I’m trying to understand why crit is often stacked higher than it needs to be based on my math:
I was looking at this in terms of gunslinger crit, if that matters.
Each point of crit = 0.03579%
Each point of spec = 0.03577%
Formula for damage increase based on crit with 100% crit damage:
(1+1*(new crit)) / (1+1*(old crit)) = % damage diff
Formula for damage increase based on crit with 150% crit damage:
(1+1.5*(new crit)) / (1+1.5*(old crit)) = % damage diff
Solving for “x” to figure out when one point of crit = one point of spec for +150% crit damage
(1+1.5*(x+0.0003579)) / (1+1.5*(x)) = 1.0003577
X = 0.33389 = ~33.4% Crit
This would mean any points in crit after 33.4% is inferior to specialization when you have +150% crit damage. So why are many builds based on such high crit?
I believe that those builds are not fully based on the dmg output, but rather the effect that they give to the identity of the advanced class.
So for example, with Arcana you have two viable builds with main stat Spec and other with main stat Crit (you can also go for Swift, but that’s not on topic). If you’re going Empress build, people would most likely go for a main Spec build as it both increase the damage output of the Ruin Skills and the Identity meter. Going Crit withan Empress build wouldn’t be so optimal (it can still be optimised, yes, but it wouldn’t be the “ideal” play style).
On the other hand, if you’re going for an Emperor build, you’d most likely go for a main Crit stat build, so your Crit Rate goes up as much as possible, so whenever you land an Emperor card you do huge PP dmg (taking in consideration you go with KBW as well)
There are some builds indeed that uses your logic to optimise it as much as possible, but it will all depend on what’s the play style of the build.
Hope I didn’t got confused 