Statistics Notes

Disclaimer

Remember: when viewing statistics on this website, they may not be totally accurate. The contributors to the wiki have done significant research (in most cases) before publishing statistics here, but there is nonetheless always the possibility for error.
Jick had stated that he usually codes drop values in multiples of five in the forums, but drop rates are kept to the nearest 0.1%, just in case. (Jick also said on a radio show in February of 2007 that this was in fact not correct, that item drop rates were only kept as integers.)
Information on specific statistical wiki-centered projects follows.

Yiab's Item Hunt

The main contributor to item drop statistics on this site was Yiab who used to ensure that the statistics were based on at least 1000 encounters with a given monster before publishing them. All item drop rates are rounded to the nearest 0.1% chance. If you wish to contact Yiab about this project, feel free. Data with only minor preprocessing is (was?) available upon request.
- NOTE: Due to the difficulty of automating combat in the Crimbo Town Toy Factory, only about 1/3 of the reliability can be attributed to the statistics that Yiab has gathered there.
A page has been written specifically to present condensed item drop statistics on a single page, in a format easily readable by computer. It is called Parseable Item Statistics. There is a similar page for the data concerning frequencies of adventures in areas, called Parseable Area Statistics.
Meat drops gathered during this project may contain the occasional error, as the primary goal of the project was to find out more about item drop rates. Meat drops appear to follow a triangular distribution. Similar to the ones mentioned above, there is a page summarizing meat drop data to be read by a computer, at Parseable Meat Statistics.
The following are approximate dates on which the statistics were calculated. If there has been a change in an area since that date, the statistics for that area may be completely invalidated.

Amateur Statistics

Some statistics may say: "Approx. X% Chance".
These are used for rough figures until Yiab posts more reliable data.

Drop-capping method

With power-creep over the years, capping drops has become more and more feasible, which a spade can use to their advantage. This method assumes that item drop rates really are always integer percent values as suggested in that Feb 07 radio show, and that the drop is either unconditional or has a condition that you can guarantee is true the entire session.

If you think the drop rate is, say, 20%, you should run a few dozen adventures with just enough bonus Item Drops from Monsters to cap a 21% drop rate. To calculate this:

(100 / 21) - 1 = 3.76
(100 / 20) - 1 = 4.00
(100 / 19) - 1 = 4.26

So you should aim for at least +376% item, but less than 400%. Adventure until the target enemy fails to drop the target item -- this will confirm that the drop rate is definitely less than 21%. Then, boost your item drops to at least 400%, but less than 426%. Adventure until either the item fails to drop (proving that the base rate is actually even lower) or you've spent an absolutely preposterous number of adventures on this. More adventures = more confidence!

After 100 wins and 100 drops at exactly +400% item drops, the probability that the drop rate is actually 19% is:

(.19 * 5.00 ) ^ 100 = 0.006

A small number, to be sure. Spend 50 more adventures, though, and...

(.19 * 5.00) ^ 150 = 0.0005

A whole order of magnitude more confidence! Is it enough? That's up to you, but know that you can't ever be 100% confident that you've really capped the drop rather than the RNG screwing with you -- but with enough adventures you can get as close to 100% confident as you like. The usual standard is 95% confidence, though you can certainly go higher if you have nothing better to do with your adventures.

Live. Ascend. Repeat.

The Fall 2017 challenge path has made spading drop rates below 50% much simpler. If (base rate * drop modifier) > 50, the item will drop every time. If not, it will never drop. RNG is no longer a factor.

Base Drop %	+Item% Needed To Guarantee Drop	Examples
5%	+900%	blasting soda
10%	+400%	filthworm scent glands
15%	+234%	A-Boo clues, rusty hedge trimmers
20%	+150%	tomb ratchet, harem pants, harem veil
25%	+100%	bird ribs, lion oil
30%	+67%	stars, lines, hot wings
35%	+43%
40%	+25%	goat cheese, bowling balls
45%	+12%
50+%	-	bubblin' crude, Gnollish toolbox, tires

This can still be used to spade drop rates above 50%, but there are fewer reliable methods of reducing one's drop rate, and much less room to work with.

Statistics Notes

Contents

Disclaimer

Yiab's Item Hunt

Below This Line is Post-NS-13 Data

Below This Line is Pre-NS-13 Data

Statistics from Others

Amateur Statistics

Drop-capping method

Live. Ascend. Repeat.

Navigation menu

Personal tools

Namespaces

Variants

Views

More

Search

Navigation

KoL Coldfront

Well-Tempered Tools

Tools