With the changes in 2.5, shadows/assassins have (in general) been finding that the rotation feels a lot more challenging, particularly on bosses with high movement. Also, there have been a lot of complaints that the “new” rotation is a lot more static and rigid than it has ever been in the past, with no time to spare for utility powers and such. This guide is an attempt to address some of those concerns, as well as generally inform. I’m going to steer clear of the “more challenging” and “new rotation” points except for the following note: nothing about the rotation has changed aside from the 12 second timer, which is 3 seconds longer than the pre-2.5 rotation.

Oh, this is also not a complete guide to shadow/assassin tanking. I’ve considered writing such a guide in the past, but frankly Fuyri and Tennebras have that pretty well covered, and while I don’t agree with every detail in their guides, I agree with everything important. Thus, I will refer you to their guides together with my general tank gearing post for more in-depth information things that are not rotation.



This video was done in full Dread Forged gear sans implants (which are Kell Dragon), mitigation itemized according to the “Average” profile from my tanking stats post, and all Force Wielder armorings and hilt. Quesh world boss (so that I get defense/shield based regen). 4 minute duration.

The point of this video is to exemplify how precisely I go about maintaining my Shadow Protection/Dark Protection stacks in a boss fight. You’ll notice that at no point do the stacks fall off. In fact, at very few points are the stacks less than 3 seconds from expiry, though this is harder to see since buffs do not have timers on them without mouse-over. The Quesh world boss isn’t the best thing for demonstrating what to do here, because it lacks significant movement, but it gets the job done. There is a brief movement phase at the 3 minute mark, which shows what you do when you’re forced to move around at critical points in your rotation (i.e. during or immediately following your channel). The accuracy debuff is maintained at all times, except for a couple of seconds where I simply made a mistake in not refreshing it slightly earlier. Kinetic Ward/Dark Ward is also given 100% uptime without premature refresh, though given that I’m defending/resisting everything the boss does, KW/DW uptime is much less exciting than it could be.

One key thing to notice here is that I don’t actually need to finish any of my channels! The first tick is always sufficient, because I never let the stacks drop in the first place. Thus, if I had to, I could keep moving 100% of the time and never, ever stand still to channel without losing any mitigation (*cough* PvP *cough*). On an actual boss fight, force management does tend to get a bit tricky if you’re not allowed to complete the channel (as Telekinetic Throw/Force Lightning is force-positive), but you have enough leeway in the rotation that you can always squeeze in enough Saber Strikes to keep up with your force.


In a single sentence, the shadow/assassin rotation can be summarized as this: use Project/Shock and Slow Time/Wither exactly on cooldown, and channel Telekinetic Throw/Force Lightning as soon as you have 3 stacks of Harnessed Shadows/Harnessed Darkness.

The structure of the rotation is exactly as follows:

proc1 > proc2 > gcd1 > gcd2 > gcd3? > proc3 > gcd4? > gcd5? > channel

When I say “procX”, I basically mean either Project/Shock or Slow Time/Wither; which is to say, abilities which proc Harnessed Shadows/Harnessed Darkness. You will always be able to use both Slow Time/Wither and Project/Shock consecutively immediately following your channel (Telekinetic Throw/Force Lightning) unless something has happened which caused the rotation to get out of sync. Falling out of sync happens most often when you have to move away from a boss and then “reopen” the fight after a phase change. Coming back into a reopening with more than 0 and less than 3 stacks results in abilities drifting out of the structure I have above. In that case, just fall back on the maxim of “use it on cooldown” and you’ll be fine.

So you use two procing abilities immediately following your channel. Provided that things aren’t out of sync, they will always be off cooldown. In fact, provided that things aren’t out of sync, the abilities should be precisely the following:

Slow Time/Wither > Project/Shock > …

This is because you always (barring falling out of sync) proc your third stack using Project/Shock due to its shorter cooldown, which means that the first procing ability off cooldown will be Slow Time/Wither, with Project/Shock coming off CD immediately thereafter. If both Project/Shock and Slow Time/Wither are off cooldown at the same time, Project/Shock has higher priority (the reasoning here will be explained in a moment).

