Ahh, good I did miss that 20% you are correct in the numbers. I still am very unsatisfied with this "proposed" change. Hopefully we can get enough people to voice there opinion and reactions to this change to reveal these numbers.

You are calculating it incorrectly. Please see the original post of this thread.

 

The 20% armor increase from Stasis (Eye of the Storm for Assassins) is additive (not multiplicative) with the 155.8% increase from the stance. If you have the Stasis/Eye of the Storm, the number that shows up in your character panel with the stance off is 120% of your base armor value. The number that shows up with the stance on is 275.8% of your base armor value.

 

Note that 275.8/120 = 229.8% = +129.8%, which is exactly the increase you see.

Psy

There is some very valuable work in this thread. Is it possible to quantify and include the contribution of cooldowns though?

In fact, for PvE our squishiness *is* almost perfectly balanced currently (at least with Vanguards).

This line in particular got me thinking that if "sustained" squishiness was balanced wouldn't cooldowns also need to be balanced in order to keep the classes even?

It looks very much as though my shadow will go from being a perfectly viable 1.2 Ops tank to being a 1.3 "well, if we can't find a 'real' tank I guess we'll let you come as an off tank as long as you have FULL BH gear....". Certain bosses and pulls already favor a heavy armor tank to the point that 2 shadows aren't viable for 1.2 Ops. Now it looks as though taking even 1 shadow will almost always be a detriment to the group.

This also makes me wonder how a shadow will ever get "geared" now, since the only time they're not going to be hurting the Op is if they already have a complete set of the gear that drops in the op. Guess they'll have to grind out the Black Hole Heroic every week for about 6 months to get the commendations that way. Gosh, that sounds like fun doesn't it?

Some more numbers for comparison.

 

Here are the modified stats (after 1.3):

 

Shadow:

 

R = Kinetic/Energy Damage Reduction: 40.25%

D = Defense Chance: 22.53%

S = Shield Chance: 64.91%

A = Absorb Percentage: 54.04%

 

Vanguard:

 

R = Kinetic/Energy Damage Reduction: 53.78%

D = Defense Chance: 13.64%

S = Shield Chance: 52.91%

A = Absorb Percentage: 62.04%

 

So a 1.3 Shadow in full Campaign gear will take 12.4% more damage than a Vanguard in full Campaign gear, including ALL defensive combat stats.

 

I do not see how our self-heals could possibly make up 12.4% of a boss's damage, even before they are nerfed by ~40% themselves...

 

 

This means without nerf, we would take 5.2% more damage than a Vanguard. That seems perfectly in line with what our self-heals are currently capable of, given that they are frequently interrupted or outright unusable due to mechanics (c.f. Kephess)

Cut down your post but good job on working out the values. Comparisons to pre- and post-1.3 nerf aren't particularly useful unless compared to other tanks so really appreciate what you've done.

Healers in my guild have always known that vanguards are easier to heal. Its not necessarily that they take less damage (5% less is negligable) but the damage they take is more consistant, whereas shadows are more spikey. Post 1.3, shadows are going to become spikier still. Where shadow's have shone so far has been aggro - its just extremely easy as a shadow. With 1.3, we lose that advantage.

As to whether we can make up that 12% damage difference through self healing, I'm not convinced. Thought I'd post some logs of me tanking (mostly rakata, switched some mods around etc) so you can see damage taken vs healing done:

EV Hardmode

Solo-tanking Jarg and Sorno HM

EC Story Mode Bosses 1-3

Kephess Story Mode Kill

As you can see from these logs, on easier fights, self-healing is indeed quite overpowered. On the tanks, for example, I usually take the easy tank who only does a couple of big hits in a row. I can practically heal myself back up before he hits me much again! However, self-healing does not scale, whereas mitigation does! This is the whole problem surrounding this nerf:

On easy fights (heroics, world bosses, flashpoints), Shadow self healing IS overpowered, we can practically keep ourselves alive (e.g. Esseles HM last boss I can heal myself for the most part). However, when things get hard and the damage taken increases, self-healing is inferior to true mitigation.

This same logic holds true in pvp. In small scale fights, the self-healing is bordering on overpowered because it mitigates a larger percentage of incoming damage. In large scale fights, however, my healing remains constant but damage taken increases significantly.

That is why I am worried about 1.3 as far as tanking goes. I'll be able to hold aggro, yes, and I'll be able to tank everything in the game. However, on the hardest stuff in the game, both the other tanks are superior to me already, and will become more so in 1.3. They take less damage and are less spikey in nature so easier to heal through.

Lots of math....

Thanks for doing the calculations. HOWEVER these values are meaningless without factoring in the indirect mitigation that an Assassin's/Shadow's self-healing provides.

Ok, so they end up taking roughly 12% more damage. When viewed as a stand-alone number that sounds pretty bad. Most people that read all that math focus on that 12% to make an argument why Assassins/Shadows are nerfed too much. But you (and the readers) don't factor in the self-healing.

