While it's reassuring to see that S-E uses basic modern standards of game RNG algorithms, telling everyone that a good algorithm is used doesn't mean jack squat if a good seeding process isn't being used, something that hasn't been acknowledged at all beyond 'the seed for each algorithm we use is different'. A seeding method that reseeds frequently but doesn't give adequate deviation between each seed is going to produce inadequate deviation by the RNG algorithm, which will crease massive bias toward a certain result.

That said, there isn't a lot of compiled evidence beyond "I observe streakiness!" Which compounds the whole problem of things. It's difficult from a client side to tell if getting the same loot drop from ST 12 times in a row is bad luck or a screwy seeding method.