I’ve only done 4 GAPs myself thus far, but as far as I can tell, it doesn’t appear to be nerfed as I’ve gotten a total of 11 EVKs during that span. But definitely some streaky RNG as I got 0, 8, 0, 3 EVKs for each 15 minute GAP. So my advice is to keep at it, your luck will turn.
I’ve done 12 GAPs. 4 of them have gotten 0 Epic keys. That’s 892 gnomes and no Epic keys.
RNG has never, EVER, adjusted the other way across a negative that large.
And before you ask, the EVKs I did get from the other GAPs, 2 had 4 (expected), the rest were all under 10%.
While technically true, your last sentence is misleading because the 4 GAPs in which you received no epic keys were presumably not all in a row, which would be a much rarer event than picking out four different GAPs in which you received no epic keys.
You certainly should not expect to get 4 EVKs/GAP on average.
Let us assume a 10% VK drop chance, which we know is an overestimate from your data. We are told that 10% of all VK drops will be EVK, so the EVK drop chance would be 1%. If you encounter 224 gnomes during a GAP, then you should find 2.24 EVKs on average—and this is already an overestimate.
We can go further. Given the results above, the distribution of the number N of vault keys in a GAP follows Poisson statistics with rate parameter λ = 2.24, so that P(N = 4) = exp(-λ) λ⁴/4! = 0.11, so it turns out that you were quite lucky to get 4 EVKs in 2 GAPs out of 12 total GAPs.
I run double GAPs. Sorry, I did’nt mention that on this thread, thought I did on another. 2 of those double gaps got me 4 keys (expected). 2 of those got me 0 keys (which is where i got the 892 gnome total from). The rest were all under 4 and all under 10%.
And, that means I did back to back GAPS with no epic keys. Twice.
Thanks for clarifying, that does make more sense. I think it’s still more likely than not that it’s just bad luck over a small sample size, especially given the rarity of an EVK drop (<1%), but I guess we’ll find out with more data.
That being said, before running a 5-hour non-stop GaP, my epic key rate was closer to 0.7%, so take that part with a grain of salt. To me it feels like, key drop rate is 7% and epic key drop rate is 0.7% (which is 10% of the key drop, which is consistent with what we knew before).
The current results suggest more along, key (regular + epic) drop rate => 9% and epic key drop rate: 0.9%. That being said, again, the data points is not enough to completely reject that it could be 8% and 0.8%.