(I play an Assassin so I'm using only the Imp names, sorry for this.)

Pre 1.3:

Dark Charge heals for 425 at level 50, with a 50% chance on every attack. So that's a 425 heal with a 50% chance

every 1,5 secs, or on average 142 HPS.

Force Lightning heals 3% per tick for a total of 12%, roughly every 25s (Wither+Shock+Wither=21s). Say you have 20k hp that will mean on average 96 HPS.

Total self-healing: 242 HPS.

Post 1.3:

Passive self-healing: 71 HPS

Active self-healing: 64 HPS

Total self-healing: 135 HPS

^That is not trivial healing, and should not be omitted! Whether it makes up for the loss in mitigation I don't know, only time will tell. But saying the nerfs are too much just because it causes 12% extra damage is not looking at the entire picture, and many people are getting the wrong ideas because of it.

Disclaimer: Im one of my guilds 2 "main" tanks, done KP/EV nightmare a billion times, EC Hardmode 4/4, blah blah blah.

When the logs came out and we ran the numbers through the parsers, we immediately starting suspecting something was a bit off (though just looking at the skill tree gave an hint really). Armor just didn't do that much of a difference, then you tack on deflection and resilience which, if used right, can mitigate an enormous amount of damage (and let you tank for a while during the craziest of enrage). The healing would often account for a huge percentage of all healing received (in some fights, over 30% of all damage taken would be self healed). Add the campaign shield relic with kenetic ward, and that few percent damage mitigation loss from armor really didnt matter much.

When switching to campaign/blackhole geared up vanguard, our healers were crying (not because they couldnt heal it, but it was absolutely much harder). We could see the number, we could see the logs. Shadows were pretty much fine at launch (slightly underpowered), and they were buffed sky high.

When 1.2 hit, i was VERY surprised to see shadows were not being nerfed. Now i'm not surprised at all to see this. The only challenge for a shadow tank in the entire game was keeping threat during the second fight on EC hardmode against campaign geared sentinels, and thats about it (and even that wasn't that hard, but they're removing that challenge now, lol). Just keeping the rotation a little tighter instead of smacking my keyboard while reading my emails will make up for the nerf

No big deal really.

OK. Let me get this straight:

Bioware nerfs the Shadow because PvP players whine and too many PvE players can solo on too many quests.

At the same time, they leave my first Shadow stranded on a PvE that's so deserted by players, it literally hasn't completed it's class quest because there are barely enough people on Illum to form a full group even if everyone went on the same mission.

I've been a friendly, helpful player who has enough friends and enough guildmates that I could finish if I was willing to plead a little. I simply made the decision to start over on another server because I didn't want to pay a monthly subscription for a single-player game.

Someday, they'll allow transfers. Then I'll complete that quest. I'll just have to have a bigger, stronger group since I'll no longer be able to pull as much of my own weight.

=====================

Another factor here is the perception that people play Shadows because they are overpowered. Well, I've played a Shadow since the Beta, when they were generally agreed to be underpowered. Mine was certainly badly underpowered before I read forums and learned a decent build and rotation. I've tried most other classes, and simply have more fun with Shadows than any other.

Some more numbers for comparison.

 

Using the known DR formulae and the known combat procedure, it is possible to calculate the fraction of incoming damage taken, including mitigation, avoidance, and shielding.

 

For purposes of this study, I compared a post-1.3 Shadow with a Vanguard. In both cases, I used full Campaign gear, with Black Hole Implants and Earpieces. I do not use Rakata because the Rakata itemization does not reflect a reasonable tanking build (ridiculously excessive Accuracy).

 

A full Campaign Shadow has the following stats:

Armor: 2613 (base)

Defense Rating: 199

Shield Rating: 546

Absorb Rating: 409

 

Apart from armor, a full Campaign Vanguard has nearly identical stats:

Armor: 4935 (base)

Defense Rating: 203

Shield Rating: 546

Absorb Rating: 409

 

To calculate the total fraction of incoming damage taken ("Squishiness") I assume that all self-buffs that can be maintained continuously are maintained. That means Kinetic Ward for Shadow and Power Screen for Vanguard. I also include all bonuses from talents and set bonuses.

 

Here are the modified stats (after 1.3):

 

Shadow:

 

R = Kinetic/Energy Damage Reduction: 40.25%

D = Defense Chance: 22.53%

S = Shield Chance: 64.91%

A = Absorb Percentage: 54.04%

 

Vanguard:

 

R = Kinetic/Energy Damage Reduction: 53.78%

D = Defense Chance: 13.64%

S = Shield Chance: 52.91%

A = Absorb Percentage: 62.04%

 

Using the known combat table and related procedure, the fraction of incoming damage taken (Squishiness = Q) can be calculated directly from these numbers:

 

Q = (1-R) * (1-D) * (1+C*M-S*A) /(1+C*M)

 

where C is the enemy critical chance (assume C = 0.1), and M is the enemy critical damage bonus (assume M = 0.5). The value of Q does not depend much on the assumed values of C and M, but they need to be included to reflect the fact that critical hits cannot be shielded.

