The September 2014 Kataclysm is over. As I did after the May 2014 Kataclysm I will now publish my research to benefit the SK community!
I created a spreadsheet to track the various data points from September 2014.
http://i.imgur.com/CUGvZUL.png
I have a second spreadsheet which combines all of the data from May and September.
http://i.imgur.com/Ka8hJSk.png
Here you will see a graph which interprets the Book drop per BK probability curve from my data. It includes the standard error of my sample data--the true probability should be somewhere between the two error curves. There is also a hypothetical curve; it shows what the book drops might look like if the drop rate were 0.5%--the percentage that most people assume is the correct one.
http://i.imgur.com/PBTZqyX.png
- Spookats:Run Alone in the Dark has two more Spookats (on average) than Weight of Darkness. This is likely due to WoD's poison theme; zombies sometimes spawn in lieu of Spookats.
- BK:Run In 423 runs, a total of 101 BKs appeared. Taking into account the confidence interval, the BK-per-Candlestick-Run is somewhere between 21.80% and 25.95%. That comes out to 1:4.6 and 1:3.9 runs.
- This DOES NOT mean you should find a BK every four runs. I myself had 19 runs in a row without finding a BK. This is an average, describing what will happen over time if you make enough runs.
- BK:Spookats There are an average of 20.94 Spookats-per-Candlestick-Run. 101 BKs appeared in 423 runs. Multiplying the average Spookats-per-Candlestick by the number of runs gives an appearance rate of 1:87.7 BK:Spookats. Applying the standard of error gives us a range of 0.62% to 1.66%, or 1:160.3 to 1:60.4.
- This does not take into account themed Spookat spawn rates, only total Spookat spawn rates.I was not aware of the themed spawn requirement when beginning my research. Next Katastrophe's research will focus on the BK-per-themed Spookats to more accurately answer this question.
- Book of Dark Rituals:Book drop chances Each person in the party had a chance of obtaining a Book of Dark Rituals. There were 1389 total Book chances (see methodology) and 5 Books were found. Since it is so difficult to find BKs and the Books are so rare it is hard to get an accurate prediction. Nevertheless we can calculate a standard error from this data--the average Book drop rate is between 0.20% and 0.52%, or 1:501.8 to 1:192.1 Book chances.
- Note: This does not mean that you cannot get a Book outside of this range. I know players who have not received a Book with 400 chances, and someone who received a Book on his second chance. My data is for the average Book drop rate.
- The three contributors (Cheshireccat, Flawedknight, Vtmoon) all received books no higher than 368. If the book drop % is at the upper bound of my standard error that could put the the 90+% book range at 440+. 370+ would be 74%, possibly explaining this phenomenon.
- At our current average, you are 97%+ likely to have earned a book by your 968th Book chance. If you haven't found a Book by then you must have done something to offend the Dark God RNG ("Lord RNG, deliverer of triumph and misery, bless us with your favor and protect us from your ire"). I suggest you visit Punch and make offerings to the Dark God in the hopes he changes his mind.
- Books were found after the party members went solo from the group, on alts, and on alts after rejoining the main party. It appears from these first-hand accounts there are no systems in place to increase or decrease book drops. I FIRMLY believe OOO would not bother building in any such blocks or party trickery to this--it costs dev time and server resources that a small game developer wouldn't waste. See http://forums.spiralknights.com/en/node/104787#comment-948173
- Almost all of the data was recorded on Elite, using the D19 "Alone in the Dark" or D24 "Weight of Darkness" missions.
- Flawedknight contributed the 78 WoD and 265 AitD numbers from 9/14. He ran on Normal.
His result is in line with the Elite data, suggesting that there is no difference between Normal and Elite (though this is not statistically proven). He also contributed one of the books from 9/14.
- Flawedknight contributed the 78 WoD and 265 AitD numbers from 9/14. He ran on Normal.
- There were a varying number of people in each party, but always at least three.
- When a Black Kat was found some members of each party would use the "go solo" trick for extra token drop and Book chances.
- The data from the two missions were kept separate. You can view them on the spreadsheet.
- The data from the two events were kept separate. You can view them on the spreadsheet.
- The Book drop rate and BK appearance rates were calculated with combined data from the two missions from both events. This precludes the assumption that the Book drop chance is equal for both missions (at two different T3 depths) and has not changed between events.
- If you wish to see the data and conclusions for May alone, please refer to my previous Cheshireccat's Classes Black Kat post.
- If you wish to see the data and conclusions for September alone, please refer to the linked image at the top of this post.
- If you wish to see the data and conclusions for May alone, please refer to my previous Cheshireccat's Classes Black Kat post.
- To better understand the math behind the confidence interval and the ranges I provide, feel free to research "binomial proportion confidence interval."
- Mewkats found: 1
- Love Puppies found: 2
- Rock Jellies found: more than 8
- Rare drops from Mewkats/Love Puppies: 0
- Anger after a week of hardcore farming and no book: brobdingnagian.
- Happiness when I found a book: equal to receiving a new puppy. There may have been dancing involved.
- Record more data on themed Spookat numbers in AitD.
- Find more books!!! :D
A heartfelt THANK YOU to FlawedKnight and VTmoon for their contributions to this data!
Thank you as well for reading! I hope this info helps you out!
Until next Kataclysm... see you in Haven!
--Cheshireccat
PS: I highly suggest you save the linked image files. You never know when a third-party host may change something and break the links.
PPS: I like pudding.
Thanks for posting this. I especially like how you included confidence intervals, to get at the uncertainty in the conclusions. Frequently when I talk about collecting data, players say, "Why? We already know the results." The answer is that we need more data to narrow down the uncertainty.