If your crit rate is high enough to give the stacks of might you claim, your crit rate is also high enough to keep bleed on the target. That's not opinion, that's just probability.
Your damage comparison is flawed. You need to start thinking in average might stacks over time rather than an all or nothing approach. If I have 10% boon duration compared to 30% boon duration I do not lose all 25 stacks, I lose 20% of the base 5 sec duration. With the +50% boon duration runes and +30% boon duration traits the greatsword might proc lasts 9 seconds. With the 50% boon duration runes and only 10% boon duration traits it will only last 8 seconds. That's an 11% decrease in the duration of might stacks. Or an 11% decrease in the time you have the +41.7% damage you say you receive via might (which actually translates closer to an 11% decrease in average stack size). I would rather do 141.7%+20% damage 89% of the time and 120% damage the other 11% of the time than 141.7% damage 100% of the time, but it won't be as clear cut as this since everything is moving through a temporal dimension.
I could go one further and show you what your average might stack might look like given certain crit chances and attack speeds (hint, default attack with 75% attack speed and might duration at 9 sec compared to 8 sec are within 2 stacks of each other on average) in a nice pretty graph I knocked up in ~10 minutes, but based on your response I'm sure you're about as good as Red Falcon at ignoring data. There is no chance in hell going from 8 secs to 9 secs on the mighty greatsword proc and from 16 secs to 18 secs on the sigil proc increases your average might stack by more than a +20% damage increase. I can absolutely 100% guarantee you will see more damage by taking the 20 points in tactics and putting 10 into strength for +10% damage and at least 5 into arms for +10% damage when bleeding.
Group support and healing are another debate entirely though.
Anyway, I don't like having these sorts of discussions because no matter how many numbers I throw out or graphs and spreadsheets I send via private message - most people reply with something along the lines of "it feels like less damage, you're wrong" and ignore it, so I'll leave this here.
PS - Your use of the word exponentially in the first paragraph is incorrect. I believe the word which best describes the benefit from +power is logarithmic growth.
Disregarding the bleed talk, because you're probably right (Even though bleeds will and do occasionally drop off) and granted for some reason I was thinking you were going to be using different runes (And also the information in that quote isn't mine, I was just referring to it as it seems reliable)
So removing my two blunders, you happen to be right, this will net you quite a bit more damage. However, the build is as it is so that you do not fall beneath 25 stacks very often/at all. Compared to your iteration this would probably be about 3-5 more might stacks on average. Clearly that does not make up for the 20% damage, but when you consider the added boon duration to FGJ you put on allies and the increased uptime of OMM, it just might make up for a majority of that. Not to mention the 200 vitality, 20% shout reduction, and the fact that your shouts heal for 1400 (Used all at once is almost the equal of an extra heal skill for your entire team).
I'm not sure how the extra 3-5 stacks on yourself and the extra stacks/duration on allies come into play, but you're probably right in that you will net more damage. But as MacG states, damage isn't the entire function of this build, and if you take away from the support aspects you start to make having shouts less useful etc etc until you come up with a new build entirely.
It is true what you said about the support aspects being a different debate entirely but I do feel as though the added party damage you gain should be reflected into your calculations for damage (As you wouldn't have as much might or fury up from FGJ, and also not for as long, if you were to take 20 points out of tactics, and also OMM wouldn't be up as often).
Regardless, I concede that your iteration may provide more damage (Would love to know for sure how the group boon duration/shout CD comes into play, could you possibly do some tests/calculations for that?) but it definitely provides less of the other facets of this build, and ultimately would dissolve into another build entirely focused on maximizing damage with no support.
As I've stated many times, the great thing about this build is it's ability to provide support and good damage
, while neither of these aspects are wasted in a way that makes you want to lean towards total damage/total support
. You can keep both areas without sacrificing so much from one side or the other that you would be inclined to go full damage or full support
. When you start to take out traits from tactics and turn them into more damage, your shouts become so much less useful that you don't need them, and whole sections of the build would be better suited in other places to increase your dps.
Personally I dislike Red Falcon, I can see the numbers and understand them. You would end up getting more damage most likely, but at the end of the day this build's niche is not that of maxing self damage, which is as you say "another debate entirely". (Minus the FGJ/OMM duration issue, which may very well increase party damage by enough to make up for your loss)
I hope you do actually read this, and I'm hoping you do test that FGJ/OMM duration thing because I am now curious. Thanks for bringing in more discussion!
P.S. Exponentially there was not meant to be taken entirely literally, it was just a turn of phrase. It was also in reference to the original build Sithicus gave me, which was indeed supposed to "Exponentially" increase damage output (Compared to that of a full support build). However, I live to please. The first paragraph has been edited to properly and literally state the type of growth involved.
(This was mostly a joke and with the newest update I have removed the word Logarithmically in favor of just calling it a "damage increase" with no specific type of growth.)
Edited by Brand, 17 November 2012 - 11:41 AM.