 

 

Q values:

1.3 Shadow: Q = 0.3083

Vanguard: Q = 0.2744

 

The ratio of these numbers is the percentage of additional damage a Shadow will take compared to a Vanguard:

 

0.3083/0.2744 = 1.124

 

So a 1.3 Shadow in full Campaign gear will take 12.4% more damage than a Vanguard in full Campaign gear, including ALL defensive combat stats.

 

I do not see how our self-heals could possibly make up 12.4% of a boss's damage, even before they are nerfed by ~40% themselves...

 

We can also make the comparison using a Shadow in the current game version.

 

Repeating the calculation but ignoring the nerf to our armor, the values come out:

 

1.2 Shadow: Q = 0.2888

 

which again should be compared to Q = 0.2744 for a Vanguard in identical quality gear:

 

0.2888/0.2744 = 1.052

 

This means without nerf, we would take 5.2% more damage than a Vanguard. That seems perfectly in line with what our self-heals are currently capable of, given that they are frequently interrupted or outright unusable due to mechanics (c.f. Kephess)

 

As a side note, the differences cannot be made up by debuffs, either. Vanguards have a 4% enemy damage debuff which can be maintained continuously, just like Slow Time. And their Smoke Grenade reduces target accuracy by 20% over 30% of the fight (18s uptime/60s cooldown), making it actually slightly better than our Force Breach accuracy debuff. Basically the debuffs (Slow Time + Force Breach vs. Static Field + Smoke Grenade) are a wash.

 

I conclude that the claimed explanation for the nerf (PvE survivability too good) doesn't hold water. In fact, for PvE our squishiness *is* almost perfectly balanced currently (at least with Vanguards).

Thanks for this.

Pre 1.3:

Dark Charge heals for 425 at level 50, with a 50% chance on every attack. So that's a 425 heal with a 50% chance

every 1,5 secs, or on average 142 HPS.

Your numbers are way off here. It's actually 65% after skill tree and many GCD's have multiple chances to proc the effect, but your number is actually way too big because the effect occuring has a 4.5s internal cooldown. So, once you proc the effect on a GCD, your next 3 GCD's have ZERO chance of procing the effect because it's on CD.

Even if the proc chance were 100%, your number would have to be cut in half.

Thanks for doing the calculations. HOWEVER these values are meaningless without factoring in the indirect mitigation that an Assassin's/Shadow's self-healing provides.

To appraise its value, you'd have to look at logs of boss fights and find out incoming dps from the encounter and scale that with the overall mitigation difference to find out the difference in incoming dps and compare that with the hps that shadow tanks get. On easy fights with low damage, the healing model would look very favorable. On hard fights with high damage, the healing model would look very poor.

Although, it's difficult to represent the difference in cooldowns and how they affect fights this way.

The big thing to realize about the self healing is that it only helps if what is causing your wipes is limitations on your healer throughput and resources. If you die from spike damage, the self healing is nearly useless because it stops when you're already dead. Real mitigation and absorb effects (Guardians) can prevent the death from occuring - self healing can't.

So, even if the self healing catches up to even in long term numbers, it's still inferior. The self healing model has to actually be substantially better in the long term numbers to be even worth considering.

The big thing to realize about the self healing is that it only helps if what is causing your wipes is limitations on your healer throughput and resources. If you die from spike damage, the self healing is nearly useless because it stops when you're already dead. Real mitigation and absorb effects (Guardians) can prevent the death from occuring - self healing can't.

The even bigger thing is that unlike armor rating, shield chance, absorb, etc. which the Vanguard relies on, the self-healing is interruptible. As mentioned by several others already, in encounters where you have packs of mobs spamming the ever present dozens of zero-CD spammable stunlock/knockfest abilities, self-healing means nothing because it's constantly negated by the control spam. Nerfing it by half is one thing, but then claiming that what's left somehow "makes up for" also gutting passive mitigation just shows that nobody at BW has actually played a shadow tank in their own high end PvE content.

The problem here is they're just doing a knee jerk nerf in response to PvP whining, basing it solely on their beloved "targets," without actually knowing or caring how the system works in practice in PvE. The result is it's going to screw the class to being something like 10-20% less effective or more difficult to play as a high end PvE tank. As with the overdone screwing of Commando DPS recently, it'll then take months of hemming and hawing and, "playtesting in response to community concerns," and random declarations of "working as intended" before they even consider fixing it, probably in 1.4 or 1.5 next year.

But hey, at least you'll have all those jacked up new "smart" AOE threat abilities, so you can get those wipes over with quicker.

