So I wanted to bring some light to the topic. How unlucky are we all?
First of all, it is correct to say that if the drop rate of an item is “p”, then it will take, on average, 1/p runs to see an item the first time.
But the probability of NOT seeing an item after 1/p tries can be quite large, which might be counter-intuitive.
Let’s take p = 0.05 as an example, a 5% drop rate. The expected number of runs is 1/0.05 or 20.
However, the probability of not seeing that item yet is (0.95)^20 ~ 0.36 which is a 36% chance. (Here we had a 95% chance 20 times in a row to NOT get the item).
In fact, if we plot the probability of not seeing a drop within the expected runs, for drop rates between 0.1% and 25%, we see that this probability only increases (up to a limit of 1/e) as the drop rate decreases.
(wolframalpha . com/input/?i=plot+%281-x%29%5E%281%2Fx%29+for+x+in+%280.001%2C+0.25%29)
So what does this mean? Well, if you want to make sure you see the item (just kidding, only with a 95% chance), you could solve for N number of runs:
0.95 >= 1 - (1-p)^⌊N⌋
–> N >= ⌈log(0.05) / log(1-p)⌉
For our example of a 1/20 drop rate, you would need to do 59 runs to see the item with a 95% success rate! That’s a lot more than 20 runs!
For fun, I decided to plot the curve of the number of “extra” runs (over the expected) that you would need to do to get your item (with a 95% chance), for different drop rates. It’s not so pretty!
(wolframalpha . com/input/?i=plot+%28log%280.05%29+%2F+log%281-x%29%29±+%281%2Fx%29+for+x+in+%280.001%2C+0.25%29)