Now you have essentially three GCDs (labeled “gcd1” through “gcd3”) which are filler while you’re waiting for Project/Shock to come off cooldown (as it will come off CD faster than Slow Time/Wither). You can use these GCDs for whatever you want. They are only significant in that they increase your DPS or utility. Thus, you can move around in these GCDs, get stunned, push adds, go AFK, etc…

Traditionally, these GCDs are filled with some combination of Force Breach/Discharge (if the accuracy debuff is falling off), Double Strike/Thrash, Shadow Strike/Maul and Spinning Strike/Assassinate. Double Strike/Thrash, Shadow Strike/Maul and Spinning Strike/Assassinate are particularly exciting as they have a chance to finish the cooldown on Project/Shock and essentially abbreviate your rotation. This proc is exciting when it happens, but ultimately it isn’t necessary! You do not need this proc to maintain your stacks, it just makes the rotation more dynamic, that’s all.

Assuming that you either get no procs or you had to fill these three GCDs with something other than Double Strike/Thrash, Shadow Strike/Maul or Spinning Strike/Assassinate, then Project/Shock will come off cooldown in exactly the “proc3” position listed in the rotation structure above. Use it immediately. You have now proc’d exactly three stacks of Harnessed Shadows/Harnessed Darkness. The optimal thing to do in terms of threat and rotational safety is to channel Telekinetic Throw/Force Lightning immediately and start the rotation over from the beginning. This is what I do most of the time. However, you are also perfectly free to inject two additional GCDs before you channel, though the second GCD is playing with fire unless your connection is very lag-free. These GCDs are new in 2.5, since it was previously utterly mandatory to channel as often as you possibly could, whereas now there is no particular benefit in channeling before the stacks are about to expire (aside from extra threat).

For the above reason, I tend to reserve gcd4 (and if necessary, gcd5) for Spinning Strike/Assassinate if it is up (since this is optimal in terms of DPS), Force Breach/Discharge if the accuracy debuff will expire in less than 6 seconds (I do this at least twice in the video), or any utility powers that I need to apply. I try very hard not to use gcd5 at all, since I know that I’m far from perfect in terms of executing the rotation, and I would rather refresh the stacks a GCD early rather than risk them falling off. If I’m not in the execute phase, no utilities are needed and the accuracy debuff is secure, I’ll just channel right away here as it is the optimal thing to do in terms of damage and threat, in addition to making the stack refresh a bit more secure.

Once the channel is finished, you start the rotation block from the top. Note that, because of the Particle Acceleration/Energize proc, it is entirely possible that your rotation will actually look like this:

Slow Time/Wither > Project/Shock > Double Strike/Thrash (proc) > Project/Shock > Telekinetic Throw/Force Lightning

It’s very nice when this happens, because it means that you’re refreshing your stacks after only four GCDs, which is to say 6 seconds or half the duration of the buff. Generally, when this happens, I will dither around a bit and maybe toss in an extra Double Strike/Thrash after Project/Shock to see if I can reset the cooldown before I channel, essentially amortizing the luck that I saw in the current block and making the next block a bit easier.

Just to reemphasize the point: the Particle Acceleration/Energize proc is not needed to maintain your stacks! If you don’t believe me, go practice your rotation on the training dummy while standing at 8 meters (out of melee range). As long as you use Project/Shock and Slow Time/Wither on cooldown, you will always be able to refresh your stacks with exactly 3 seconds to spare (corresponding to the gcd4 and gcd5 slots). Always. Because Particle Acceleration/Energize isn’t essential, or even a survivability gain (as it was pre-2.5), you shouldn’t fear using Shadow Strike/Maul or Spinning Strike/Assassinate instead of Double Strike/Thrash whenever they’re up, despite the fact that they have a lower proc chance. You also shouldn’t be concerned about space in which to use your utility powers, like Force Breach/Discharge, since foregoing the proc on Project/Shock is not going to have any effect on whether or not you can refresh the stacks in time.

Thus, while the position of your procing abilities (especially the first two) is extremely rigid, most of the shadow/assassin rotation is actually quite dynamic and free to change and shift in response to situational requirements and DPS-increasing procs (such as Shadow Wrap/Conspirator’s Cloak). The addition of gcd4 and gcd5 in particular is very, very welcome and represents a significant loosening in the strictness of the rotation from what it was pre-2.5, where gcd4 and gcd5 couldn’t exist due to the need to channel ASAP.

