I apologize, I was not calling you out.
the math is impeccable.
the company is suspect.
What I've mentioned in a few posts is that what the RNG aproximates as random is based more likely on a pseudorandom number generator which uses a seed for each user.
this seed selection can explain both why there is seeming randomness and the fact that there is bias that can exist.
I would argue that using a poor prng would be better for a 'gatcha' because it almost guarantees that there is an approximation of an even distribution without needing to have one. This seed can also explain why certain users see multiples of the same god.
secondly - assuming that the rates don't change over time is incorrect as well. Gungho can and does manipulate the drop rate for every dungeon by the hour (increased drop rates 1.5% over the week for "x" hours). There is no reason to think that the rates during the godfest have to stay the same for each god over the course of the godfest.
I definitely was not trying to be picky or overly technical, I was actually arguing the opposite - that by trying to be picky and technical we are trying to assign numbers to something that is being given to us by a for profit company that has recently had to drop its stamina rates in JP because it is not fairing as well as it was 1 yer ago. I do not believe for a second that Gungho has my best interest in mind but they want my money.
I play and will be addicted to this game until the next, better one comes out - I have supported the game more than most (I have purchased 15 - 85 stone packs at full price) I have also posted my PADherder and do have a great selection of mobs/cards.
I do not believe for a second that the drops are random.
Yeah, sorry for my somewhat aggressive reply. But I stand by what I said, peope love seeing patterns where there are none. Saying "I get a lot of gold/bad eggs in a row" or "some people get multiple dupes" is not, in any way, evidence of a controlled RNG, because this SHOULD happen a lot if the rng acted random.
"I would argue that using a poor prng would be better for a 'gatcha' because it almost guarantees that there is an approximation of an even distribution without needing to have one. This seed can also explain why certain users see multiples of the same god. "
That's the thing. I don't know how many gods there are, but I counted 13 pantheons with 5 gods each, and 15ish godfest exclusives. So, let's say 80ish gods in total.
As surprising as it is, If you pull just 10 gods out of a pool of 80, with each gods equally likely, you basically have a near-50% chance of getting a dupe. (For calculations, google "Birthday problem")
If, on the other hand, the gods are NOT equally likely (as in, godfests/galas/etc), your odds of pulling a dupe will go up, not down, unless you only aim godfests/galas where you own none of the gods at all.
Maybe Gungho seeds player, but I'm just saying that, we're not going to be able to notice from supposed patterns and forum hearsays, and it's harmful to the users to pretend otherwise. People start making up theories and spending money because they think they cracked the code.
If anything, if I was gungho and I wanted to seed the players, I'd make it so they're less likely than usual to get 7 bad pulls in a row. 7 bad pulls is enough to make people quit and stop spending money, they'll think the odds are rigged, even though it's perfectly normal.
