I24: Switch Boxing or Kick with Tough

If you mean technical reasons, there's no technical reason it can't be made an inherent for all players. If you mean non-technical reasons, there's tons of those.

Because Boxing and Kick would still be skipped?

Hmmm...seeing as how all my concepts are based around crime fighters who thusly train to be better fighters and have some sort of Hand to Hand experience...I'd say Fighting Pool is one of the EASIEST pools to fit into a concept.

The problem with the Fighting Pool is that the Tough/Weave percentages should of always been higher for the squishies and less for the melees.

Instead of going "Oh, this AT should have high modifiers in this area" the pool powers should go the better route of...

Fighting Pool "More benefit for the squishies as the melees are already tougher...so while they're peaking out, the squishies are building up from nothing so getting more out of it"

Leadership Pool should of always had the same stats across the board. Whether that means upping everyone to the best Leadership numbers, lowering them to the weakest Leadership numbers, or meeting somewhere in the middle.

No reason one AT should be a better "Leader" than another.

You're probably only looking at scrappers or blasters.

For Tanks:
Maneuvers: 2.275% Defense
Weave: 5% Defense

Wow, holier-than-thou-art much? What right do you have to judge people for enjoying the numbers behind the game versus playing the game however you enjoy it?

I24 Proc Chance = (Enhanced Recharge + Activation Time) * (Current PPM * 1.25) / 60*(1 + .75*(.15*Radius - 0.011*Radius*(360-Arc)/30))
Single Target Radius = 0. AoE Non-Cone Arc = 360.

In D&D circles, this is known as the Stormwind Fallacy.

That said, if we allow your premise that most people do take it soley to min/max, and got rid of it entirely, wouldn't the min/maxers just then find something else to take soley to min/max? Wouldn't that eventually lead to getting rid of just about every power in the game?

Um... What? What are you referring to by "now?" And how is my SR Scrapper as effective against Incarnate critters as a SO SR Scrapper is against non-Incarnate critters? My SR Scrapper IS using Singe Origin enhancements, or at most Common Inventions (I'd have to look). Sticking to just that simpler way of building was something I held very dearly, but it's clear to me now that the game is balanced around the "average" player, and SOs and Commons are apparently significantly below average.

I don't follow what you're saying.

Originally Posted by Arcanaville Samuel_Tow is the only poster that makes me want to punch him in the head more often when I'm agreeing with him than when I'm disagreeing with him.

I realise that's probably the case and my terminology was inconsistent, but I was under the impression you got what I was referring to - that the final critter to-hit before accuracy remains the same between having 30% defence against common enemies and 16% defence vs. Incarnate enemies. I'm honestly not sure what your "normalised" mitigation approach is. I remember asking for a specific formula and an explanation of why it was chosen a few times over the years, but for the life of me I can't remember one or an explanation in general.

So I think you misread the question. Could you explain to me how you reached your "mitigation" number step by step so I can see what I'm missing and where our terminology differs?

OK, I could work with that. In what cases is that not true, and can it be true when considering just one character's own mitigation independent of any others?

Originally Posted by Arcanaville Samuel_Tow is the only poster that makes me want to punch him in the head more often when I'm agreeing with him than when I'm disagreeing with him.

The standard mitigation definition - which also matches what people think mitigation generally means - is the percent of attacks or damage that are nullified relative to what would have hit otherwise. So we don't normally say everyone starts off with 50% mitigation, because those attacks wouldn't have hit anyway. And if we say that, we then have to say everyone starts off with 35% mitigation against even Bosses. And 25% against even AVs. And so on. And it also means 50% resistance becomes 75% mitigation against minions.

When I say "the standard matches people's expectations" people sometimes believe that mitigation should be calculated in one way or another, but the rosetta stone is "what is 50% resistance?" Almost everyone says that's 50% mitigation, and from there the standard follows. 25% defense generates the same average mitigation, so 25% defense must also be 50% mitigation. It must be 50% mitigation vs minions, bosses, and AVs, because it generates the same average result as 50% resistance verses all those enemies.

That basic principle then translates into math from first principles. Mitigation is the number of attacks that miss divided by the number that would have hit, because that's what the expectation above is equivalent to. For even minions, the number that miss is your defense, and the amount that would have hit is minion base tohit, so mitigation = defense/basetohit = 25%/50% = 0.5, or 50%.