Project/Shock vs Slow Time/Wither

I mentioned earlier that you always want to use Project/Shock before Slow Time/Wither whenever you have the option to do so. The reason for this is quite simple: it gives you a free additional GCD without making the channel timing any tighter. Consider these two alternatives (assuming no procs):

Project/Shock > Slow Time/Wither > gcd1 > gcd2 > Project/Shock > Telekinetic Throw/Force Lightning
Slow Time/Wither > Project/Shock > gcd1 > gcd2 > gcd3 > Project/Shock > Telekinetic Throw/Force Lightning

Both rotations are bounded entirely by the 6 second cooldown on Project/Shock. In every rotation block, you are essentially waiting for your second Project/Shock to come off cooldown in order to get your third stack of Harnessed Shadows/Harnessed Darkness. This is true regardless of which ability (Project/Shock or Slow Time/Wither) you use first. However, using Project/Shock first means that the rotation block is at its absolute minimum in terms of duration, whereas putting Slow Time/Wither first adds a mandatory GCD at the start of the block, delaying the end of the cycle by 1.5 seconds.

Thus, another way to look at the rotation is purely in terms of Project/Shock. Basically, we have the following:

gcd1′? > Project/Shock > gcd2′ > gcd3′ > gcd4′ > Project/Shock

Assuming no procs, this is the shortest block duration possible. The four GCDs around Project/Shock (annotated with the ‘ character to differentiate from previous use of the labels) will always include exactly one Slow Time/Wither. Thus, every rotation block has at least two spare GCDs (out of five) in which you can inject any one-GCD ability without consequence. In the optimal case, the Slow Time/Wither usage will come between the two Project/Shock activations. When this happens, the optional gcd1′ simply disappears, leaving only the three GCDs in the middle and shortening the entire block by 1.5 seconds. This is an optimal case, since it means that you have built 3 stacks of Harnessed Shadows/Harnessed Darkness with an enormous amount of time to spare. This leaves you the widest margins for utilities and fight mechanics such as movement, stuns, knockbacks, tank swaps, etc.

In a sense, the optional gcd1′ positioned before Project/Shock in the above block diagram is actually part of a set of three GCDs which can be placed anywhere in the rotation block. More fully:

gcd1′? > Project/Shock > gcd2′ > gcd3′ > gcd4′ > Project/Shock > gcd5′? > gcd6′? > channel

When we use Project/Shock first in this block, we are liberating ourselves to place gcd1′, gcd5′ and gcd6′ anywhere in the rotation without delaying the channel (or remove them entirely, if that’s what is situationally best). When we cannot use Project/Shock first in the block, we are forcing ourselves to use one of these three precious GCDs in a very specific place: at the start of the block. This is true even when that GCD is being spent on Slow Time/Wither. This removes flexibility, since it means that we can no longer drop these GCDs if we experience downtime during the block (i.e. spend the GCDs on “empty” space, likely being knocked back or moving). It makes a situational decision about the tail end of the rotation before we know what the situation of the fight will be at the tail end of the rotation, which is why we don’t want to do this if we can help it.

This implies a very strong maxim about the shadow/assassin tank rotation: USE PROJECT/SHOCK AS OFTEN AS POSSIBLE. When you delay the first Project/Shock in your rotation block, you are effectively betting your stacks on the hope that you won’t have any delays in using your second Project/Shock. Don’t do this! Project/Shock will always, always be off cooldown at the very latest in the second GCD of every block.

I’ll make this simple: if you don’t use Project/Shock in either the first (optimally) or second GCD after you channel Telekinetic Throw/Force Lightning, you’re doing it wrong. If you’re having trouble maintaining your Shadow Projection/Dark Protection stacks, this is probably why.

Note that this implies the ideal opener for a shadow/assassin tank, just based on priorities, is the following:

Force Pull > Project/Shock > Slow Time/Wither > Force Breach/Discharge > Double Strike/Thrash > Project/Shock > (Force Potency/Recklessness) Telekinetic Throw/Force Lightning

Following this opener, you will have a block which looks like this:

Slow Time/Wither > Project/Shock > gcd1 > gcd2 > gcd3 > Project/Shock > gcd4? > Telekinetic Throw/Force Lightning

Barring any fight mechanics – because you know, those totally aren’t a thing – all of your subsequent rotation blocks will look like this second block above, with varying abilities placed in gcd1 through gcd3 (and potentially eliminating these GCDs if Particle Acceleration/Energize procs early) and optional abilities being applied in the one GCD following Project/Shock as necessary.

