Catwhoorg's DRR Theory of Survivability V1.0


Bill Z Bubba

 

Posted

Catwhoorg’s DRR theory of Survivability.

Introduction
This guide is written with the most in depth insight in the game from a regen scrappers perspective, but I have played all AT’s and the guide is, I hope, generally applicable.

The DRR stands for Defense – Resistance – Regeneration. Cats theory is best thought of as “Survivability is a stool, with one leg it wobbles, with two it can still be knocked over, with three legs it is as steady as a rock”.

Now it is complicated by the fact that each of the DRR components can be further divided.
A +DEF is functionally the same as –Tohit on the opponent. (The two are additive to each other.)

A +Res, is compounded or replaced by a –dam on the opponent. (The direct replacement is exact, the two combine in a multiplicative manner, not an additive manner)

Innate regeneration can be boosted by heals (By self or others).

Plus/Delta

Now each of these types survivability boost have their own strengths and weaknesses.
Defense
Strength :- Ghetto Status resistance, in that when someone is whiffing with a status attack, that attack does not affect you.
Weakness :- One shotting. If an attack is going to kill you if it connects, it eventually will connect and down you go. It may take a lot longer, or many encounters but eventually the dice will roll against you.
Weakness 2 :- Scaling. Whilst the combination of defense is being addressed (such as the recent FF changes) the scaling of defense is still very non-linear. If a mob has a 50% acc, and you have a 45% def, that mob will hit only 5% of the time. If the mob has a 65% acc, and the same 45% it will now hit 20% of the time. Minion to boss level accuracy increases the hits four fold, and those hits from the Boss will hurt much more.

* Note Stargazer has done some wonderful testing which shows that there are still stacking issues with +Def from different sources.

Resistance
Strength :-True one shot protection.
Weakness :- Internal scaling issue, a little bit of resistance is not worth as much (Damage taken = 100-resistace) (Going from 10->20% is worth much less than 80-90% resistance)

Regen/Healing
Strength :- Low down time.
Weakness :- One shotting, you cannot regen if you are defeated.


Synergy

This is where the stool component comes in, what you try to do as a team, or as an individual, is to cover a minimum of two of DRR, and ideally you want to cover all three. Want to turn a regen scrapper into a I2 God- like one, throw him/her fortitude (+ DEF and nice other effects), and a sonic bubble or two (+Res). What hits through the + def is weakened by the +res and then regened back quickly.
Status effects will be reduced (though the +def) and one shot resistance improved (+res)

Breaking down the pairs
Defense/resistance – a great combination of being hit less, and when hit hurt less. The risk here is attrition, with nothing to help you , it is easy to be worn down in a long fight. Green Inspirations covers this gap very nicely. INV tanks (and scrappers) are actually in this category once they get Invincibility.

Defense/regeneration – an awesome combination (Cat usually has fortitude placed on him). This combination reduces the alpha risk, and the one shot risk, but does not eliminate it. (that 5% chance of being hit is always there). This is easily replicated solo, by the use of purple inspirations and a regen /self heal. Many /SR scrappers take the medicine pool for the self-heal, which effectively gives them this combination.

Resistance/regeneration – if you have to take a pair this is the one to get. The smooth decline in HP afforded by significant resistance, meshes so well with the smooth replacement of HP by regen or healing. The resistance grants the one shot protection that is regens weakness.

It is interesting to point out that since the last patch everyone has the capability to address the missing 3rd leg of the stool if they have two of them through the use of inspirations. The sturdies are the hardest to get a significant effect from, but it is feasible with the larger versions.

Effective Regen rates.

Now this is where the real regen bias comes in. It is easiest (for me anyway) to compare different mixes of DRR, by reducing everything down to an effective regen rate multiplier.

For defense (Muliplier = ACC/(ACC-DEF))
Assuming a 50% accuracy (even con minions)
5% = 1.11 x regen rate (50/45)
10% = 1.25
20% = 1.67
25% = 2.00 (one small luck)
30% = 2.50
40% def = 5.00
45% def = 10.0 (To hit floor reached)

Here is a graph which shows the way this multiplier scales with opponent accuracy. It is interesting to note, that given a fixed defense the higher accuracy will push the scaling down, the higher accuracies can actually be penalized MORE by boosted defense than by damage resistance.