This extends to things with accuracy. The amount of hits a boss will normally land is 1.3 * 50% = 65% - Boss intrinsic accuracy is +30%. The actual amount of hits that will land (on average) is 1.3 * (50 -25) = 32.5. So the amount of misses generated is 65-32.5 = 32.5. We then have mitigation = 32.5/65 = 0.5, or again 50%.

The base number of hits (with no defense) will generally be Accuracy * Basetohit. The number of actual hits will be Accuracy * (Basetohit - Defense). The number of misses will be Accuracy * Basetohit - Accuracy * (Basetohit - Defense). That reduces to Accuracy * (Basetohit - (Basetohit - Defense)) = Accuracy * (Defense).

Thus, mitigation = Misses/BaseHits = (Accuracy * Defense) / (Accuracy * Basetohit) = Defense/Basetohit.

QED

Against incarnate critters, base tohit is not 50% its 64%. An SR scrapper with 30% defense is in some sense behaving like they have 14% less defense - 16%. But that's relative to a base 50% world. Its slightly misleading in that most people believe that's comparable to 28% resistance, but against Incarnates that's not true: 30% defense is comparable to 46.9% resistance, because its equivalent to 46.9% damage mitigation.

So 30% defense is like 16% defense when facing incarnate critters, but not like 32% resistance. Probably most people would make that intuitive but incorrect leap, because in any situation where base tohit is not 50%, people's intuition about the relationship between defense and resistance will be wrong.


If you're talking about how much harder something is to a defense set outside of comparisons to other things, its usually fine. If you're talking about a disadvantage relative to other things, its sometimes misleading. Mathematically correct so long as you do not generalize, but probably misleading.

I think this is the part where our definitions sort of part ways and where I'm not entirely certain what you're doing. To me, basic probability would imply that if your chance to be hit, i.e. the statistically average but implausible amount of times you'll be hit, is Accuracy*(BaseToHit - Defence), then the chance to be missed will be 1 - (Accuracy*(BaseToHit - Defence)), or basically the chance that anything BUT the former will happen. We're certain that the chance to be hit will never exceed 1 (100%) because final to-hit is bound between 0.05 and 0.95 in every case. I guess if you're trying to nullify the impact of enemies having a base to-hit of 50%, meaning they'll miss even completely unprotected characters I could maybe see it, but I just don't follow the WHY of the formula.

Personally, I have no problem accepting that every character, by virtue of existing, already has 50% defence, with defence sets adding more to it. It makes defence much easier to interpret when you consider it together with base to-hit, because how its impact changes is much easier to track, at least to me. I recall us actually having a similar discussion once upon a time, and I also recall insisting on using a "variant" of mitigation that basically measured survival time, after taking account of regeneration, resistance, defence and max hit points. I even had a formula for calculating this, but my subscription to Mathematica ran out since the university doesn't see it as necessary. What I'm saying is I don't mean to "disprove" your formula, I just don't "get" it. Maybe it's just me, but a formula which tries to be more intuitive by obfuscating how it actually works is actually less intuitive to me because I'm always much more comfortable if I "get" the basic workings of a system without needing them to be translated for me. And City of Heroes isn't so complex as to be impossible to "get."

That's probably my own personal kinks, but "mathematically correct" is what I usually shoot for and I tend to not draw wider comparisons or quote numbers unless I have the research to back it up. I actually considered running a few basic calculations to see how much extra resistance one would need to attain starting from nothing to go back to where you were in the regular game, vs. the 14% defence SR has lopped off its stats. I wanted to say double that, but it's not actually true off a base of 64%, and I also know that how much a change in defence helps depends on what you change from and what you change to. Trouble is, though I used to have a very good grasp of this just in terms of mental math, I haven't worked with the numbers in so long I've forgotten and would have to spend some time re-familiarising myself with those.

That said, I'd still rather work with the raw numbers than with representations of them, and I'm just more comfortable comparing statistics that represent something actually measurable in-game, as opposed to a representative comparison stat. Again, I'm not saying you're wrong, just that what's instinctively more comfortable for me may just be weird.