The problem here is they're just doing a knee jerk nerf in response to PvP whining, basing it solely on their beloved "targets," without actually knowing or caring how the system works in practice in PvE. The result is it's going to screw the class to being something like 10-20% less effective or more difficult to play as a high end PvE tank. As with the overdone screwing of Commando DPS recently, it'll then take months of hemming and hawing and, "playtesting in response to community concerns," and random declarations of "working as intended" before they even consider fixing it, probably in 1.4 or 1.5 next year.

The funny thing is, I don't see this changing much at all for PvP.

Due to the fact that self healing does not scale at all, self-healing becomes pretty useless in all fights except small scale stuff. You are more helpful if you cc/interupt/dps/taunt other targets than if you self-heal. The change to armour rating wont affect a thing: 4% less mitigation against less than half the attacks you take in pvp wont even be noticeable. Combine this with the nerf to adrenals/relics and there will be less spike damage = we live longer anyways = we can heal more times during a fight. For example, if a 1v1 with a marauder currently takes 30 seconds and you use TK to heal twice for a total of 24% total health, after 1.3 that same dual might take 45 seconds, allowing you to use TK three times (8% each) for a total of 24% total health....

The difference remains at the top end of pve tanking only. Self-healing made shadows superior for all the easy content because the self heals healed through a significant proportion of damage. On the hardest content where you are taking 1k+ damage per second, the self healing becomes worthless and mitigation much more important. Given that easy content is, you know, easy, it doesn't matter who tanks it. What matters is how classes perform in the toughest of situations. If vanguards and guardians mitigate more damage in the hardest content and are less spikey in nature, then shadows are inferior. This is the way it already is in game, but the threshold where mitigation overtakes self healing is very high.

In easy fights (heroics, world bosses, flashpoints), Shadow self healing IS overpowered, we can practically keep ourselves alive (e.g. Esseles HM last boss I can heal myself for the most part). However, when things get hard and the damage taken increases, self-healing is inferior to true mitigation.

This is, in my opinion, the core of the problem.

Shadows are currently overpowered in 95% of the content. A nerf was in order for that. However, they nerfed our weakness in that 5%.

We are still probably going to be better in that 95%, although not by much, but we will be significantly inferior in that 5%, which are the hardest fights for healers/tanks to begin with.

Our two strengths were self healing and superior aggro management, IMO. Those were both nerfed (one straight nerf, one by buffing everyone else). Our not insignificant weakness is that we could take more spike damage.. and i'm dumbfounded that that was also nerfed.

Lots of good points made above. Unfortunately, I'm afraid nobody is listening. In the past they have proven themselves willing to gut a class's role in PvE to appease PvP QQing (c.f. Scoundrel DPS). I suspect you will see about as many Shadow tanks in top tier operations after 1.3 as you do DPS Scoundrels now (i.e. basically none).

Regarding variability of damage, using the combat table formula earlier in the post and the known probabilities of various combat results (avoid, shield, etc), it is possible to work out the variability of our damage taken compared to Vanguards. (For the mathematically inclined, this is the standard deviation of our "Squishiness" on an attack by attack basis.) As before, this is assuming full Campaign gear

Vanguards: Q = 0.2744 +/- 0.1928

Shadow 1.2: Q = 0.2888 +/- 0.2271

Shadow 1.3: Q = 0.3083 +/- 0.2424

The spikeyness is the number after the +/-, which shows how much our damage taken varies from attack to attack.

The variability ("spikeyness") of our damage taken is already about 18% higher than Vanguards. After 1.3, it will be 25% higher than Vanguards. This is in addition to the Squishiness advantage (currently 5.2%, after 1.3, 12.4%) they also enjoy.

So @#$% depressing.

The spikeyness is the number after the +/-, which shows how much our damage taken varies from attack to attack.

 

The variability ("spikeyness") of our damage taken is already about 18% higher than Vanguards. After 1.3, it will be 25% higher than Vanguards. This is in addition to the Squishiness advantage (currently 5.2%, after 1.3, 12.4%) they also enjoy.

.

/snip

Yeah this is the bit that concerns me why take a shadow anymore when we look like our damage taken is gonna be so god damm spikey, the much more predictable damage of the other two tanks could leave us out in the cold on many ops as why give your healers a nightmare heaing our spikey damage when you dont need too

I just think they shoudl nerf us bit by bit instead of getting the nerf mallet out, as i said before i hate the self healing on TK as on some bosses its useless git rid of it put all our self healing on our stance proc and adjust that to give a nice predictable amount of self heals then gently adjust our armour or healing as required after that

Lots of good points made above. Unfortunately, I'm afraid nobody is listening. In the past they have proven themselves willing to gut a class's role in PvE to appease PvP QQing (c.f. Scoundrel DPS). I suspect you will see about as many Shadow tanks in top tier operations after 1.3 as you do DPS Scoundrels now (i.e. basically none).