Force Management

I’m not going to say much here, since honestly on most fights, force management is really, REALLY easy. This is probably the number one reason why vanguard/powertech and (especially) guardian/juggernaut tanks consider shadow/assassin tanks to be extremely easy-mode. We just don’t have to think about managing our force 90% of the time.

However, there are situations where this isn’t the case. This is most noticeable when off-tanking, since we miss out on almost 40% of your resource generation by not shielding/defending attacks (the skill tree tooltips are bugged, FYI). There are also some bosses which don’t hit frequently enough to keep up with the force drain of our rotation (Dash’roode and Grob’thok being decent examples). Finally, there are some situations even on average hit-rate bosses where force can become an issue for mechanical reasons. For example, if you get unlucky with the droids on Nefra, you may find yourself being forced to consistently break your channels in order to move out of the circle. When this happens, the rotation becomes much more resource-constrained than is ideal.

Fortunately, there’s more than enough room in the rotation to account for these situations. The solution is to essentially go into all-out efficiency mode, where “efficiency” is code for “boredom”. This rotation looks like the following:

Slow Time/Wither > Project/Shock > Saber Strike > Saber Strike > Saber Strike > Project/Shock > Saber Strike > Telekinetic Throw/Force Lightning


This rotation has the following costs: 20 + 26 + 0 + 0 + 0 + 26 + 0 + 30 = 102 force. The total duration of this rotation is 9 GCDs, or 13.5 seconds. During that time, assuming no regen procs from shield/defense, we will regenerate 10.4 * 13.5 = 140.4 force, which means that we are actually force-positive here by 38.4 force. Thus, if you find yourself in a situation where you’re running very low on force, throw in a rotation block which looks like the above and you should find yourself back into safe waters.

The important thing, really, is just to make sure that you have enough force to keep Project/Shock on cooldown and use Telekinetic Throw/Force Lightning before the stacks expire. The easiest rule of thumb to accomplish this is to always ensure that you have at least 26 + 30 – 15.6 = 41 force at the end of gcd3. Remember this rule of thumb for resource management and you should have absolutely no force-related issues maintaining your stacks.

Most of the time, I don’t pay very much attention to my force bar. This is because, most of the time, I don’t need to.

Expert Refinements

One particularly nice refinement to the rotation structure given above was suggested by Tenebras. Specifically, rather than channeling Telekinetic Throw/Force Lightning as soon as you have 3 stacks of Harnessed Shadows/Harnessed Darkness (as used to be optimal), glance at your Shadow Protection/Dark Protection buff to see if you have a little bit of time left on it. If you have a GCD to spare, then inject a Double Strike/Thrash, Spinning Strike/Assassinate or Shadow Strike/Maul. It doesn’t actually matter what you put here, as we’ll see in a second.

Practically, this alters the structure to be the following:

Project/Shock > Slow Time/Wither > gcd1 > gcd2 > Project/Shock > gcd3 > Telekinetic Throw/Force Lightning

The massive, massive advantage to injecting gcd3, rather than channeling immediately, is that you now have aligned the cooldown of Project/Shock with the tail end of Telekinetic Throw/Force Lightning. It doesn’t matter what you do in gcd3, Project/Shock will always be off cooldown following your channel! This means that you can reliably put Project/Shock first, and thus cut a full GCD off of your next proc delay on Telekinetic Throw/Force Lightning (as explained in the previous section).

This is most definitely something that requires a bit of practice though. If you had some delay before your first Project/Shock, or if you had to delay your second one for whatever reason (less common), then you need to straight into your channel to avoid losing your stacks. The easiest way to do this, actually, is to just count GCDs. The buff from Shadow Protection/Dark Protection lasts precisely 12 seconds, which is 8 GCDs. You want to leave a spare GCD to account for lag and movement and such, so that’s 7 GCDs that you can use. Project/Shock and Slow Time/Wither take the first two, and Project/Shock rounds out the second-to-last GCD. You will always have at most 2 GCDs between Slow Time/Wither and Project/Shock (possibly less), which means that you can rely on there being not just one, but twoGCDs to spare at the end unless you had to delay your first Project/Shock.