Originally Posted by Arcanaville Samuel_Tow is the only poster that makes me want to punch him in the head more often when I'm agreeing with him than when I'm disagreeing with him.

You can't mitigate what would have never hit in the first place.

Here's a different angle at the problem. Suppose you have 90% resistance. Virtually everyone calls that 90% damage mitigation. But lets say you also have soft-capped defenses. So, for even minions, only 5% of all attacks get through. But that means your 90% resistance is only reducing that 5% down to 0.5%, cutting 4.5% of the original damage.

Is 90% resistance only offering 4.5% damage mitigation in this case? What does that even mean?

What we normally say is that 5% of all attacks *land*, and 90% resistance mitigates 90% of what lands. So its legitimately 90% mitigation. If you flip that resistance off and on, while under soft-capped defenses, you'll see incoming damage drop by 90% of what it is when that resistance is off. Its that relative difference that is the basis of how we judge mitigation for resistance, and essentially no one argues with that.

Defense must be judged in the same way, or it becomes impossible to compare the two together in a consistent fashion.

Yes, that's exactly what it means. I don't see why it's a problem. Piling 90% resistance on top of 5% final to-hit reduces that 5% by 90% points of itself, down to 0.5%, exactly as you explain below. I don't see why that would be as confusing as I infer you're presenting it as. That's how defence as a probability and resistance as a percentage decrease work. To me, THAT'S intuitive.

Here's what I don't get - why is it supposed to mean anything more than what it already means?

I think your example is missing a crucial element - what happens if you're not measuring that resistance from 0% all the way up to 90%, and instead measure what would happen if you piled 40% resistance on top of 50% resistance? What if you piled 60% on top of 30%? What if all players who enter Dark Astoria suddenly took a resistance debuff of 28% that ignores buffs and enhancements?

I guess what I'm missing here is a definition of what you mean by "mitigation." I've seen the formula, but I don't quite follow why you're using the numbers you are.

In trying to understand what you mean by reading and re-reading your previous post, I keep coming back to:

I don't get what this is saying. How are you counting misses, when you're calculating with percent and what you're getting is, by definition, also a percentage? How do you transfer probability into misses without going through some specific example, like say "this many misses out of 100 attacks?" Because, again - you seem to be trying to work towards some kind of general "mitigation" metric without taking all stats into account, but you CAN'T include regeneration into this without taking a specific situation. I say this because regeneration is additive to damage, rather than being multiplicative, so how effective regeneration is as opposed to resistance and defence should take into account how many attacks are made, how much damage they do and how much hit points the character has both as a basis for regeneration and as a pool for it to work off of.

What I'm saying is, if you want general mitigation, then you can't afford to compare "misses" one time and DPS another. To me, in order for this to be truly generic, it must all go through the same basic stat, which is how well a given level of protection protects a given health pool from being depleted. Because, ultimately, that's where the rubber meets the road, isn't it? It's why I don't get this counting of hits and misses and comparing them to each other.

I guess what I don't get above all else is this:

Why would you ever, under any circumstances, multiply accuracy by JUST defence when no formula would ever allow this? Every example of accuracy always includes base to-hit in brackets so I genuinely can't see how it would slip out of there. Are you saying that's (Accuracy*(Base - (Base - Defence))/(Accuracy*Base)? Because this genuinely makes no sense to me.

Why makes no sense, if I want to be specific, is why we're counting misses when that's not what probability gives us. If we want to compare base to base and defence, wouldn't it make more sense to divide the former by the latter and get a ratio? Because that way, I can say that going from 40% defence to 45% defence cut my chance to be hit in half, thus it very likely doubled the time I can survive. JUST comparing misses and going with the above formula would suggest I have mitigation levels of 80 vs. 90, suggesting I went up only "10 mitigation," which... Doesn't ring true to me.

I guess if your point is to only ever compare protection against the base to-hit I could maybe see that working, but I've honestly never been in a position where this has come up, which is to say I've never been in a situation where I've gone from base to high levels of defence and had to consider the difference. Maybe it would be useful for comparing sets to each other, that I can't support or deny, but what it doesn't help me with is exactly what I brought up to begin with - what happens to my survivability as my final defence numbers fluctuate. What happens to me when my defence levels shoot up via Inventions? What happens if I drop to Premium and have to give them up? How does this relate to a varying base to-hit?

