[needs further investigation] Shiny Chests likely misconfigured

Just throwing a comment in here, July 16 2am pst approx, just pulled 15 shiny keys 1 at a time and pulled 0 Grub Wildshot. Thought that was kind of odd for something that’s supposedly 20%. Not impossible to hit 0, just odd given other people’s current experiences.

I did pull Grub Wildshots from Shiny Dust pulls so I’m not overly hurt by it

Jeto, I don’t know how to word this in a diplomatic way, your understanding of probabilities sucks. :sweat_smile:

You are challenging the very basics of stochastics here, somewhat similar to claiming that 0 isn’t a number. So again, opening 50 chests and getting 5 hits, at a supposedly 20% chance, is the very definition of an outlier. It’s not a highly significant outlier, on its own still well within what could be considered bad luck. However, this wasn’t an isolated incident, and there were outliers that surpassed this one by several magnitudes.

Going with your coin example, if you flip it 100 times and only get heads twice it doesn’t guarantee it’s a trick coin, it would still be unreasonable to believe otherwise.

If you had to open 500 chests to end up past 35 tokens it actually proves beyond any reasonable doubt that something is off. At 20% chance your expected hits is 100, and your expected tokens gained from each hit is 1.65, for a total of 165 tokens on average. Stochastics say you should be reaching 35 tokens at around 106 chests opened, at the peak of the bell curve, with the chance of needing less/more chests dropping off fast the further away you get from 106.

Please don’t get this wrong, I appreciate the effort you are putting into this. However, all observed numbers, especially the ones you provided yourself, serve as very strong indicator that shiny chests are broken. I understand this is a touchy topic, it still needs to get looked at by someone experienced in that area. Is there any chance to take a shortcut here, without having to go back and forth a dozen times, like with similar incidents back when Kafka was in charge? Pretty please?

Perhaps kingdom progression shouldn’t be locked behind shiny troop probability…

If you crunch the numbers and your math actually maths? It’s a lot more common than you’d believe.

Assuming 15 “independent” attempts at the token, with a 20% chance to pull the token? The probability that you don’t pull a single one of those tokens in 15 draws? P = (1 - 0.2) ^ 15 which comes out to roughly 3.5% of the time. Something like 1-in-28 in thereabouts.