So, will critical damage mod sets now be useless?

Lol. The best part about this thread is that you guys are trying to minmax on a game that is COMPLETELY nonhomogeneous. Almost every damage dealer either has some boost to stats that changes how they stack stats (self buffing crit damage, or offense up, or stacking offense of health converting to offense or whatever) in addition to risk/rewards for stacking stats (incurring a taunt for critting, taking an extra attack, clearing debuff on a critical, reducing cooldowns on a critical, inflicting specific debuff on a critical, etc.). That's not even taking into account that for at least 60% of all characters, crit dmg/offense are not your go to set.

The real, honest truth is that every character needs individual consideration, and has best in slot stats that interact with abilities, especially the newer characters.

If you guys want a real hard and fast rule, it's this: use crit damage on characters that yield high crits (wedge, CLS, FOE, Rey, etc). This will naturally spike their peak damage, and make for more interesting and exciting gameplay. Save your good crit dmg sets for them, and use offense on anyone else.

I don't know about the rest of you, but I'm going with "If your character will have 75% or more crit chance including mods and composition bonuses (R2, crit chance buff, leader bonus), then mod for crit damage, otherwise mod for offense.)"

And yes I can do the math, but I prefer to just play the game. One of the advantages of not fighting for the top spots I guess.

Love this thread!

So many variables like character basics team comp and mod sub stats. Theory crafting is still theory . At the end of the day it’s not about my number is bigger than yours but how you use it.

"crzydroid;c-1624479" wrote:

Finally, I am sorry to point out that while working through your process, I did discover an error in your setup

Now I will endeavor to show how our two methodologies should actually arrive at the same result. You want to take the increase in damage from offense set over no set: o (x)/f (x). You want to do likewise with crit damage, so cd (x)/f (x). You then want to compare these values to one another, so (cd (x)/f (x))/(o (x)/f (x)). You can see that the two denominators in both the numerator and denominator of the resulting compound fraction actually cancel, by multiplying it by f (x)/f (x) (or 1). You wind up with cd (x)/o (x).

Incorrect representation of this from the start here. Offense set ratio would be represented by BO + SO / BO CD is represented by CD + SB / CD.

I'm not using any of the same figures in either or these ratios. To represent them with the same variable as you have done is confusing, misleading, and most of all, inaccurate.

In other words, the mere fact that you have used the same variables to represent the parts of my formula for the offense set, and the parts of my formula for the crit damage set means that there is no way you could possibly have found an error in my math with your reasoning.... because you aren't using my math.

"Woodroward;c-1625014" wrote:
See it's like I figured out that the offense set gives an 8.3% increase in offense. So my representation for the offense set is 8.3% Sure I could figure out what he hits for on a crit, how often he crits and what he hits for on a non-crit.

To simplify let's say someone hits for 10k before set bonus is considered. With the offense set, they would hit for 10,830. On a crit they would hit for 19,200 without the offense set and 20793 with it. I can calculate the average damage by using crit chance. let's say 40% so, ((20793 * .4) + (10830 * .6)) - ((19,200 * .4) + (10000 * .6)) = 1135.2 The average damage increase for the offense set would be 1135.2. Now let's see what % of the non offense set damage that is. (19,200 * .4) + (10000 * .6) = 13680.

1135.2 / 13680 = 0.082976... 8.3%. It worked out the same with or without going through all that extra math (not counting a slight variation due to rounding). So the % damage increase is the same regardless of crit chance. This means that we don't have to differentiate between crits and non crits for the offense set since we cover both by using % instead of flat damage.


"Instead, the final step of your calculation essentially takes
(x *(1-c))/(1.083x*(1-c)) + (2.22*c*x)/(1.92*1.083x*c). This is, as you put it, improper math."

No. no this is not an accurate representation of my formula at any point. You are sticking factors into my equation that were already factored out before it hit the paper, like crit chance. It's irrelevant. It is only skewing your results to include it. It is to avoid skewing of results that I made sure to factor it out as early as possible.

The ONLY step of my formula (that's what's great about it, it's only an ongoing calculation that can be solved linearly, not an equality/inequality that requires many many steps) is the one I have shown. It doesn't have multiple steps. It doesn't require doing things in "chunks" as you call it. It is linear, it is quick, it is streamlined, and it is 100% accurate.

This is my formula, what you have shown is avidly not

(1 - (Physical damage listed in panel / physical damage - offense set physical damage) * 100) / (1 - (222/192) * 100)


I don't know if I can explain to you any clearer that you are the one making the mistake that I was making when I first addressed you in this thread... skewing the results by improperly including constants that should be factored out.