Basically what you're doing is taking the interval between "base to-hit" and 1.0 and stretching it into the interval of 0 to 1.0, which I just don't see being as meaningful. As you stretch the interval, you smooth over the sharp increase in survivability as both defence and resistance shoot up. You've said it yourself, and argued with me on this point, no less, that 5% defence can mean the world or it can mean nothing depending on what level of defence you pile it on top of. The way you've re-stretched the interval, that 5% defence adds up to 5% mitigation regardless, and that doesn't ring true to me. You've basically taken defence and stretched it to always represent what chunk it has of the base-to-cap interval and presented it as a linear progression.

Let me be scientific for a moment. In the formula of defence/to-hit, to-hit is a constant, or at the very least a parameter. Your variable here is defence, and what you've constructed is a linear function, with "to-hit" serving as nothing more than to modulate or translate the value of defence into something which corresponds to the chunk it covers from the "base-to-cap" interval. You may have equalised the numbers, but in so doing I believe you've obscured their actual impact. Why? Simple example:

A 1000 hit point character running at base to-hit and taking a 100 point attack every second will survive for 20 seconds. The same character with another 5% defence will, instead, survive ~22.22 seconds, or a little over 2 seconds more. A character with 40% defence will survive for 100 seconds, whereas a character with 45% defence will survive for 200 - an increase of 100 seconds. Both additively and multiplicatively, the difference is not linear. That's the source of my problem.

I guess I don't get exactly what, physically speaking, is it that "mitigation" is measuring and how can I use it to predict my own survivability in a practical situation? Because from what I can see, we're comparing stats, rather than their actual impact on performance. I'm probably wrong and not seeing something, but that's what I'm seeing. The difference between 40% defence and 45% defence should not be "10 mitigation," or else the difference between 0% defence and 5% defence should be something different than "10 mitigation," because the jump in actual survival isn't equal between the two.

*edit*
This, I think, is where I disagree. I don't believe you need to judge all stats the same way, because all stats don't work the same way. What we should be judging stats by is their impact in practice, or at least in some kind of "clean" theoretical testing environment. I know that the game's environment is a bit too complex and varied to really predict, but at the very least, we can say that "you can survive a long time if you meet about this much DPS." Just something which grounds the math in reality, is all.

Originally Posted by Arcanaville Samuel_Tow is the only poster that makes me want to punch him in the head more often when I'm agreeing with him than when I'm disagreeing with him.

Well the thing is Resistance and Defense basically do the same thing: help keep you alive by reducing the damage that you take. As such a common metric for judging their effectiveness makes a LOT of sense. Being able to say that a power provides X mitigation and another provides Y is a lot more useful if X and Y share a common definition.

In the same way that when looking at Healing, Regeneration and Absorb we tend to consider all of them in terms of HP/Sec. We acknowledge that they are radically different in implementation but HP/Sec is a good starting point for comparison.

Mitigation doesn't determine survivability. Its inverse, what I call Admittance, does (that one hasn't caught on though). No level of damage avoidance equals survivability. What hits you is either enough to kill you or not, regardless of how much is mitigated.

We compare defensive powers by how strong they are, which is a measure of how much they can mitigate. But that's a relative value, useful to compare mitigators to each other. We actually measure survivability by calculating admittance, whether anyone calls it that or not. Its 1-Mitigation, but then translated into absolute (not percentage) values. Its often done in reverse, by calculating the maximum survivable damage. per window of time.

Its grounding in reality that is really a flaw in your version of mitigation. By attempting to bend it to serve the function Admittance really serves, it completely breaks any sense of connection to reality. My definition of mitigation is an operative definition. Which is to say, its defined based on how we would measure it. We measure the mitigation of X by flipping X off, measuring how much damage we take, flipping X on, measuring how much damage we take, and subtracting. Or, because its more convenient in the long run, dividing to get a percentage.

