Sanctuary Onslaught: Khora Parts No Longer Dropping?

[Raspberry]

After obtaining all but Khora's Neuroptics blueprint, I've started doing runs of only 6 zones to get to rotation B, then extract if I don't get the neuro. However, after 20+ such runs, I've noticed that I have received no Khora parts at all (not chassis from rotation As, not neuroptics from rotation Bs), which is quite mind-boggling. Although 20 runs in a row with three rotations each without a single Khora part drop is still in the realm of possibility:

((1-0.0833)^(20x2) for two chassis) x ((1-0.0769)^20 for neuroptics) = 0.00622, or 0.622% chance of occurring

...I am starting to believe that the drop chances might be wonky. I'm going to keep at it for the next few hours of my playtime because I don't have many better things to do, but the chances of still not getting a single Khora part after more runs are going to look more and more rigged.

Chance of 25 6-zone runs in a row with no Khora parts: 0.00175, or 0.175% (still possible, albeit very unlucky for me)

30 runs: 0.000491, or 0.0491%

35 runs: 0.000138, or 0.0138%

40 runs: 0.00003587, or 0.00387% (pretty much shouldn't be happening)

FYI: I am running REGULAR Sanctuary Onslaught, not ELITE Sanctuary Onslaught, so I am making no mistake on my choice of mission.

Feel free to correct my math if I'm wrong somewhere. I'm not a huge fan of probability and statistics.

UPDATE: Eventually got it 3 runs after I made this post. :P

Edited April 27, 2018 by Raspberry

[Firetempest]

Just been unlucky and it can get worse Drop	Rotation	Expected Num.      Zones	99% - 99.99%
Chassis Blueprint (8.33%)	A	44	316±106
Neuroptics Blueprint (7.69%)	B	102	694±232
Systems Blueprint (5.64%)	C	136-144	952±320
Khora Blueprint (5.64%)	C	136-144	952±320

Edited April 27, 2018 by Firetempest

[Raspberry]

@FiretempestThere's not getting Neuroptics, then there's no longer getting Khora anything. I'm a lot closer to those 102 expected runs for the neuroptics, since I've done plenty of 4 rotation runs before starting on my rotation B farm. Hopefully, I won't need to actually run 102+ times.

[taiiat]

one must remember calculating the probability of not winning the lottery is farce math - it isn't applicable to actual situations with high confidence. or, only doing it for academia purposes, and not using it to draw conclusions.

so, if the expected average of a 7.69% Chance is 13 instances - having high enough confidence to really come to a conclusion comes within ~100x that, or ~1300 samples.
projected math that determines what the probability of not being wrong based off of a very small data set is... dubious math that shouldn't really be applied to things in the real world tbh. :/
because using math from the same vein of thinking can say ridiculous things like the probability of not winning the lottery is already nigh infinitely small even with only playing like 100 tickets. but that's just simply not true.

anyways also importantly, is that the weights of the chances aren't listed in the public document so we can't fully trust the numbers as part of the data is being omitted. mostly what that means is there's some variance to expect in the real world vs the numbers on the public document. from what we know in the past it could be upwards of a 6:1 variance, but usually Mission Rewards would be 2:1 if weighted at all.
(see also: lying by omission, a hard written part of the dictionary regardless of whether people like it or not)
which sorta just means use the document as a relative scale more than hardset numbers.

Edited April 27, 2018 by taiiat

[-SoH-ViktorD]

I got 35 hours into the farm so far, got all (multiple times) except for the BP. 4 fcking systems... Really DE? I understand it's a farm, but you should at least give a 2 systems per 1 bp ratio...

[Chrisplaysmore]

Well I got a huge number of Chassis, a single system 0 Neuroptics and 0 BP. I should be close to 100 runs by now ...

I know i am not a lucky person but still!

[Naturesocks]

There might be something with the dropchance, but it's not impossible to get her. I got her last part (neuroptics) yesterday after, so at least they are optainable.

[RedJaeger]

Been farming multiple times each day, 8 rounds per run, ever since the update came out in April. That comes out to well over a hundred runs. I've only managed to get one chassis, and absolutely nothing else. Ever. It sucks.

[Aaxar]

Well, i had the complete set and some, in the first 7 runs.

[MrMrs]

Alright so here's a funny story. I once got the neuroptics if I remember correctly along with the chassis but I was a $&*&*#(%&s and didn't know how to extract from the mission so I didn't get the parts). I have now been reduced to farming over and over to get nothing at all.

Edited August 11, 2018 by MrMrs
lol I misspelled stuff

[-AoN-CanoLathra-]

On 2018-04-26 at 11:12 PM, Firetempest said:

Just been unlucky and it can get worse Drop	Rotation	Expected Num.      Zones	99% - 99.99%
Chassis Blueprint (8.33%)	A	44	316±106
Neuroptics Blueprint (7.69%)	B	102	694±232
Systems Blueprint (5.64%)	C	136-144	952±320
Khora Blueprint (5.64%)	C	136-144	952±320

There is a problem here, though, which is the last two entries in the last column.

If you extrapolate this across the playerbase, approximately 1000 players will have to run Sanctuary Onslaught to C rotation 2544 times just to get Khora.

But, if you extrapolate out one more fraction of a percent, you find that around 100 unlucky players would have to run Sanctuary Onslaught to C rotation over 10,000 times to get Khora.

This illustrates that DE really doesn't have anyone who really understands probability or gambling theory, or if they do, they aren't doing the drop tables. Having something be extremely rare is fine psychologically as long as the opportunities for success happen at frequent intervals and the other rewards are at least entertaining or engaging to some small extent. However, if you fail to do this properly, the human brain will become discouraged quite quickly, because while are brains are horrible about comparing two competing factors (low chance vs high frequency, or high chance vs low frequency), we are overly good at conflating two coinciding factors (low chance and low frequency). The problem, therefore, is not the low drop chance, or the long interval between opportunities, but the combination of the two.