I think the hidden issue at heart of all this is really whether or not they are commited to making every spec not just "viable" but competitive in both PvP and PvE aspects of the game.

It's been a common nightmare in virtually every big MMO trying to keep a given spec balanced in both sides of the game. Often times the better forms of balance have occured when a developer just concedes a spec as "designed for PvP" or "designed for PvE" and let it simply be subpar on the other side of the game.

Generally speaking, I'm not against that style of game design. I'd rather have some really good balance than dubious balance across the board.

For example, I'd be entirely okay with Infiltration being the PvP spec and Balance being the PvE spec and both not particularly good in the other side of the game. However, for that to work, Infiltration BETTER BE competitive with any class' dps spec in PvP - and that's not the case. And Balance better be competitive in PvE, and it probably is already outside of maybe 1 or 2 overtuned specs in other classes. I would be entirely happy with that, especially if they get the dual spec feature built in.

I suspect what has actually happened here is that internally they have conceded Kinetic tree as "primarily a PvP spec." And that's probably not unjustified in the sense that the vast majority of the players using the tree in endgame content are probably using it for PvP. If I had to pull a number out of my posterior, I'd probably venture that there are 4 or 5 PvP Kinetic users for every 1 legit endgame PvE tank out there. Just being prudent they pretty much have to treat it as a PvP spec because that's what the player base is using it for. So, they balance it for PvP and let it be subpar elsewhere and lucky for them all the easy PvE content helps hide it being subpar.

The problem is in longterm game health, they only have 3 tank specs in game, and if content actually gets difficult and compelling to play at a high level, they now only have 2.

The eventual solution is probably two things. One, find a way to buff PvE mitigation that translates poorly into PvP. Kinetic Ward is probably a prime target for accomplishing that. Two, is making at least one other Shadow spec more attractive for PvP.

First off, thanks for running the numbers, LagunaD. I did quite a bit of this after the announcement. My numbers came up quite close to yours, but I wasn't as thorough. I think that your analysis is very good, but only insofar as it goes.

12.4% is a scary number. It is not, however, the percentage of boss damage we will take beyond what a Vanguard takes. Rather, it is the percentage of the damage a Vanguard takes that we will take above and beyond their damage numbers. In other words, it's not a very useful value unless you really want to compare strictly the mitigation values of the various tanks. It does not give us any insight into how viable shadows/assassins are as tanks in 1.3.

The number we actually want is the added percentage of boss damage that we will take. This value is fairly straightforward to calculate:

(1-.2744) - (1-.3083) = 0.0339

That is, we take 3.3% more boss damage than an equally geared Vanguard. Still pretty scary, but having this number allows us to factor in the self-heal, something which your original analysis neglected. The value of the self-heal, relative to straight mitigation, is a function of how much damage the boss is doing. Thus, we can compute the requisite boss unmitigated DPS that will be required to exceed the capabilities of the shadow's self-heal by assuming a self-heal of 130 HPS (a number I have both computed and verified in combat logs):

((1-.2744) - (1-.3083))x = 130

x = 3834.81 DPS (pre-mitigation)

In other words, a boss would need to put off raw, unmitigated DPS of about 3835 before the overall survivability of the Vanguard exceeds that of the Shadow. Now, when Kephess is dishing out hits in excess of 17k, it seems like this is a comparatively modest target. However, I have analyzed combat logs to find the actual, unmitigated DPS of several bosses, and I can only think of one (Foreman Crusher) that actually hits harder than 3.8k DPS. Note that Foreman Crusher only achieves such a high average due to his Frenzy ability, which is a predictable periodic phase amenable to cooldowns. Most bosses hit between 1.5 and 2k DPS. (this is only counting weapon/kinetic damage, since those are the survivability numbers we are examining)

There is an argument that the value of the self-heal is less due to mechanical reasons on many bosses. However, my combat logs say that this is really a non-issue. The last time I tanked Kephess, I held a steady HPS of nearly 210. That's only a small bit shy of my maximum HPS (220, tested on combat dummy). The only fight which actually causes my HPS to drop noticeably is LR-5 in Lost Island HM. I think we can all agree that is a weird fight though. Besides, his damage is actually very low once you negate Incinerate.

Just for kicks, let's look at what the threshold of self-heal inadequacy is for shadows pre-1.3 (again using your numbers):

((1-.2888) - (1-.3083))x = 220

x = 11282.1 DPS (pre-mitigation)

I'm pretty sure shadows needed to be nerfed. I'm also pretty sure that these nerfs are just about exactly what the numbers say is necessary. Things will become problematic in upcoming tiers of content, where bosses will (presumably) hit hard enough to cross the 3.8k threshold. Until then, vanguards and shadows appear to be nearly even.