Human beings are not good at observing small changes in things accurately, but when the changes are high enough they can judge gross changes in survivability. If you have 40% defense, a 5% defense power will halve the incoming damage you take; conversely it will double the damage you can survive. Players can notice that. On a relative basis, 5% defense is offering an incremental 50% increase in damage mitigation, which exactly matches players direct in-game experience. They know its *not* "only" taking out another five hits out of a hundred.

Its five out of a hundred theoretical attacks. Its five out of ten actual hits. Both mathematically true, but only one of them is something anyone cares about.

Players cannot observe what doesn't happen. They do not observe that that power blocks 5 more attacks out of 100, because they cannot observe the 100. They observe that they used to get hit by 10, and now they get hit by 5. That's half, and that's the value of that power in that circumstance.

The circumstance by which we judge totals is relative to starting from zero: zero defense, zero resistance. We compare to that standard, because players do not judge a power on the basis of what it does absolutely, but what it does relatively. What benefit does it convey in reality, right now, relative to what I start with. All other definitions of mitigation are academic curiosities, but don't match players experience.

The fact that the "standard model" of mitigation is essentially the *only* one that is used when mitigation is discussed publicly is a testement to the fact that it as demonstrated over the years to match people's in-game expectations, even when for various reasons it doesn't always match people's calculation expectations.


Also I should point out that while I say this is "my" system, that's because its always been the way I've done it, going back to when nobody was doing it. But its not my system in the sense I'm its lone proponent or even its singular defender. The system doesn't really need defending: its the system: the system everyone uses now, the system that has been reinvented many times over by other players, the *only* system that has proven to be effective at functioning as the foundation for productive mitigation discussions. People do occasionally come up with their own proprietary systems, but there is no minority circle of people using another system incompatible with the standard one anywhere I'm aware of besides within their own heads. Lots of other systems have been promoted as the "better" one, but they all eventually get discarded because no one can find a practical use for them.

I'm not trying to argue if it's a good or a bad system, so quoting user numbers doesn't really help. What I'm trying to do is understand it, because you quote this supposedly much easier to understand system and the numbers that come out of it don't mean anything to me and, worse, seem to contradict what I'm pretty sure is true in the way I see my own defences and resistances. I just think your mitigation metric is simply saying something very different from what I thought it was saying, and is used for something significantly different from what I'd assumed it was used for.

I'm glad other people are using it, but other people are using Facebook on their iPhones and I don't "get" either of those, either. What I'm saying is just because I don't get it, it doesn't mean it's bad, just that I probably can't make use of it.

Originally Posted by Arcanaville Samuel_Tow is the only poster that makes me want to punch him in the head more often when I'm agreeing with him than when I'm disagreeing with him.

What does it already mean? It's mitigating 4.5%, but 4.5% of what? What does that tell me? 4.5% isn't very much; does that mean 90% resistance isn't much better than no resistance? And then it's 5.85% against a boss, and 8.19% against a +4 boss, and 45% against the same +0 minion if I had no defense, and so on, so is it variable and unreliable? Any of these numbers is useless (because you have to know everything else about the situation to put the number in context), and also misleading (because it doesn't directly tell you what the resistance is doing).

The problem is that looking at it this way doesn't give you usable information. We figure out what the power does by looking at what the power does: 90% resistance prevents 90% of the damage you would take. It makes you ten times tougher. You don't need to know anything else about how much defense you have, or the rank or level of the enemy attacking you or what attack they're using. Going from zero resistance to 90% resistance makes you ten times tougher, period, the end. What if I also add enough defense on top of that to prevent half of the damage I would have taken? Then I'm twice as tough as ten times as tough as I was before. That's twenty times as tough. I don't really care that half of the incoming attacks were missing already - those attacks were missing anyway without my powers, so I can't attribute them as being mitigated by my powers. Nor do I care that bosses were already missing 35% of the time instead of half the time - my resistance still prevents 9/10 of their damage and my defense still prevents half of their hits and I still take 1/20 of the damage I would have taken with no defense and no resistance.
This part, btw, is because mitigation cares only about the damage prevented. Mitigation measures the difference. The hit chance formula, expressed simply and uncluttered by all the clamp() and whatever that isn't relevant here, is:
Accuracy*(Tohit - Defense)
And we can distribute that through:
Accuracy*Tohit - Accuracy*Defense
How many attacks were hitting with zero defense? That's the same formula again, just with 0 plugged in for defense:
Accuracy*(Tohit - 0) = Accuracy*Tohit

So what is being mitigated by the defense - what is the difference between the chance to be hit, with and without defense? That's just subtraction:
[Accuracy*Tohit] - [Accuracy*Tohit - Accuracy*Defense]
Distribute through: Accuracy*Tohit - Accuracy*Tohit + Accuracy*Defense
The first two terms subtract to zero: 0 + Accuracy*Defense
And then we're left with just the Accuracy*Defense term.

