Would they even dare try this considering that Granblue/Cygames got raked over the coals for pulling this exact stunt?
As they say in professional sports, "It is only illegal if you get caught". How much goodwill does GH have to risk?
----------------------
Another popular phrase is "The fix is in."
Enough anecdotal evidence has been compiled today on Reddit that some disturbing trends have been identified. As a data summary dump from the statistical analysis there for those that are interested...
The high level summary:
Gold rate: 25% (1 in 4)
Silver rate: 75% ( 3 in 4)
Seems pretty normal, right?
But when the data was compiled, it was quickly noted that one gold card per rarity tier appears to be listed at a lower rate than what would be suggested by the card's star rating (as was suspected earlier on this page by others). Plus, troll golds are much heavier weighted than other golds in the draw.
The low level summary:
n ~ 2700 (at the time of this post)
Silvers: 5 silvers, at 15% each; equally weighted (75% total)
Troll Golds:
5* Kano + Ruka, 7.5% each (15% total troll gold).
** Take away the troll golds, the real gold rate is 10% (1 in 10).
But, the fun does not stop there. Although there is some room for interpretation, as the percentages become small within that 10% and fit within the margin of error currently, there is a good feel what the rates probably are to add up to the 10%.
REAL GOLDS:
5* Akechi - 3% (1 in 33)
6* Ra - 3% (1 in 33)
6* Bastet - 2% (1 in 50)
7* Satsuki - 1% (1 in 100)
*** There are two possibilities for the 8* cards, as their draw rates are extremely tiny. Both theories are listed below.
-- Theory A (less likely, but possible because of margin of error)
8* Gadius 0.5% (1 in 200)
8* Eschamali 0.5% (1 in 200)
-- Theory B (more likely scenario based upon compiled data)
8* Gadius 0.667% (1 in 150)
8* Eschamali 0.333% (1 in 300)
The collab machine is definitely a trap.