Thief sPvP | Tournament Sword/Dagger, Shortbow | Team fighter S/D Heavy Boonsteal Harassment Thief, by Arga

I'll go over these skills briefly here, and go more into detail on them later.
Sword Auto Attack: deals a decent bit of damage, with the most damage coming from the last skill in the chain, which has a skill-specific coefficient of 1.3 and gives weakness (essentially 25% damage reduction and 50% lower endurance regeneration) and cripple, two very powerful conditions. Also, the first attack is often used in a chain of other attacks on the sword or evades, as explained later in this guide.

Infiltrator's Strike/Shadow Return: This is an extremely useful skill in spite of its lack of damage output and high overall initiative cost. This skill basically allows you to "link" to a point 1200 units away, making you able to teleport between it and an enemy simply by using this skill. This is extremely useful for juking enemies, fighting 1v2's (or more), catching up with enemies, securing stomps, and a very wide variety of other things. It is also one of the two main condition cleansers for this build. Tactics explained in the gameplay section.

Flanking Strike/Larcenous Strike: two more very powerful skills, although they are, in a few different ways, polar opposites of one another. FS is a gap closer that gives you an evade as well, but deals fairly little damage (its skill-specific coefficient is .05 less than either of the first two attacks in the auto attack chain). On the other hand, LS is a very short-range attack that can leave you vulnerable, but deals massive damage (it has a coefficient of 1.5, or twice that of FS). This skill, along with the auto attack and Infiltrator's Strike/Shadow Return, form the backbone of this build. Each has its niche in the build, and together these skills work together very well.

Dancing Dagger/Cloak and Dagger: two highly situational skills. The first skill is going to be used mostly to chase down enemies that are running away from you, or are running towards a point. Cloak and Dagger is an emergency escape button that, when it hits, gives a player a few extra seconds to run away, and when it misses, usually completely screws over a player. This skill can also be used occasionally in 1v1's against opponents to give you some time to grab some seconds of CD as well as planning time. Neither of these skills are going to be used much in this build.

Shortbow: Shortbow serves a few different purposes. For one thing, it is a good way to mess with enemy heals- just throw down Choking Gas and your opponent's heal is automatically reduced by 33%. It's also useful against multiple enemies, of course, given that you're fighting with multiple allies. In this case, the amount of damage that this weapon deals grows hugely, and it often becomes a better choice than Sword even though Sword has higher coefficients to its skills. This weapon is also good for mobility during games, as Infiltrator's Arrow is an invaluable skill for getting you on top of ledges, away from enemies, and generally to where you need to go much faster. This weapon has many other uses, discussed later on in the gameplay section, which is why this build is so good as opposed to dual Sword/Dagger builds, which sacrifice the ability to have a Shortbow in exchange for advantages with weapon-swapping.

Utilities
Roll for Initiative: a very good stunbreaker with an evade on it. It also increases your initiative by an average of .1 per second assuming you use it every 60 seconds, although you often won't use it every sixty seconds as using it when you have 9 or more initiative is squandering the use of some of your initiative regeneration (when you have maximum initiative, you're no longer producing initiative, which you could be using to increase your damage potential instead). Because of its great initiative regeneration, as well as its ability to get you out of sticky situations, this is a great utility to take. The only downside is that it takes so long to CD.

Infiltrator's Signet: provides passive initiative regeneration of .1 per second, and is good as a stunbreaker as well as a method of securing stomps, especially against thieves. If they teleport away while you're stomping them, simply use the signet and you'll be able to finish your stomp easily. It can also be used instead of Shadow Step as a method of creating a Shadow Step stomp, by using Shadow Return while stomping and then using the signet to teleport to your enemy. Overall just a very solid choice.

Shadow Step: extremely useful as a Shadow Step stomping mechanism (an extremely useful and often deceptive maneuver when you're in a 1vX, and especially when stomping a guardian). It is also a very good stunbreaker (and- hopefully it doesn't have to be used as this- a dual stun breaker), condition cleanser (mostly for cleansing many conditions at once while Basi Venom is on CD), positioning weapon, and early-game development utility due to its long range. Another very solid utility to keep on your bar.