Well, here's the thing: The system Arcanaville is describing is the only one that gives you the right answer to the question "what impact will this ability have on how long I survive in combat?".

And that, it turns out, is the only question we care about for defense and resistance.

The goal of evaluating powers which are intended to increase survivability is to see how much they increase survivability.

Hmm. Okay, lemme try a thought experiment. Let's imagine that there is a vast simplification of tax code, along the lines of the famed "New simplified IRS form" some comedian joked about:

How much did you make last year? ______
Send it in.

Imagine that the tax code is simplified such that you have your salary, and then you send in some percentage of it.

So when it starts out, you're making $100K (a nice round number), and The Tax Rate is 10%, so you get $90K. Obviously, if the tax rate increases by 1%, you lose $1K. So if it were 15%, you would get $85K, and so on.

But like most people, you don't actually care nearly as much about either your salary or the tax rate as you do about your take-home pay. That's what you have to match up against your mortgage and stuff. So if you're evaluating a possible tax increase, you aren't going to just say "well, 5% more means I have 5% less money", because it's not 5% of your money, it's 5% of the theoretical "before-taxes" value, which isn't what you're spending from.

So say the tax rate is 10%, and they decide that they need to fund the Federal Basketweaving Administration, and they increase it by 10%. The tax rate has increased by 10%. The tax rate has doubled. And the effect on you is that you lose 1/9th of the money you were taking home.

On the other hand, say that there are a few special categories, and since you're in the much-maligned group "people who have sometimes disagreed with Arcanaville on the forums", your tax rate is 80%. And then taxes are increased by 10%, so it goes to 90%. The tax rate for you has only increased by 1/8th (12.5%). But! The amount of money you get has halved.

So even though the absolute increase (10%) is the same, and the increase relative to what it already was (12.5%) is lower for you than it was for the people who were paying 10% before... The effect on you is much larger. It's half your income!

Damage works the same way. No one cares what the attacker would be doing if they hit 100% of the time and there were no resistance, except as a number to shove into equations. What we care about is the damage we are taking with or without a given power.

So say that I can currently, with regen floored (so we can ignore regen, and just count how long it takes me to take my health in damage), stand up to a given mob which deals only S/L damage for an average of a minute, and I get Tough. How long will I last now? We have no idea at all. Because we don't know what my resistance was before.

If I had no S/L resistance before, 15% resistance will reduce the damage I take by 15%, and I should last roughly 15% longer.

If I had 80% S/L resistance before, and I'm a tanker or brute, 15% resistance will increase my resistance to 90% (due to the cap). So it's reducing the damage I take from "20% of theoretical" to "10% of theoretical". Which means I'll take about half as much damage, which means I'll live twice as long. If the cap were 95%, I'd live four times as long.

And "how long would I live even without regen" is a pretty useful comparison to make!

So basically, the most useful thing to measure is how much damage you're taking ("admittance"). But powers all give mitigation. And converting between mitigation and admittance doesn't require much math. The trick is that, to accurately evaluate the benefit to you of a power, you have to do that conversion twice: Once on your mitigation without the power, once on your mitigation with the power. The compare the results, and you have the actual change in your experience.

So, say I'm building a fire tank because fire farms are simply THE most exciting content the game has to offer.

I start out with a 22.5% fire resistance power. How much does this help?

Without the power: I have no mitigation. I take 100% of incoming damage.
With the power: I have 22.5% mitigation. I take 77.5% of incoming damage.

Now, what happens when I add another 22.5% fire resistance power?

With only the first power: I have 22.5% mitigation. I take 77.5% of incoming damage.
With both powers: I have 45% mitigation. I take 55% of incoming damage.