Nearly even, that is, on gear alone. There is a slight subtlety which your original numbers appear to have ignored: relics. Shadows receive dramatically more benefit from the proc absorb relic than vanguards do (due to having a higher shield chance). In fact, if you compute mitigation contribution for BiS relics for both a shadow and a vanguard, the shadow's relics are worth a bit more overall. This further skews the statistical survivability values in favor of shadows. I didn't do the math on this one, because the self-heal alone is enough to justify the nerfs. I just thought it was an interesting point to raise.

/signed. Apparently this game is made only for Guardians and Juggernauts. And NO i am not gonna roll one. Good Riddance

I rolled them and it's why I went to the Sin/Shadow in the first place because there was more reward for the hard work. A Sin/Shadow had a nice balance of DPS and Tanking while a Jugg/Guard could tank just fine - their DPS was lacking. IMO Sin/Shadows Tank specs should be about the base line as to say where all Tanks should meet, not nerf one because there're issues with one of the others; that's the worse thing you could do for "balance" and why so many MMOs get it wrong.

12.4% is a scary number. It is not, however, the percentage of boss damage we will take beyond what a Vanguard takes. Rather, it is the percentage of the damage a Vanguard takes that we will take above and beyond their damage numbers. In other words, it's not a very useful value unless you really want to compare strictly the mitigation values of the various tanks. It does not give us any insight into how viable shadows/assassins are as tanks in 1.3.

 

The number we actually want is the added percentage of boss damage that we will take. This value is fairly straightforward to calculate:

 

(1-.2744) - (1-.3083) = 0.0339

 

That is, we take 3.3% more boss damage than an equally geared Vanguard. ...

Actually at least with the way you are both wording it, his number is correct.

With the way he worked his Q number, if a boss hit repeatedly for 100 damage, a Vanguard would end up taking 27.44 damage on average and the Shadow would take 30.83 damage on average. In that case, the Shadow is taking 12.35% more damage than the Vanguard.

I would also note he seems to be generously giving us 100% uptime on Kinetic Ward charges.

Actually at least with the way you are both wording it, his number is correct.

 

With the way he worked his Q number, if a boss hit repeatedly for 100 damage, a Vanguard would end up taking 27.44 damage on average and the Shadow would take 30.83 damage on average. In that case, the Shadow is taking 12.35% more damage than the Vanguard.

This is true. His 12.4% number is accurate, but it really doesn't help us determine relative viability. All that number does is tell us how much better a vanguard's mitigation is than a shadow's. Great. We already knew that vanguards have better mitigation. The more interesting question is what is a shadow's *survivability* vs a vanguard's survivability.

The way you determine this is by assuming a certain value of unmitigated DPS (represented by 'x') and determine the damage taken by both classes. This gives us a scalar that we can compare to the static value of the self-heal (130 HPS) to see whether or not a vanguard has more overall survivability than a shadow on the same boss. This is precisely what my post computed.

If we look at your example again:

30.83 - 27.44 = 3.39

3.39 is exactly 3.39% of 100, the boss damage number you assumed for your example. Thus, the shadow takes 3.39% of boss damage *in addition* to the damage that a vanguard will take.

So in other words, LagunaD's numbers weren't wrong, he just answered the wrong question with his final computation. A better way for me to phrase my 3.3% number would be "percentage of boss damage that will be mitigated by a vanguard and not mitigated by a shadow".

I would also note he seems to be generously giving us 100% uptime on Kinetic Ward charges.

In my experience, that is not an overly-generous assumption. The only time my KW falls off (assuming I haven't made a mistake) is in trash pulls or extremely rare burst phases on certain boss fights. Trash pulls don't interest me, since being the "best tank for trash" is a rather unimpressive award. As for the burst phases, there are comparatively few of these (e.g. Foreman Crusher's Frenzy ability), and you're generally hitting a CD at that moment anyway, in which case KW doesn't really matter. Boss swing timers are surprisingly long (take a look at your combat logs).

This is true. His 12.4% number is accurate, but it really doesn't help us determine relative viability. All that number does is tell us how much better a vanguard's mitigation is than a shadow's. Great. We already knew that vanguards have better mitigation. The more interesting question is what is a shadow's *survivability* vs a vanguard's survivability.

 

The way you determine this is by assuming a certain value of unmitigated DPS (represented by 'x') and determine the damage taken by both classes. This gives us a scalar that we can compare to the static value of the self-heal (130 HPS) to see whether or not a vanguard has more overall survivability than a shadow on the same boss. This is precisely what my post computed.

 

If we look at your example again:

 

30.83 - 27.44 = 3.39

 

3.39 is exactly 3.39% of 100, the boss damage number you assumed for your example. Thus, the shadow takes 3.39% of boss damage *in addition* to the damage that a vanguard will take.

 

So in other words, LagunaD's numbers weren't wrong, he just answered the wrong question with his final computation. A better way for me to phrase my 3.3% number would be "percentage of boss damage that will be mitigated by a vanguard and not mitigated by a shadow".