For Resistance (Muliplier = 100/(100-resistance))
5% = 1.05 (one small Sturdy)
10% = 1.11
20% = 1.25
40% = 1.67
50% = 2.00
60% = 2.50
70% = 3.33
75% = 4.00 (Most cap here)
80% = 5.00
85% = 6.67 (Dwarf form cap)
90% = 10.00 (Tanker cap)

See how for even level minions the defense % is worth twice the resistance %

So to get you effective regen rate multiplier, you take the defense component, and multiply by the resistance component.
Regen rates of typical powers
Health = +40% (+120% 6 slotted with SO)
Fast healing = +75% (+225% 6 slotted with SO)
Integration 100enh +50% Unenh ( 350% 6 slotted)

Here is the response for a even minion (50% accuracy)

Here is the response for an Even Boss (65% accuracy)

Here is the response for a theoretical (90% accuracy)

All the above three are at the same scaling you will notice. See the how the synergy really kicks in with the combination of high def and high resistance.

To get a net regen you take the above multiplier and apply it to your regen rate.

So at a regen rate of 100% (baseline) and you have 50% resistance and pop 2 lucks versus even minions.
10(d) x 2(r) x 100%(r) = 2000% Effective Regen rate

A stone tank with 6 slotted rooted (+300% regen) Health (+120% regen) and granite armour 6 slotted for defense (60% resistance, 44% defense) versus an even con minion and then that theoretical 90% accuracy foe

Even minion 8.33(d) x 2.5(r) x 520%(r) = 10829% effective regen rate.
90% accuracy foe 1.96 (d) x 2.5(r) x 520%(r) = 2548%. Note the two last terms haven’t changed, that drop is entirely due to the non-scaling nature of the defense.

So lets see how Cat lines up. He is running with fortitude cast (5 Def slots = 60% defense) against +2 minions (60% base acc IIRC). He has INT and FH 6 slotted. (+575% regen rate)
12.0(d) x 1.0(r) x 675%(r) = 6900% effective regen (the above stone tank has 4875% in this case)
Of course the problem is that we haven’t accounted for a) Dull Pain (multiplies HP by 1.4, and hence regen rates, and reconstruction with Cats slotting 50% of his HP every 30sec)

His real effective regen rate is actually closer to 12.0(d) x 1.0(r) x 1345%(r) = 16140%, however there is exactly the same risk of one-shotting as before.

The same logic can be applied to heals. A basic 5heal/1end slotted heal other heals for ~69% of a defender HP. Put a well slotted fortitude on that defender, and it suddenly has a value of 690% (even con minions). This is, a little nonsensical at first, unless you think about what that number actually means . Because the incoming damage has been reduced 10 fold, one heal other repairs now, what 10 uses would have been needed to do before.

Throw fortitude, regen aura (+500%) and AB (+500%) on someone against those same even minions, 12 x 1 x 1100 = 13200%. Spam 10 heal others in a minute and that adds in another 6900 taking you over the 20 000% effective regen rate.

Ever wonder why empaths despise the RU healer tag ? With fortitude and regen aura they are throwing out +Def like a /SR scrapper and boost regen rate to near (or heck exceed at times) a /regen scrapper. Do that right and healing becomes very much an afterthought.

Rads and darks are pretty cool, because they cover all three legs to a greater or lesser extent. – acc, -dam and a heal available.

Cautionary note:

Effective regen rate is at best a concept, a useful way of saying would I be better off with this combination than that combination. THERE IS NO ACCOUNTING FOR THE ONE SHOT. It is also a maximum figure, as regen or heals are rarely used at full effect all the time.



@Catwhoorg "Rule of Three - Finale" Arc# 1984
@Mr Falkland Islands"A Nation Goes Rogue" Arc# 2369 "Toasters and Pop Tarts" Arc#116617

 

Posted

I have been trying to summmarize my almost identical thoughts on what I call the Layered Defense Doctrine. If you only have one layer of defense, your enemies only have one obstacle to overcome. Extra layers are always beneficial. The one thing that keeps popping up is that as far as scrappers and tankers are concerned most sets get two layers within their defensive powerset. Ice and Super Reflexes don't. All things being equal, pure resistance is less effective than a mix of resistance and defense. Pure defense is worse than anything else.

Those are all in regard to what I call passive defense though. Active defense is categorized in Control (mezzed enemies cannot attack, therefore deal no damage) The Slaughterhouse Approach (kill them before they have a chance to fire off a shot) and Debuffing (rather than increase your resistance/defense just decrease their accuracy/damage output). It is a lot harder to quantify active defenses except to say that debuffing enemy damage resistance for a more effective slaughterhouse approach seems to be better than most other aspects.