And the meaningful change to me is: 55%/77.5% ~= .71, or 71%. I will take 71% as much damage with two powers as I did with one. So if with one power on, I was taking 100 damage per second of combat, with two on I will take about 71 damage per second of combat.

Third 22.5% power? I go from 45% to 67.5% mitigation, leaving me with 32.5% of incoming damage. If I was taking 100 damage per second with two powers, I'd be taking about 59 per second with three.

Now I go get me some IOs, and add about 30% to each power. That increases my mitigation to 87.75%. If I was taking 100 damage per second with all three powers unenhanced, I'll be taking about 38 damage per second now that I've I/Od them. If one of the sets has a 2.25% fire resistance buff, I'm up to 90%, and I'd be taking only 30 damage per second from that same input.

You'll note I've ignored defense completely. That's because, no matter what defense is doing, it's multiplying both sides of the comparison by the same amount. Similarly, if I'm evaluating defense, I ignore resistance, because whatever defense does, resistance is multiplying both sets by the same amount.

So how do I compare defense and resistance, then?

Simple! Do one comparison in which I change only defense, and see what the effect is, then another in which I change only resistance, and see what the effect is.

But in each case, I have to use the existing value, not 0, as my baseline to get accurate information. Because that is what answers the question I care about, which is "how much will this change my life."

What you seem to be doing is what everyone else does when they calculate survivability. 40% defense is 80% mitigation. 45% defense is 90% mitigation. Everyone knows not to presume that means 45% defense is 13% better. You seem to think that the standard mitigation calculations assert that. They don't. The standard model says if you're going to compare 40% defense to 45% defense, you compare incrementally. 40% defense gets hit 10 times out of 100, 45% defense gets hit 5 times out of 100 (oversimplifying here for the even minion case). Thus, 45% defense is twice as good.

What you've done is redefine the term mitigation to mean something it does not mean. Mitigation is what gets blocked, but you've decided to redefine the term to mean what gets let in, just because you don't find what gets blocked to be useful. Rather than do that, because that's wrong, I defined Mitigation to be something that actually relates to mitigation, and I use a completely different set of calculations to calculate survivability, and called them "Survivability" so I didn't have to radically alter the definitions of the words.

Other players do likewise: both Werner and Starsman I believe create and use metrics that is essentially a survivability rating to compare different levels of defense, resistance, and regen. There's nothing stopping you from doing likewise. But if you call it mitigation in defiance of the english definition of the word, you'll only be introducing confusion into the conversation.

OK... I am not 100% where the confusion is here, I see two lines of thought being cross referenced but cant track it back to the origin... this been a long conversation.

The "offending" logic here seems to be this bit by Arcanaville:

If this is not it, then ignore this post. But here goes:

Defense causes enemies to miss. But enemies also can miss on their own. You may not consider there is a difference, but the game has one.

Look closely at enemies attacks, and you will see sometimes they pop a MISS, sometimes they pop a DEFLECTED or AVOIDED, sometimes ABSORBED will pop up (and that will become very confusing once Bio goes live.)

When you measure defense, you are not trying to see how many attacks didn't land, you are trying to measure how many will be Deflected or Avoided compared to how many will Hit.

We dont tend to care, miss is miss after all. But the stat is just adding deflections and avoidance.

Counting misses as "mitigation" is as seeing some one trip 100 meters away and take credit for knocking him down.

Mind you, the chat log only says MISS or HIT, so it's hard to count Deflections and avoidances (they just flavor text, all are just deflection.)

Also, one other bit, I hate adding a percentage sign to defence rating, mainly because defense is not a percentage of nothing. Defence is a rating that gets evaluated against a critter rating to then determine how many attacks are you likely to deflect. It has a potential of swinging up or down, but thanksfully most of the enemies in the game only have 50 defense plus varying accuracy rating.

If your foe has 50 base ToHit, then 25 def is equivalent to 50% damage avoidance (or deflection, or mitigation, whatever you rather call it.)

If the foe has 64 base ToHit, then 25 def is only 39% damage avoidance.

As for regeneration and how everything else adds up... maybe later. Don't want to complicate this further, but I personally dont consider healing or regeneration to be "mitigation".

I split survivability between mitigation and restoration.