We all are now, spiraled into your head.

Would be nice if you'd kept track of the actual number of Kats in total, to better determine what the current spawn rate is per kat, rather than just a level average.

Still neat though

So... Assuming in Weight of Darkness there is a 20% chance...

So a couple of nights ago, we did some farming of Purr Terror. 15 depths total ( a couple WoD snuck in) results were 1 Bk / 294 total kats.

For the past few days, we did WoD exclusively and got from 36 runs 719 kats, and 14 Bk. There you go. 15 black kats from approx 1000 kats.

Scarlet Fortress doesnt seem to give them as common as I would hope so. I counted them because the same # spawn every time.

Draycos, probability doesn't work like that, doing a 1% chance 100 times does not give you a 100% chance. I can't quite remember how to calculate it though, I think it's the Binomial Coefficient formula. Any help?

If you're referring to the original post --- 19 out of 100 levels sampled had kats, so he estimates the chances of finding a kat in a level at 19% --- then that is correct statistics. But it should also include an estimate of the uncertainty in that "19%". This is sometimes called a margin of error. To compute the uncertainty, read about "confidence interval for proportions".

If you're referring to post #9 --- 1% chance of getting a book from a kat, so you expect to do 100 of them, to find a book --- then that is also correct statistics. The key word here is "expected". It basically means "on average". Some players will find a book after only 50 kats, and some after 215 cats, and some after 105 kats. But the average will be 100 kats (if the chance is really 1%).

Note especially well that "expected" does not mean "guaranteed". There is no guarantee at all. You could do 1,000,000 kats and not find any book. But that is highly improbable.

The correct probability would be for the 100th BK would be 1 - [probability of not finding a book in 100 tries]=1-[.99]^100=.634 or 63.4% chance.

We can argue all day long without getting anywhere, if we don't even agree on what we're trying to compute. Assume for the sake of argument that 1% of black kats drop books. Let's say exactly what we mean:

* The probability of finding a book anytime in your first 100 tries is 63%, as Quartic-Forums just computed.

* The probability of finding a book anytime in your first N tries crosses the 50% mark at N = 69. So you will probably find a book, if you try 69 kats.

* The expected number of tries, to achieve the first success, is 100, as I said. To learn more, read about Bernoulli trials and the geometric model. Or see here.