Talk:Humanity (gameplay mechanic)/@comment-5440610-20140830205415/@comment-71.23.225.154-20141119035100

The item discovery stat is not a percentage. We all know this to be true. (Any starting character with 0 humanity shows a 100 point item discovery stat...  if this were equal to 100% item discovery, it would mean that every enemy you defeat would drop something in addition to the souls they add to your total soul count.  Since this is not the case, we know it is not a direct "overall drop percentage".)

In answer to your question about testing game mechanics in order to learn the exact percentage granted, the solution is extremely simple and obvious (though terribly time-consuming):

All you have to do is "farm" enemies for desired drops. For example, defeat slimes in the depths in order to attain lots and lots of green titanite shards and large titanite shards. I'd suggest you start with a character with 0 humanity and have them defeat 100 slimes (without increasing their humanity). If a total of 10 titanite items drop from the control group of slimes, you could assume the drop rate is about 10%.

Next, defeat 100 slimes while your character has 1 humanity (while keeping him/her at exactly 1 humanity). If a total of 12 items drop from this control group, the drop rate is about 12%.

And so on and so forth.

It's important to realize a few things about percentages, though:  If a coin in real life has a 50% chance to come up heads, the following outcome may occur when flipping the coin ten times:

heads, heads, heads, heads, heads, heads, tails, heads, tails, heads

Just because you flipped heads the first time, doesn't mean the second coin flipped is "obligated" to be tails just to balance the math. And, if you get heads the first 3 times, your mind may "want" the next coin-flip to be tails as you watch in disbelief as it's also heads. Try it in real life.

Another way to state this phenomenon is that each coin flip is independent of the others. It has no "memory" of the outcome of the coin flipped before it, or the coin flipped after it. So, in our slime-farming example listed above, you may get only 9 goodies instead of 10 even though the drop rate is 10%. Or, you may even get 13.

Mathematicians know this. So, they don't rely on a test with only 100 instances. Or only 1 test.

Defeating 1000 Slimes would give a more accurate result (than only defeating 100) at any given humanity level; furthermore, defeating 10,000 would give us even more extensive and accurate data.

To further complicate things, different players on different days may have different results. This may be due to a "luck roll" each time you log in. In World of Warcraft, it is said that the game "rolls" for your character's luck when you first log in. This is why some days players report having enemies drop epic after epic after epic item. But other days, it feels like "pulling teeth" to get the enemies to even drop something common. This should be taken into consideration, too.