So somewhere there is a communication error here. I am not talking about any calculations regarding the offense set. That set does provide a constant increase in damage.

I was referring to the fact that you leave the non-crit portion out when relating to the critical damage set. While critical damage sets only affect the crit portion, in comparing them to offense sets, which affect all portions, you need that non-crit portion to tell what value of critical chance will provide more damage overall. As soon as you introduce critical chance into the question, the problem necessarily becomes one of both crit and non-crit damage--for that is what critical chance is: It's that weight which determines how often a hit is a crit and how often it is not. Even if you get more critical damage with a cd set, you will get less non-crit damage than you would have with an offense set. When you then frame the question in terms of what critical chance provides an equal mixture of crit and non-crit for each set, you need the non-crits in there.

I've actually amended the next part of what I was planning on saying in terms of calculations, because I realized you were simply approaching this in a completely different way--while my steps for showing what you did show an answer resulting from an improper combination of fractions, the approach assumes you were thinking of solving for critical chance in the first place. I provided a mathematically equivalent vetsion of your formula (go ahead, try it out), but your question was framed differently.

You took

(1-1.083)/(1-(2.22/1.92))

(Again, this is your formula without the *100--I just put it in decimal form up front, with the result in decimal form as well. Go ahead and try it, and see if you get the same answer).

Anyway, this formula is taking the percent increase in offense set crits vs. non-set over the cd set percent increase in crits over non-set.

The result is is the proportion of increase in crit damage with an offense set to the increase in crit damage with a cd set. It is ONLY concerned with the damage increase for one set over the other ONLY when you crit. This isn't the question we want to answer. It also bears no relationship to crit chance. Crit chance, in this scenario, for all intents and purposes, is 1.00. Either it is actually 1.00, or it is some other non-zero value, but since you are only looking at crits in determining the effectiveness of one set over the other, the value doesn't matter. It doesn't tell us anything about a crit chance breakpoint for which the mixture of crit and non-crit is equal for both sets. It is important to look at this mixture because even as cd sets provide more on a crit, they provide less on a non-crit.

When you solve your equation and come up with 53.12%, that result is completely independent of any level of crit chance, because you are only looking at crits. It does not mean there is a 53.12% crit chance breakpoint. It means an offense set only provides 53.12% of the increase in damage over a cd set for only those instances where you crit. You have provided a perfectly accurate formula for that number, but it is not the question we are asking.

If you are only concerned with crit damage, and a cd set provides more crit damage, then it will ALWAYS be so for crits. But you are not concerned with crit chance in this scenario. When you start asking questions about crit damage breakpoint, you are now concerned with non-crits as well. For that is what critical chance means--it is fundamentally tied to the mixture of crit and non-crit. So it is absolutely vital that both those components be included in any calculation of critical chance. It is also helpful to include the variable of interest in the appropriate place in the equation.

So putting my formula in the same form as yours, you would have

(1-1.083)/(1+1.083*1.92-1.083-2.22)

As you can see, the numerator is the same, but the denominator is different, taking into account the relationships between the sets when the fundamentally important component of non-crit damage is left in. This set up is appropriate to the question we are asking, and finds a crit chance breakpoint of 37%. You will also see that this is mathematically equivalent to the formula I listed above--it is simply a computational form as opposed to a theoretical or definitional one. Those terms are all still there, it's just some of them were pre-calculated, mostly by the game itself. And you can see in my reponses to other posts that I am an advocate of using the game panel offense in practice.

"Aluxendr;c-1625070" wrote:
Lol. The best part about this thread is that you guys are trying to minmax on a game that is COMPLETELY nonhomogeneous. Almost every damage dealer either has some boost to stats that changes how they stack stats (self buffing crit damage, or offense up, or stacking offense of health converting to offense or whatever) in addition to risk/rewards for stacking stats (incurring a taunt for critting, taking an extra attack, clearing debuff on a critical, reducing cooldowns on a critical, inflicting specific debuff on a critical, etc.). That's not even taking into account that for at least 60% of all characters, crit dmg/offense are not your go to set.

The real, honest truth is that every character needs individual consideration, and has best in slot stats that interact with abilities, especially the newer characters.

If you guys want a real hard and fast rule, it's this: use crit damage on characters that yield high crits (wedge, CLS, FOE, Rey, etc). This will naturally spike their peak damage, and make for more interesting and exciting gameplay. Save your good crit dmg sets for them, and use offense on anyone else.

Welcome to the discussion.

Yes, it has been a general rule of thumb to put cd sets on characters that crit a lot. But this thread started because of the question as to how the new set bonuses and 6e mods would affect the decision making process.