Yeah, it's just a matter of wording and what you want to use as a point of reference.

Personally, I don't find full unmitigated boss damage to be a particularly useful point of reference because no one goes out and tanks anything with no mitigation. Content would even be impossible to do that way.

The best point of reference would probably be to some "average tank." How much damage more or less does someone take compared to an average tank? The problem with that is defining what average tank is. For class comparison, we could take all three tanks, average them together, and then call it that...and then see where everyone is compared to that.

It's a lot easier to just compare to another tank directly though, even if you have to do it twice for both of the other tanks.

Why I find that more useful is that the person who cares is the healer, so I would look at it from their perspective. They're thinking, "My Vanguard tank usually takes 11000 twice in 4 seconds here. The Shadow tank takes 12360 in 4 seconds and almost dies. Why is the Shadow tanking again?" So, what they see is the 12.4% difference.

The hope in Shadow design is that they would hopefully heal themselves for at least 2720 to make up that difference. Although, if the tank does die from the damage spike, they won't have a chance to heal it back up.

In my experience, that is not an overly-generous assumption. The only time my KW falls off (assuming I haven't made a mistake) is in trash pulls or extremely rare burst phases on certain boss fights. Trash pulls don't interest me, since being the "best tank for trash" is a rather unimpressive award. As for the burst phases, there are comparatively few of these (e.g. Foreman Crusher's Frenzy ability), and you're generally hitting a CD at that moment anyway, in which case KW doesn't really matter. Boss swing timers are surprisingly long (take a look at your combat logs).

It's a reasonable assumption with a disciplined and knowledgable player.

There are situations, even if just by bad rng, where the charges can be run out no matter what the player does though. So, even if it's unlikey, it will happen at least on rare occasion. Also, even the best player makes errors occasionally. As a tank looking to avoid wipes, there tends to be a lot of emphasis on what happens when rng does go completely sour because after all...you can be perfectly fine for 6 minutes of a fight and then all of a sudden, bam...bad rng string and dead tank -> wipe. It only takes one instance of bad luck to fail, no matter how unlikely. A lot of healers would rather heal more damage on average than have to worry about getting that one surprise damage spike out of nowhere. Clearly we're the most dangerous tank in terms of spike damage - especially after patch - so we pretty much have to be less resource intensive to heal long term in order to be at all competitive.

So, it's not that I'm saying the assumption is unreasonable, I'm just noting that the assumption was made so that people are aware that the numbers can technically be even worse if the charges ever go down.

Personally, I don't find full unmitigated boss damage to be a particularly useful point of reference because no one goes out and tanks anything with no mitigation. Content would even be impossible to do that way.

I'm not saying people are tanking without mitigation. I'm saying that if we look at unmitigated DPS, we can multiply that by the mitigation scalar to achieve the *mitigated* DPS, which is the number that is actually interesting. Comparing post-mitigation DPS between tanks is absolutely useful, since it lets us see exactly how much better or worse our mitigation is in terms of actual damage points. This in turn is absolutely crucial. Without this value, we cannot factor in the value of the self-heal, which is a major part of our survivability.

The best point of reference would probably be to some "average tank." How much damage more or less does someone take compared to an average tank? The problem with that is defining what average tank is. For class comparison, we could take all three tanks, average them together, and then call it that...and then see where everyone is compared to that.

Looking at unmitigated DPS is effectively looking at an "average" tank, since it is easy to take the unmitigated DPS number and adjust it forward to whatever tank you happen to be looking at.

Why I find that more useful is that the person who cares is the healer, so I would look at it from their perspective. They're thinking, "My Vanguard tank usually takes 11000 twice in 4 seconds here. The Shadow tank takes 12360 in 4 seconds and almost dies. Why is the Shadow tanking again?" So, what they see is the 12.4% difference.

No, what the healer sees is the raw HP value. In other words, they don't see a 12.4% difference, they see 3.39% of the boss's damage output difference, *minus* the value of our self-heal. That last bit is crucial. The shadow will get hit harder by the same boss than an equally-geared vanguard. However, unless the boss is consistently landing monstrous hits (well beyond anything in the current PvE endgame), the shadow will come out in the end with more HP.

For example, nightmare Karagga hits for 1.5k unmitigated DPS. Thus, a shadow will take an average of 50.172 DPS beyond what a vanguard will tank (exactly 12.4% more damage than the vanguard). However, a shadow's self-heal is 130 HPS (in 1.3). Thus, the *vanguard* will have a net HP loss of 79.828 relative to the shadow.

So, the healer will actually be asking the question, "why is the vanguard taking more damage? why is he tanking rather than the shadow?"

The hope in Shadow design is that they would hopefully heal themselves for at least 2720 to make up that difference. Although, if the tank does die from the damage spike, they won't have a chance to heal it back up.