10 in Deadly Arts: this does a few different things. First of all, it should be mentioned that the 100 power coming from 10 points in Deadly Arts amplifies the damage you deal by only very slightly less than 30 in Critical Strikes (30 in CS deals about 1.4% more damage than 10 in DA). Of course, crits and power have an inverse relationship. As you increase power, crit stats (precision and critical damage) become more powerful while power amplifies your damage by marginally less; as you increase crit stats, the other crit stat and power both become more valuable. Ultimately, however, by the end of the build crafting, for an sPvP build, these effects tend not to justify putting in 30 in CS as opposed to 10 in DA. The issue is that you only have a limited amount of points you can invest, be they trait points or actual stat points, and by putting in 30 into CS, you're essentially over-inflating the value of your offensive stats so that you've spent a ton of points increasing your offensive amplification and thereby keeping you from putting those same points into, say, more defense amplification. Also, I should mention that a large number of the traits that you take in CS are near useless, and that Executioner tends only to be good for burst against weakened targets, as otherwise it tends to be fairly counterable just by timing your heals correctly.

That aside, by putting 10 into DA, you're able to take Mug, which provides a small but, IMO, decent heal and a bit of damage as well, as well as a poison. Since you can steal often with this build (due to the 30 in Trickery), 10 into Deadly Arts becomes extremely valuable. It can provide heal mitigation against your opponent as well, due to Serpent's Touch. Just cast steal as your opponent is healing, and their heal will be reduced by 33%.

30 in Acrobatics: this serves multiple purposes. First of all, increased boon length is always something nice to have, especially given that we run traits like Thrill of the Crime and Bountiful Theft in this build. This trait line also gives you plenty of vitality (which, at lower levels of health, is a more useful stat for increasing how much damage you can take than toughness is) and good offensive amplification via Fluid Strikes. Vigorous Recovery in Acro gives you plenty of vigor, which is really nice for this build. Quick Recovery gives you a great source of passive initiative regeneration (which pairs with Infi Signet to give you an average of .3 initiative per second). Finally, Assassin's Reward is such a powerful trait, especially for this build, that it's really difficult for me to describe how useful it is. By regenerating 69 health per point of initiative used, and with high initiative regeneration, while your heal skill is still on cooldown, you can heal a lot with Assassin's Reward. I find this trait superior to Quick Pockets (which would turn this build into an S/D S/D build) as it gives you the opportunity to take a Shortbow and thereby be more flexible and mobile. QP also, of course, sacrifices all of the benefits of taking AR, and is one more thing to have to build around and think about in battle, making it somewhat inefficient.

30 in Trickery: this is a fantastic traitline with multiple virtues. The first of all is of course the initiative regeneration. Hastened Replenishment, Kleptomaniac, Preparedness, and the reduced Steal CD all increase initiative regeneration. This trait line also makes your steal significantly more powerful, and gives you your other major source of boonsteal in this build outside of Larcenous Strike, which is, as an individual skill, more valuable because skill has an instant cast as well as 900 range (1500 if traited). This trait line gives you Lead Attacks, which is an extremely flexible minor trait for increasing your damage. Typically, it'll be used most actively against more bunkered builds: you'll simply resort to your auto attack more and build up initiative to the point where you deal 15% more damage per attack. Finally, this trait line gives you the opportunity to swap to Shortbow and deal more condition damage, if necessary (and throwing down lots of condis with SB is actually more common than you might expect). Overall a very powerful trait line.

As a thief, I like to keep up with the current metagame as much as possible. I think that this build brings some unique ideas to the table that only scratch the surface of how deep and complicated this game is.

I showed you some math on this build, for example. What I did not show you were various facts about the math, like the fact that the equation for the marginal benefit of an additional point of precision in a build is entered in as

=((1-(1/21*0.01+B19)+((1/21*0.01+B19)*(B11+B4)))/(1-B19+((B19)*(B11+B4)))-1)*D21+1

on my spreadsheet for build crafting. However, these are the kinds of things that underly this game, making it... Insanely complex. I don't claim to know everything about the mathematical structures that we find in GW2; for example, it's incredibly difficult for me to give you accurate figures about the marginal benefit of, say, additional initiative regeneration, for example, or the effect of cripple on damage-dealing ability. Let it suffice to know, however, that this build does at least contain some of the ideas embodied by the mathematics in this game, which is something that most builds don't do. The majority of other thief builds, for example, will tell you that Berserker's Amulet is superior to any other amulet, even though, as I showed (to some extent) earlier, it's not. I've had people come and tell me that I don't know how to play thief because of this. I've had people whose teams I've applied to turn me down because I don't run a Zerker Amulet. What none of these people realize are that these are perceptions based off of ridiculous ideas with no true basis in reality, and hopefully this build and guide has helped you at least consider this basic concept. Granted, there are thousands of other factors that go into consideration here, and which are the current subject of my mathematical and economic studies of this game, but in both my play testing and what math I have been able to uncover from the game, Soldier's amulet tends to be hugely successful, while zerker's isn't quite so much.

