Data for experiment.

Also, no cooldowns were used to avoid clipping issues.

GCD 2.5:

VB : 1861854 ticks - Uptime: 93.09%
WB : 1895459 ticks - Uptime: 94.77%
Ticks: 2000026

GCD 2.4:

VB : 1807865 ticks - Uptime: 90.4%
WB : 1841453 ticks - Uptime: 92.07%
Ticks : 2000045

GCD 2.3:

VB: 1824566 ticks - Uptime: 91.23%
WB: 1865213 ticks - Uptime: 93.26%
Ticks: 2000024

GCD 2.2:

VB: 1825939 ticks - Uptime: 91.30%
WB: 1875632 ticks - Uptime: 93.78%
Ticks: 2000032

GCD 2.1:

...

Ugh I'm too lazy to write out the rest.

Conclusion: Experiment variance far outweighs actual uptime differences. SS has little to no effect on human (simulated, with recast at sub 1 second, no clipping) DoT uptime.