This is part of why shadow gear is so endurance heavy. It's also why shadow practical survivability is ideal in the 80% of HP range (90% in 1.3). Healers simply cannot allow shadows to get as low as they can with a vanguard. Depending on your healer, this can result in the *false* conclusion that vanguards are easier to heal. This conclusion is false because the vanguard is actually losing more HP over time, it's just that the shadow's "safety" zone is non-linear, while a vanguard's is perfectly linear from 100% to 0%.

This means that shadows fair much better with burst healers. I absolutely *love* the scoundrel healer who runs with us. He and I have been working together so long that I actually factor his healing rotation into my net "mitigation" on certain fights, while on his end, he is aware of the periodicity of my rotation and when my big self-heal is about to land. I can also do this with our sage, but it's easier with the scoundrel because of how the class is designed.

So, it's not that I'm saying the assumption is unreasonable, I'm just noting that the assumption was made so that people are aware that the numbers can technically be even worse if the charges ever go down.

Makes sense.

For example, nightmare Karagga hits for 1.5k unmitigated DPS. Thus, a shadow will take an average of 50.172 DPS beyond what a vanguard will tank (exactly 12.4% more damage than the vanguard). However, a shadow's self-heal is 130 HPS (in 1.3). Thus, the *vanguard* will have a net HP loss of 79.828 relative to the shadow.

 

So, the healer will actually be asking the question, "why is the vanguard taking more damage? why is he tanking rather than the shadow?"

This is the part that I absolutely get back on the same road with you. It's also where I think the tanks ought to be...if we're a vulnerable to scarier damage spikes, we need to be able to provide healers with a bit of a resource advantage. It's also arguable that some or all of that resource advantage goes away if they have to resort to recovering a health bar with bomb heals a little more often.

The other big thing is that reality is dependant on tanking content being very easy...if it gets harder, the situation flips around and we could end up supplying all liability and no advantage.

For example, anstalt/Svard linked a log earlier of a Jarg & Sorno HM solo tank where the tank took 1.4k dps after mitigation…which would be in the neighborhood of 4.5k Unmitigated. There we would be taking 174 dps more than a Vanguard, which would probably outstrip what our healing can provide after patch. The other dynamic that can play out is if tanks take appreciably more damage in 16 mans, then the healing model looks worse there comparatively.

This is the part that I absolutely get back on the same road with you. It's also where I think the tanks ought to be...if we're a vulnerable to scarier damage spikes, we need to be able to provide healers with a bit of a resource advantage. It's also arguable that some or all of that resource advantage goes away if they have to resort to recovering a health bar with bomb heals a little more often.

Depends on the class. For a Sage, absolutely yes. For a Scoundrel, burst healing is second-nature and resources are renewed in a matter of seconds. Scoundrels *prefer* periodic HP bombs over steady drainage. Commandos are somewhat weirder, and fall between the two. They do, however, have immense burst potential, so I wouldn't see them as having much of an issue with the "periodic bomb" damage profile.

I think the idea of balancing heal/tank combinations by trading off enforced burst healing vs throughput over time is a really interesting one. That may be what BioWare is thinking here, which would be cool.

The other big thing is that reality is dependant on tanking content being very easy...if it gets harder, the situation flips around and we could end up supplying all liability and no advantage.

Truth. As I said in my original post, if a boss hits for more than 3.8k unmitigated DPS, the vanguard will be a better tank in terms of overall HP loss (survivability).

For example, anstalt/Svard linked a log earlier of a Jarg & Sorno HM solo tank where the tank took 1.4k dps after mitigation…which would be in the neighborhood of 4.5k Unmitigated. There we would be taking 174 dps more than a Vanguard, which would probably outstrip what our healing can provide after patch. The other dynamic that can play out is if tanks take appreciably more damage in 16 mans, then the healing model looks worse there comparatively.

You can't really reverse the net "damage taken" number back to a pre-mitigation value. For example, when I tank nightmare mode Foreman Crusher, I have a "damage taken" number of about 1.1k. However, I have actually crawled through the raw logs and done the math to determine that his unmitigated weapon/kinetic DPS is about 4.2k. By comparison, I take about 900 DPS on nightmare Karagga, but his damage is only 1.5k. So, it's all a question of attack+damage types. Jarg and Sorno both do a fair amount of damage, but most of Jarg's damage is elemental, while most of Sorno's damage is energy. Solo tanking them together would absolutely be a lot of unmitigated damage though.

Your point stands that harder-hitting content will overwhelm our self-heal and leave us in a bad place relative to the other two tanks. Right now though, the average pre-mitigation DPS of the hardest-mode bosses in the game is going to sit around 3k. That's right about where it needs to be for shadows to be even with vanguards (assuming slightly imperfect self-heal usage). Once we get the next tier of content, BioWare will need to do something to scale our self-heal more effectively (perhaps a new set bonus).