Of course, there are other unique aspects to this build. The ability to maintain high initiative regeneration without having to run S/D S/D is a huge plus. The unique trait choices- 10/0/0/30/30- also happen to be extremely good ones, IMHO. Really, just about everything in this build flows so consistently within itself, that I do believe that this build could be the new thief meta, if not for the biases facing the thief community.

There's a lot I can say about this build left; for example, it caters to beginners to the weaponset or the thief class fairly well, but at higher levels is so extremely complex that it would make other builds look somewhat pathetic. Its 1vX'ing ability and team support are both incredibly useful, allowing you to get into the midst of battle, deal great damage, and all the while take relatively little damage yourself. But anyways, I'm tired; I'll make sure to edit this guide more in the future. If anyone is interested in knowing about the build or the mathematics, just PM me on the forums, and I'll try to get back to you as soon as possible.

Thanks guys!
~Arganthium

Here, I'll just put down some questions that people have asked me about the build and the guide:

Q: I’m not sure I completely followed your arguments about offensive power and your comparison to jumper’s dps. Maybe you could walk us through it a bit slower?
I also think it’s complex as your build is going to clearly have advantages when it comes to things like boon ripping (which S/D already does so well anyway).
Along those lines, I wonder about taking both Bountiful Theft and Vigorous Recovery together. Aren’t you sort of over-killing the vigor? With either of those you’d have like 90% up-time on vigor – with both, you only lose vigor if someone kills your boons. Instead you could be taking power of inertia, initial strike (since you seem to be fond of auto-attacks), flanking strikes (for another flat damage boost – on top of your 15%) or even uncatchable. Although Bountiful theft does combo well to rip a lot of boons at once with larcenous strike – how often do you actually pull that off? How often do you need to rip 4 boons, and couldn’t do it with larcenous strike alone? Is it enough to justify the trait? Meanwhile, there are some other great traits that you could be taking that would synergize great with this build.
Anyway, interesting ideas, and I’m excited to give it a try.
Edit: I realized that part of the goal of your build is team battles – so perhaps bountiful theft is worth it for the team benefits? So that leaves you with maybe 60-70% vigor up-time (not including random boons from those lyssa runes). I think power of inertia may then be worth it over vigorous recovery. That’s an extra 4-5 might stacks most of the time in a build like this.

A: So basically, when I evaluate different builds mathematically, I always compare them to some type of baseline. Coming from a thief background, that tends to be the amount of damage that a thief does when he/she is wearing full armor but has zero additional stats from runes, sigils, traits, etc. Then, I tweak the traits here and there, and, basically, my little mathematical GW2 machine pumps out numbers saying by how much I’m increasing my damage, relative to the original thief with full armor but zero additional stats. At one point, I stated that the amount of damage this build does, over the course of its lifetime, is equal to about 5.47 times the amount of damage a thief with full armor but no additional stat points (for the sake of simplicity, I’ll just call it a “base thief” for the rest of this comment), assuming that stats are the only things that change, and not enemy reactions, player reactions, etc, etc. Basically, I’m measuring the effects of changes in different stats, ceteris paribus (if you know what that means ).
Following from that, my build turns out to amplify the amount of damage I can take by 2.38 times, and the amount that I can deal by 2.3 times. Multiply those together, and you get damage over lifetime, relative to the base thief. This value is 5.47, or 547% of the damage of the base thief. On a similar string of though, Jumper’s build increases lifetime damage by a factor of 526.8% relative to that same base thief.
Of course, this is just discussing stat points. This does not take into account whatsoever other factors like healing, init regeneration, boonstealing, etc. I prefer to write about that kind of stuff in other places, as things like those can sometimes be subject to some degree of bias, which it is my goal to eliminate. Hopefully that explains what I meant to you well enough. I know that I can be bad at explaining things sometimes. xD
Anywho, my original 10/0/0/30/30 build did actually run Inertia (you can see a few vids of me with it here. The thing about it is, for a lot of classes, applied vigor doesn’t often last particularly long, so it’s often hard to steal a significant enough amount of vigor to really do you any good. Also, because I have to be very specific about when I want to heal, I usually don’t keep up particularly good vigor uptime just from boonsteal and Bountiful Theft (btw, on a side note, Bountiful Theft is good because of the vigor, and all-around team boonsteal, and is especially good against boon-heavy builds- just check out my quadruple boonsteal combination). I mean, when I practice against golems, I’ve been able to increase my vigor uptime by a significant amount of time (hit about 1 min 45 secs when I was training on golems before a tourney today), but in practicality, you’ll typically not have vigor even nearly that long because of the timing on your heals. Personally, I’ve found that the evades are really good for holding points, which is my main objective. However, I’ll take your idea into consideration and try PoI a bit more soon. c:
Oh, and btw- quadruple boonsteal is far more common than you might think. Especially against HGH engineers and Shout Guardians. In their situations, the quadruple boonsteal combo is just… Absolutely godly, but it’s also necessary because- for example, with HGH- additional boons flowing in “buffer” the more important boons for you to steal, like Might, meaning that an opponent could potentially generate boons fast enough for you to be unable to steal the boons you really need to steal. BT, however, is a much safer boonsteal (insta-cast) that also provides team support, and allows you to get around the boon-buffering problem.

