Talk:Mesa/@comment-139.60.142.196-20171214071828/@comment-173.244.48.17-20171214221855

If you do the math from the perspective of getting the set in no particular order as suggested, then the odds for any number three or greater runs should work out as subtracting the sum of probabilities a given part isn't yet acquired for each part, so we do the math for each part 10 times and remove it from 100%. The odds we don't get neuroptics for a given drop is 1.0 - .3872, .6128, or 61.28%  Odds for that 10 times is (1.0-.3872)^10, 61.28% of 61.28% of 61.28% of 61.28%... Chassis is same number. Systems is 1.0 - .2256, .7744, 77.44%. That for 10 runs is (1.0 -.7744)^10. 77.44% of 77.44% of 77.44% of 77.44%... So bothering to punch it into a computer, 1.0 - (.6128)^10 - (.6128)^10 -(.7744)^10 works out to only .9075017. At any given moment your odds are 90.75017% to get all three items at some point in the next 10 runs. And 1.0-(1.0-.744)^8 - (1.0-.6128)^8 - (1.0-.6128)^8 is only .8308907. So 83.0890% in the next 8 runs. Eight to ten runs only gets us all three blueprints 83.08907% to 90.75017% of the time when rounded to 5 decimals. More than 9% of the time 10 runs isn't sufficient.