Keep in mind that while EF is a lot stronger on total healing(/absorb) output, it is only useful during the tick, while SS' absorb effect will apply to the next blow. Once EF does less healing than SS absorbs per tick, it doesn't really matter that EF still puts out more total healing, it won't help as much as SS. EF might tick twice as often, but that advantage is mostly negated by EF ticks only being applicable in between attacks, whereas SS is also useful before an attack.
Thus, for a string of 3 attacks, we can rely on 1 SS tick and 1 EF tick.
I'm actually a little confused how BoG stacks are included in Slootbag's sheet. Changing the selfcast multiplied from 2 to 1.5 is easy enough, but what about the bonus for BoG. It lists a 2.5 multiplier, but each stack increases output by 10%, right? So shouldn't it have been 1.5 (or at most 1,61051 if it was multiplicative, which I doubt is the case).
EDIT: I checked the video again, and totally forgot to include Mastery. Then looked back at the sheet, and noticed that mastery was included in the formula. *headdesks*
Testing it with 2.2 (20% Mastery + 4 BoG) and EF still comes ahead per tick.
Testing it with 1.9 (20% Mastery + 3 BoG) and EF starts to lose above 100k vengeance.
Of course, it's quite possible for players to have more than 20% Mastery.
At 25% Mastery (and 3 BoG stacks), EF and SS tick about evenly hard at 200k vengeance, with SS winning at higher vengeance levels.
At 30% Mastery (and 3 BoG stacks), EF still wins all the way up to 400k vengeance (though SS will take over a little higher than that).
These might sound like high numbers, though not too high if you assume BoM present, which we should, since we provide that buff ourselves, and so many classes can provide stats.
Either way, those numbers all assume 3 BoG stacks. At 4 or 5 BoG stacks, EF is a clear winner. And even at 3 BoG stacks, it counts per EF tick, while EF ticks twice as often as SS.
TIP: It might be useful to add a "BoG stacks" field to the sheet, just like there's a "Mastery" field.