Comment: The thing I dislike about investing so heavily in trickery, is that power based builds give up so much potential raw damage and pressure, for team utility. Utility that is easily brought by other means, or other classes with less detriment to damage output.
If you want to refill your allies’ endurance, Signet of Agility does that, instantly, on a very low CD, although comes with the caveat of not knowing their endurance and guessing.
I think the fact that you can steal 4 boons is wonderful, but you’re also allowing your target to live longer, indirectly giving them more opportunity to reapply those boons anyway. Currently, the most powerful boon to be stolen, Might, doesn’t steal the full no. of stacks last I checked.
Furthermore, many classes bring their own fury/vigor/swiftness close to 50-100% uptime, and the sharing radius is not huge. Building all this utility into steal also makes you think twice about using it on CD, despite it being only 21s-ish. Do you open with it at 100% HP and waste the majority of the Mug potential? Do you use it to chase/finish someone and/or stay alive but miss out on boonsharing because there’s no one near you?
Every second you keep it off CD though, you’re wasting all those trait points which are reducing its CD. I guess what I’m saying is, you mention that Critical Strikes traits are useless/wasted/avoided, and I feel like that applies to many traits for all classes.

Response: While I might agree with you about the lack of damage potential from not using DA or CS by instead investing heavily into Trix, the other thing you have to remember is attacking rate. I’ve tried out builds before that only invested heavily into those stat points that mathematically did the most for my build; all of them failed miserably. The truth of the matter is, every trait in Trix (well, that we use in this build, anyways) is good, and the only thing that’s a real shame about it is that it builds up condition damage. That’s the real reason to invest so heavily in to Trickery; while the direct advantages are somewhat lacking, it provides absolutely excellent initiative regeneration as well as a large number of boons (and the quadruple boonsteal, which is incredibly useful against certain builds). IMO, the offensive advantages granted by investing into Trickery thereby is greater than the offensive advantages granted by investing as heavily into DA or CS. Neither of those trait lines provide particularly good initiative regeneration options.
Also, you’re missing out the point behind Steal. It’s an all-around class counter. It doesn’t matter if opening with it sacrifices some Mug potential if the total damage that I deal as a result of using the steal early on as opposed to later is greater than the healing I would have gotten from it later on. Also, remember that when you aren’t using steal, that it’s not on CD. That’s much more important than you may think; the ability to keep up large amounts of fury, might, swiftness, and vigor uptime is huge. Either way, stealing is a very powerful countering weapon, and every single part of it is useful, even if not all parts are used together simultaneously.
And, like you said, it is true that, for many classes, the traits taken are often quite lackluster, as is true for CS. The thing is, however, that this shouldn’t have any effect on our build crafting; we don’t need to build crappier builds because our opponents are building crappier builds. We should aim to make our builds powerful relative to all builds, and that’s why I don’t approve of CS; one, maybe two of the traits are useful whatsoever. This build aims to get rid of as many such unnecessary traits as possible in its trait choices. Also, Signet of Agility is a perfectly viable option, but I don’t take it because of the relatively short radius, the fact that it doesn’t stun break, and the utility otherwise granted by other utilities that I already use. But if you want to try playing with it, go ahead.