This is really the kicker. If you can use Project/Shock immediately following Telekinetic Throw/Force Lightning, then you’re absolutely safe to inject the extra melee attack prior to channeling. If you had to delay by a single GCD, then you’re still safe, but you need to start watching out. If you delay by more than one GCD, or if you have to move a lot in the middle of the rotation (delaying the second Project/Shock), then you probably need to go straight into the channel just to be safe.

This is a situational adjustment to the rotation, and definitely something that takes a while to become accustomed to. It makes the biggest difference in DPS once you hit the execute phase, since it allows you to use Spinning Strike/Assassinate almost on cooldown, rather than only once per rotation cycle. It also increases the effective crit chance on Project/Shock by a very sizable amount, as well as increasing the number of Shadow Strike/Maul activations throughout the fight. In short, it’s a pretty noticeable DPS jump, on top of giving you better control over the margins on refreshing your buff.

The only place where I categorically do not use this trick is in the opener. Double Strike/Thrash just doesn’t do enough threat for me to be confident holding agro while using it. Additionally, injecting that extra GCD throws off the timing of the rotation with respect to the taunt debuff, forcing me to either delay my first taunt (almost always resulting in an agro rip), or to potentially break my Force Potency/Recklessness-enhanced channel in order to AoE taunt back if agro is ripped immediately after the taunt debuff falls off (in the exact middle of the channel). In short, it’s not great. Thus, I still use the “channel as soon as possible” opener, and then start injecting the pre-channel melee attack as soon as I have the margins after the first rotation.

Hard Priority Queue

In summary, you want to observe the following priority queue for optimal shadow/assassin mitigation, threat and DPS:

  1. Telekinetic Throw/Force Lightning (if 3 stacks of HS/HD and <1.5s on Shadow Protection/Dark Protection)
  2. Telekinetic Throw/Force Lightning (if 3 stacks of HS/HD and MECHANICS INCOMING!!!1!)
  3. Project/Shock (if <3 stacks of HS/HD)
  4. Slow Time/Wither (if <3 stacks of HS/HD)
  5. Force Breach/Discharge (if accuracy debuff expiring)
  6. Spinning Strike/Assassinate (if above 51 force)
  7. Force Potency/Recklessness + Telekinetic Throw/Force Lightning (if 3 stacks of HS/HD)
  8. Shadow Strike/Maul (if above 46 force)
  9. Telekinetic Throw/Force Lightning (if 3 stacks of HS/HD)
  10. Double Strike/Thrash (if above 49 force)
  11. Saber Strike

You’ll notice that optimal use of Telekinetic Throw/Force Lightning falls in multiple places on the priority queue, depending on your procs, the fight, available cooldowns (i.e. Force Potency/Recklessness), and so on. This is part of why the shadow/assassin rotation is actually a lot more dynamic than it seems at first glance, since really the only two rigid elements of the rotation are (1) Project/Shock on cooldown, and (2) Telekinetic Throw/Force Lightning before your stacks expire.

Those of you who have read Tennebras’s guide (and those of you who are in fact Tennebras himself) will probably notice that this priority queue does not include any facility for exploiting the Force Potency/Recklessness stack removal glitch at the tail end of the Telekinetic Throw/Force Lightning channel. This glitch does technically increase DPS (and TPS) by allowing you to effectively get three stacks of Force Potency/Recklessness rather than two. I think this is a neat trick, but it unfortunately relies on using Slow Time/Wither last in the preceding rotation block in addition to saving a Particle Acceleration/Energize proc. This implies at least two GCDs of delay following your second Project/Shock in the preceding block (one for Double Strike/Thrash and one for Slow Time/Wither). This is technically sustainable, but honestly it is playing with fire (the expiry of your stacks) for a fairly marginal increase in DPS. Two GCDs following the second Project/Shock is safe (with 1 GCD to spare) provided that Project/Shock was the very first thing in the preceding rotation block, but this is rarely the case. More often, Project/Shock is on cooldown for 1 GCD at the start of a block, which means that barring lucky Particle Acceleration/Energize procs between the first and second Project/Shock, this two GCD delay is literally within milliseconds of dropping the stacks. My personal verdict is that it just isn’t worthwhile. Too much risk and not enough reward (the net gain is about 17 DPS, assuming nearly perfect play and a static fight).


The Shadow Protection/Dark Protection stacks are eminently maintainable. The above is how you do it. The end. 🙂