Since offense sets are multiplicative and crit damage sets are additive, there's a level of crit damage from other sources for which the cd set does not provide as much of an increase relative to the offense set, even for characters who have high or maximum critical chance. This is true under the current system; however, the threshold is unlikely to be reached with current sources of bonuses (maybe Asajj Ventress could reach it after enough deaths).

The new bonus for offense sets, however, decreases that threshold. Simultaneously, slicing a critical damage primary mod to 6e provides mre distance towards reaching the threshold.

The balance to this is that Woodroward pointed out that offense set bonuses only apply to base offense. Therefore, the more offense you add from mods, the more the threshold increases again. Furthermore, those offense values can be sliced as well, further raising that threshold back up.

So the thread is about how, if at all, are the situations where you use one set over the other changed by these new mods.

Of course, as you can see from some of these later responses, some people will just keep on playing the way they play and having fun.

"crzydroid;c-1625488" wrote:

"Woodroward;c-1625014" wrote:
See it's like I figured out that the offense set gives an 8.3% increase in offense. So my representation for the offense set is 8.3% Sure I could figure out what he hits for on a crit, how often he crits and what he hits for on a non-crit.

To simplify let's say someone hits for 10k before set bonus is considered. With the offense set, they would hit for 10,830. On a crit they would hit for 19,200 without the offense set and 20793 with it. I can calculate the average damage by using crit chance. let's say 40% so, ((20793 * .4) + (10830 * .6)) - ((19,200 * .4) + (10000 * .6)) = 1135.2 The average damage increase for the offense set would be 1135.2. Now let's see what % of the non offense set damage that is. (19,200 * .4) + (10000 * .6) = 13680.

1135.2 / 13680 = 0.082976... 8.3%. It worked out the same with or without going through all that extra math (not counting a slight variation due to rounding). So the % damage increase is the same regardless of crit chance. This means that we don't have to differentiate between crits and non crits for the offense set since we cover both by using % instead of flat damage.


"Instead, the final step of your calculation essentially takes
(x *(1-c))/(1.083x*(1-c)) + (2.22*c*x)/(1.92*1.083x*c). This is, as you put it, improper math."

No. no this is not an accurate representation of my formula at any point. You are sticking factors into my equation that were already factored out before it hit the paper, like crit chance. It's irrelevant. It is only skewing your results to include it. It is to avoid skewing of results that I made sure to factor it out as early as possible.

The ONLY step of my formula (that's what's great about it, it's only an ongoing calculation that can be solved linearly, not an equality/inequality that requires many many steps) is the one I have shown. It doesn't have multiple steps. It doesn't require doing things in "chunks" as you call it. It is linear, it is quick, it is streamlined, and it is 100% accurate.

This is my formula, what you have shown is avidly not

(1 - (Physical damage listed in panel / physical damage - offense set physical damage) * 100) / (1 - (222/192) * 100)


I don't know if I can explain to you any clearer that you are the one making the mistake that I was making when I first addressed you in this thread... skewing the results by improperly including constants that should be factored out.

So somewhere there is a communication error here. I am not talking about any calculations regarding the offense set. That set does provide a constant increase in damage.

I was referring to the fact that you leave the non-crit portion out when relating to the critical damage set. While critical damage sets only affect the crit portion, in comparing them to offense sets, which affect all portions, you need that non-crit portion to tell what value of critical chance will provide more damage overall. As soon as you introduce critical chance into the question, the problem necessarily becomes one of both crit and non-crit damage--for that is what critical chance is: It's that weight which determines how often a hit is a crit and how often it is not. Even if you get more critical damage with a cd set, you will get less non-crit damage than you would have with an offense set. When you then frame the question in terms of what critical chance provides an equal mixture of crit and non-crit for each set, you need the non-crits in there.

I've actually amended the next part of what I was planning on saying in terms of calculations, because I realized you were simply approaching this in a completely different way--while my steps for showing what you did show an answer resulting from an improper combination of fractions, the approach assumes you were thinking of solving for critical chance in the first place. I provided a mathematically equivalent vetsion of your formula (go ahead, try it out), but your question was framed differently.

You took

(1-1.083)/(1-(2.22/1.92))

(Again, this is your formula without the *100--I just put it in decimal form up front, with the result in decimal form as well. Go ahead and try it, and see if you get the same answer).

Anyway, this formula is taking the percent increase in offense set crits vs. non-set over the cd set percent increase in crits over non-set.