 

Posted

Nice. Thanks for the work, Catwhoorg


Be well, people of CoH.

 

Posted

Excellent job, Cat! Great analysis with the numbers to back it up. The one thing that might make your analysis more effecive is to see how this DDR concept would apply to various builds, possibly using a DDR scale to reflect overall defense.

One more factor that should be considered but would be very difficult to quantify: Pets. Many pets, by drawing aggro, substantially increase "defense" by reducing the probability of attacks on the player. Some pets also debuff. Illusion Controllers have a unique power in Phantom Army. Is there anything else that can completely ignore the "one-shot"? What about the effect of Deceive/Confuse?

(It probably goes beyond the scope of your anaysis, but I had to throw in something that it appears you did not think of.)


LOCAL MAN! The most famous hero of all. There are more newspaper stories about me than anyone else. "Local Man wins Medal of Honor." "Local Man opens Animal Shelter." "Local Man Charged with..." (Um, forget about that one.)
Guide Links: Earth/Rad Guide, Illusion/Rad Guide, Electric Control

 

Posted

Thanks for the comments folks.

One PM'ed comment I got as well was the scaling factor - Basically flooring Def against a 90% to hit opponent was worth more then against the 50% so the number was artefically boosted.

But after examinination you know - it is worth more. Floooring accuracy of a 90% to hit opponent is worth more than that of a 50%.
Its my opinion that the numbers should reflect this.



@Catwhoorg "Rule of Three - Finale" Arc# 1984
@Mr Falkland Islands"A Nation Goes Rogue" Arc# 2369 "Toasters and Pop Tarts" Arc#116617

 

Posted

So, what do the graphs look like now, with ED & CoV launch & what not?


 

Posted

The grpahs are fundamentally the same, its just harder to move up the axes.

The examples in the text are wrong and will be reworked, in the age old time of soon.

Im thinking November, but no guaruntee - I want to adress the scaling in the same update and haven't gotten that one cracked.



@Catwhoorg "Rule of Three - Finale" Arc# 1984
@Mr Falkland Islands"A Nation Goes Rogue" Arc# 2369 "Toasters and Pop Tarts" Arc#116617

 

Posted

It'll be interesting to see how the formulae's apply when they fix Defense Scaling. If I understood what the dev's were saying, an even con Boss's tohit will now be calc'd as (50 - Defense) * (1 + ToHitBonus). Where the tohitbonus is 15%. (currently 65 tohit against a 0 def target).

So, that'd be ( 50 - 0 ) * ( 1 + 0.15 ) = 65. Whereas say, a 35% defense (easily attainable for some sets), would make that ( 50 - 35 ) * ( 1 + 0.15 ) = 17.25% ToHit (Under the current system, that boss would normally have a 30% tohit, 65-35=30).

Which is rather interesting I think. especially with your DRR theory. Taking a 65% ToHit down to 17.25% is a multiplier of 3.77. The dev's say that 1% Defense = 2% Resist. So, compare that 3.77 multiplier to the multiplier for 70% resist, which is 3.33. I like this upcoming change to defense scaling, while it won't help with One-Shotting, it'll protect more against higher level mobs than resist (excepting one-shots). Anyways, love your post and the formulae you came up with to support your theory.


Damage Proc Mini-FAQ

Just noticed Damage Proc Mini-FAQ wasn't working with new forums, it's been updated.

 

Posted

Yeah the defense scaling chnages things.

I'm re-working the plots right now, in anticipation of I7 hitting test.



@Catwhoorg "Rule of Three - Finale" Arc# 1984
@Mr Falkland Islands"A Nation Goes Rogue" Arc# 2369 "Toasters and Pop Tarts" Arc#116617

 

Posted

So, what's the status on this?
Did it show under a different title, or just not happen because I7 and later changes have changed things completely?


 

Posted

Its in progress - Have the I7 version plost done and the text mostly re-written. Would have finished it last night but for the CoP trial activation.

With all the alt SG's Im in, it is going to be a busy week in game



@Catwhoorg "Rule of Three - Finale" Arc# 1984
@Mr Falkland Islands"A Nation Goes Rogue" Arc# 2369 "Toasters and Pop Tarts" Arc#116617