DNC has Dual Wield, let the numbers be 5 and 6 respectively.

Chances of connecting all hits with 95% accuracy:
Dancing Edge: 0.95^6 = 0.735
Pyrrhic Kleos: 0.95^5 = 0.774

Making Pyrhhic Kleos ~4% more likely to connect fully than Dancing Edge with capped accuracy. The case is even better with lowered accuracy. Assuming the first hit of both WS have capped accuracy,

Chances of connecting all hits with 85% accuracy:
Dancing Edge: 0.95 * 0.85^5 = 0.421
Pyrrhic Kleos: 0.95 * 0.85^4 = 0.496

Making Pyrrhic Kleos 7.5% more likely to connect fully than Dancing Edge with 85% accuracy. And the lower your accuracy, the more the figures favor a lower-hit WS.

If 2 WS have similar damage outputs, then you'd be inclined to choose the one with less hits since there are less chances to whiff. If 2 WS have a different number of hits, but the properties (fTP, mods, or chance of crit) are exactly the same, then of course you'll pick the one with more hits, because why not take another hit if you can for no loss?

This was my point, though it was originally articulated poorly.

Edit: I also see that my original post has been rated down quite a bit - is this due to my foot-in-mouth statement, which I've tried to rectify in this post, or due to a disagreement that Pyrrhic Kleos can add more total damage than Dancing Edge in uncapped accuracy situations due to its additional effect, which was the message of that post?

Edited, Sep 7th 2010 7:48pm by apresmoi

Your post got me to thinking, so I did some math to check:
Let's assume we have a 5 and 6-hit WS of the same damage total, both at capped accuracy. Let the damage total be called 30x, since 30 is divisible by both, it'll make the calculations easier. I'll simplify the equation a bit by having each hit of a WS do the same damage (in other words, not accounting for fTP), let it be 6x in the 5-hit WS, and 5x in the 6-hit WS. I'll compute the expected value of each.

Odds on 6-hit WS:
0 hit: 1.58 x 10^-8
1 hit: (2.97 x 10^-7) x 6C1 = 1.78 x 10^-6
2 hit: (5.64 x 10^-6) x 6C2 = 8.46 x 10^-5
3 hit: (1.07 x 10^-4) x 6C3 = .0021
4 hit: (.0020) x 6C4 = .031
5 hit: (.0386) x 6C5 = .232
6 hit: .735

Expected Value of 6-hit WS:
5x(1.78 x 10^-6) + 10x(8.46 x 10^-5) + 15x(.0021) + 20x(.031) + 25x(.232) + 30x(.735) = 28.502x out of a maximum 30x

Odds on 5-hit WS:
0 hit: 3.12 x 10^-7
1 hit: (5.94 x 10^-6) x 5C1 = 2.97 x 10^-5
2 hit: (1.13 x 10^-4) x 5C2 = 1.13 x 10^-3
3 hit: (.0021) x 5C3 = .021
4 hit: (.0407) x 5C4 = .204
5 hit: .774

Expected Value of 5-hit WS:
6x(2.97 x 10^-5) + 12x(1.13 x 10^-3) + 18x(.021) + 24x(.204) + 30x(.774) = 28.508x out of a maximum 30x

I'm sure the difference is due to rounding at some point in the calculations, so I see that my earlier statement was indeed wrong, that over time, it doesn't matter if you're more likely to connect with all hits due to less swings, because the damage over a long interval will be the same.

I'm sorry if that was readily apparent to everyone else, I guess I just saw getting a max return more often as a plus, not accounting for the greater loss when misses do occur.

All that aside, really that was just a minor - and apparently incorrect - point in my post. The best reason I prefer Pyrrhic Kleos is because of the evasion down effect.

tl;dr - doctorugh is right, there is no difference.

Edit: I did more thinking and realized this isn't the full picture. The above is only true in the case of capped accuracy.

If accuracy is < 95% on a WS, then less hits in a WS will be beneficial:
Say you're accuracy is 85%. Then your first hit will have 95% accuracy, and all the rest will have 85% accuracy. The less the amount of hits, the greater the proportion of 95% accurate hits as compared to 85% accurate hits.
This can be seen in the case of a 1-hit versus a 2-hit most clearly.
With a 1-hit WS, the WS accuracy would be 95%, and 95% of hits would land.
With a 2-hit WS, the accuracy for the two hits would be 95% and 85%, resulting in 90% of hits landing.

So, in the case of <95% accuracy, the 1-hit WS will have proportionally less misses. The benefit to less hits will become greater as accuracy gets lower. This can be extended to 5 and 6-hit WS as well, where it's not as large a difference, but there is a difference.

Edited, Sep 7th 2010 10:59pm by apresmoi