https://stowiki.net/wiki/Specialization:_Energy_Weapons_Officer#Crit_Chance_variant
https://stowiki.net/wiki/Specialization:_Projectile_Weapons_Officer#Crit_Chance_variant

I recorded a PUG ISA run using a single CritH doff and a single CritD doff and tallied the buff stacks every 5 seconds. I flew a CSV-built Akira (7 cannons + Soliton). The run took about 97 seconds and I parsed 317k DPS (actually a personal best in this ship, parse below).
Side Note: The buff icons are crazy, laggy, and often report wrong times (like briefly displaying 30 seconds before returning to 4 seconds, and such). I couldn’t tell which icon was which (CritD vs CritH; they use same icon), but I did watch their cooldowns to help track them individually.
Time Stamp | Buff1 Stacks | Buff2 Stacks | Note |
14 | 0 | 0 | |
19 | 1 | 1 | Opening Wave |
24 | 3 | 1 | |
29 | 3 | 1 | Left Transformer took first hit at 30 sec |
34 | 3 | 2 | |
39 | 0 | 2 | |
44 | 1 | 3 | |
49 | 1 | 3 | |
54 | 1 | 3 | Transitioning to Right Transformer |
59 | 0 | 0 | Right Transformer took first hit at 60 sec |
1:04 | 2 | 0 | |
1:09 | 2 | 0 | |
1:14 | 3 | 0 | Right Transformer died at 1:16 |
1:19 | 3 | 0 | |
1:24 | 3 | 1 | |
1:29 | 0 | 1 | |
1:34 | 0 | 2 | |
1:39 | 1 | 2 | |
1:44 | 1 | 3 | |
1:49 | 1 | 3 | |
1:54 | 0 | 0 | Cube died at 1:54, coincidentally, but stacks wore off on their own. |
Dropping the initial data point at the beginning gives us a convenient 20 data points. If we count up the number of times a 0 appears in the first column, we can find the percentage of time that no buff was active. We can do that for each buff count (0, 1, 2, 3/max), and for both columns:
% of time with 0 stacks | 25% | 30% |
% of time with 1 stack | 35% | 25% |
% of time with 2 stacks | 10% | 20% |
% of time with Max stacks | 30% | 25% |
The important takeaway is that, with only one copy of the doff, I only had about 25-30% uptime with max stacks, and conversely, I went about 25-30% of the time without any buff stacks.
ADDITIONAL NOTE: Because the specific buffs (critD vs critH) could have swapped places at the 59 second mark, it might be worth it to swap the data points from that point on also, creating essentially 2 sets of data (only one is actually correct, but we can’t know which). Running the numbers for all 4 sets of data gives:
% of time with 0 stacks | 25% | 30% | 35% | 20% | |
% of time with 1 stack | 35% | 25% | 30% | 30% | |
% of time with 2 stacks | 10% | 20% | 10% | 20% | |
% of time with Max stacks | 30% | 25% | 25% | 30% |
Very similar, but not identical. And before we worry about it too much, we should remember that this is all based off of a single ISA run. Not exactly a sufficient sample size for exact numbers, but sufficient to show that a single doff was not enough to keep max stacks up in this run.

One reply on “Crit Doffs”
Thanks for the testing and sharing of the results!