The result is is the proportion of increase in crit damage with an offense set to the increase in crit damage with a cd set. It is ONLY concerned with the damage increase for one set over the other ONLY when you crit. This isn't the question we want to answer. It also bears no relationship to crit chance. Crit chance, in this scenario, for all intents and purposes, is 1.00. Either it is actually 1.00, or it is some other non-zero value, but since you are only looking at crits in determining the effectiveness of one set over the other, the value doesn't matter. It doesn't tell us anything about a crit chance breakpoint for which the mixture of crit and non-crit is equal for both sets. It is important to look at this mixture because even as cd sets provide more on a crit, they provide less on a non-crit.

When you solve your equation and come up with 53.12%, that result is completely independent of any level of crit chance, because you are only looking at crits. It does not mean there is a 53.12% crit chance breakpoint. It means an offense set only provides 53.12% of the increase in damage over a cd set for only those instances where you crit. You have provided a perfectly accurate formula for that number, but it is not the question we are asking.

If you are only concerned with crit damage, and a cd set provides more crit damage, then it will ALWAYS be so for crits. But you are not concerned with crit chance in this scenario. When you start asking questions about crit damage breakpoint, you are now concerned with non-crits as well. For that is what critical chance means--it is fundamentally tied to the mixture of crit and non-crit. So it is absolutely vital that both those components be included in any calculation of critical chance. It is also helpful to include the variable of interest in the appropriate place in the equation.

So putting my formula in the same form as yours, you would have

(1-1.083)/(1+1.083*1.92-1.083-2.22)

As you can see, the numerator is the same, but the denominator is different, taking into account the relationships between the sets when the fundamentally important component of non-crit damage is left in. This set up is appropriate to the question we are asking, and finds a crit chance breakpoint of 37%. You will also see that this is mathematically equivalent to the formula I listed above--it is simply a computational form as opposed to a theoretical or definitional one. Those terms are all still there, it's just some of them were pre-calculated, mostly by the game itself. And you can see in my reponses to other posts that I am an advocate of using the game panel offense in practice.

See this is what I'm trying to explain to you. I'm not comparing average damage for the sets. That's a complicated messy waste of time. Instead I'm comparing only the increase each set provides. When does a crit damage set increase the value of a non-crit? It doesn't.

The offense set provides a 8.3% increase on crits and non-crits. The crit damage set provides only a 15% increase on crits. The 1.0 offense and the 1.92 crit damage can be safely factored out of both sets, as it is exactly the same for both. By only comparing the crit % damage increase to the offense % damage increase, I am taking into account all damage immediately, and then immediately factoring out all damage not related to the sets.

The results I am obtaining from this formula are completely independent of any level of crit chance on the mods, because it was factored out. It has no bearing on whether an offense set or a crit damage set can produce more offense at this level of the equation.

In the end this ratio makes a comparison of the actual damage of the offense set to the POTENTIAL damage of the crit chance set.

If an offense set provides a 77% increase in damage 100% of the time compared to an crit damage set that provides 100% damage 77% of the time (because it crit 77% of the time), then they are equal. So the % increase in damage than an offense set provides compared to a crit damage set IS the crit chance breakpoint where a crit damage set becomes better than an offense set. If the result is 1.21, then at 121% crit you'd be better (assuming the damage continued to rise which it doesn't, hence anything over 1 means offense is just better) Anything over 1 means a crit damage set isn't capable of producing more damage than an offense set.

Otherwise, it shows that a crit damage set is CAPABLE of producing more damage than an offense set, but only if it can produce results at that % of the time. This is not in any way a calculation error, it is a massive simplification with 0 loss of accuracy. If you are experiencing different results it is because you are somehow improperly including constants you don't need to, like crit chance, base offense, and base crit damage. For instance this formula right here is incorrect:

(1-1.083)/(1+1.083*1.92-1.083-2.22)

Since you insist upon including things you don't need to. you should at least have the proper format:
(1.083 - 1) / ((1 * 2.22) - (1 * 1.92))

See in your version you are simplifying to just the offense bonus on one side, then adding base offense and base offense and set offense together, multiplying it by a crit with no cd damage set, then subtracting base offense and set offense and subtracting the crit multiplier with a set and triangle. I see why you keep getting a different result than me now. This is not even close to accurate math.


Again, this question is NEVER asking, "which of my mod sets is better?" It is asking, " which set of mods is the best that can be gotten?"

I think I'd like to know which mod set has a higher average damage dealt in a battle.

"ImYourHuckleberry;c-1625539" wrote:
I think I'd like to know which mod set has a higher average damage dealt in a battle.

This can easily be obtained by comparing % damage increases rather than calculating out all the flat values and averaging them. That way is just longer and has far more room for error.