Talk:Trick-or-Treat Bag/Drop rate
Google's Excel[edit]
May I suggest using that instead of the wiki? Unless you plan on advancing the script a bit so it's easier to add to and update. Ihysoal 03:43, 25 October 2012 (UTC)
I forgot how to indent but here's a quick excel view: puu.sh/1isgl
Multiples[edit]
If I open 1 bag and I get "1 piece of candy corn" and "2 pieces of candy corn", should I tally that as 1 "3 pieces of candy corn" or as both 1 "1 piece" and 1 "2 piece"? I'm assuming the former because the total quantity from one bag is more important than the stack splits, but correct me if I'm wrong. — Burfo
The latter, as far as I'm aware. Ihysoal 03:58, 25 October 2012 (UTC)
- Okay. That seems strange to me but I'll do it that way then. — Burfo
New columns needed[edit]
I opened 368 bags and want to record my findings, but I need columns for some new items added, including the Rotten Egg and Recipe: Major Sigil of the Night (Weaponsmith version). I've never messed around with adding new columns to the drop tables before so I don't want to mess it up. Fallen Angel Eyes 17:33, 25 October 2012 (PDT)
Remove the single candy corn form drop research?[edit]
As you always get 1 piece of candy corn from a bag and one thing of something else, wouldn't it be better to remove the single candy corn drop for the drop research so that the numbers would represent only the second item in the bag?Shol 01:20, 26 October 2012 (PDT)
- I don't think so, because that something else may be a second 1 piece of candy corn. — Burfo 07:20, 26 October 2012 (PDT)
- No, I mean only removing the single candy corn that everybody will get. We can simply calculate this for the previous submitters by subtracting the number of opened bags from the number of single candy corn stacks. I propose this so that the percentages reflect the chance you have for your second item. On a second thought, we could also do this by simply changing the way the percentages are calculated. That would be easier.Shol 08:34, 26 October 2012 (PDT)
Totals that don't add up[edit]
I have bolded the totals that dont't add up. I hope the users are able to correct their inputs, but I recognize this might be impossible. Is there a standard way on how to handle totals that don't add up?--Shol 07:04, 28 October 2012 (PDT)
1250 bags[edit]
I opened 1250 bags but didn't record the individual results so I have no idea how many 1-2-3-4-5 candy corns or 1-3 of the other things I recieved. Here are the totals:
3056 candy corn, 1 Recipe: Major Sigil of the Night, 18 Plastic Spider Tonics, 12 Concentrating Halloween Tonics, 15 Candy Corn Tonics, 3 Halloween Tonics, 21 Bowls of Candy Corn Custard, 27 Pieces of Candy Corn Almond Brittle, 55 Plastic Fangs, 31 Chattering Skulls, 46 Nougat Centers, 16 Glazed Pear Tarts, 5 Glazed Peach Tarts, 26 Strawberry Ghosts, 5 Spicy Pumpkin Cookies, 5 Toilet Paper, 6 Rotten Egg
I converted the candy corn into personalised bags so I'll post their results over there. It's a pretty good sample size for comparing the differences even without the specifics. purple llama 04:56, 30 October 2012 (UTC)
- A guild mate just re-wrote a bit of it for those of us who don't record that (see: batch data) (I had 2k bags opened) Killer Revan 05:35, 30 October 2012 (UTC)
374 bags[edit]
Open 374 Candy bag and obtain (plus the 1 candy) 153*1 candy 112*2candy 9*3 candy 12*4candy 9*5 candy 17*1 nougat center 2*3 nougat center 13*1 chatering skull 3*3 chattering 8*1 plastic fang 5*1 spicy pumpkins cookies 10 Bowl of Candy Corn Custard 3 Glazed Pear Tart 1 Glazed Peach 9 Strawberry Ghost 4 Piece of Candy Corn Almond Brittle 1 rotten egg 1*3 halloween tonic 2*3 plastic spider tonic
If somebody can edit
PS:It's my first time on a wiki, plus I am not English User: Ovnidemon
- I have added this. Shol 18:39, 2 November 2012 (UTC)
Table[edit]
I think the table is pretty messy, plus the code line suggested for use to add the data doesn't match the count of items on the table. I tried redoing it, adding the missing items, but I'm at the point where I'm not sure if we should just start from scratch and re-add the counts from each user or clear it, remake it and then ask them to kindly add their data again. (Being the latter the most reliable way IMO, since people have been using [presumably] a line that doesn't really match the table rows.
Missing elements: Endless Halloween Tonic; Recipe: Superior Rune of the Mad King.
What do you guys think is better to do with it at this point?
Beleeth 22:00, 2 November 2012 (UTC)
- No user will remember what they added before, and I dont think people look at the code line either. They just copy paste the line from the user before, and change their values. I would leave it for now. Especially since we wont see much changes soon as Halloween is over. We could always start fresh next halloween. There might even be a change in drop rate next year if they add new items. However, I don't see the use of the "batch data" table. As far as I'm concerned that can be removed, as it is not very accurate. Shol 22:28, 2 November 2012 (UTC)
- I agree on removing the batch data table. As for the main table, I think i will add the missing items on the last rows, add zeros to everybody's else's line and add my data. Beleeth 23:16, 2 November 2012 (UTC)
The point of the batch data[edit]
So here is the thing why I and others preferred to use a batch data format. Last Halloween, I opened 2304 bags (normal ones, not personalized). At a number this high, it would be incredibly tedious to take note of every single bag you open and check if you get 1, 2, 3, 4, or 5 pieces of candy corn. It’s just not practical to do that; the interface does not even allow you to do that in a good way, given that the items just stack for a long time on the right side of the screen.
So while it may be a bit less usable to get specific statistical information about opening a single bag, this is really the only way you can expect people with a lot of bags to provide the data. And let’s be honest: No one cares about how likely it is to get 1, 2, 3, 4, or 5 candy corn from a single bag. In the end, they would like to know what is the average amount of candy corn you get from a bag. And you can get that pretty well from a “batch” collection. Just that those allow collecting data from a much larger subset and as such will be a lot more precise about the probabilities. poke | talk 14:04, 26 October 2013 (UTC)
- I support this. Some have requested this for any containers that have multiple items, variable counts, and/or especially those with multiple loot tables. However, others have argued that the binary data is more important, i.e. whether something dropped is relevant but how many of each isn't as important.
- Since nothing has changed since before the game was released, I just ignore the wiki data and instead look to Reddit, Guru, and the trading sites for batch data. 75.37.23.146 22:55, 26 October 2013 (UTC)
- After considering these arguments, I think I agree with you guys. The batch data seemed like a big mess to me and some others, which is why I deleted it. However, I now see that the batch data actually provides more relevant data than the precise data. As was stated above, nobody cares how many 1,2,3,4 and 5 stack drops of candy corn are received, but rather how many candy corn you get per bag on average. I'll reverse the last edit I made that removed the batch data, and I'll fix the table format so that it is readable. Poster of Leet (talk) 19:16, 27 October 2013 (UTC)
Halloween 2014[edit]
I move the old drop research to make room for new research. I opened only 10 bags and already got 24 Plastic Fangs, which makes it likely that both the number of fangs one gets and also the probabilities have changed a lot. So we should start fresh. poke | talk 00:05, 22 October 2014 (UTC)
- Yes, the rewards have greatly changed, both in numbers and in mechanics. Each bag contains four item groups, each is randomly selected from the possible items. It’s also possible to get the same item multiple times (e.g. I had a bag with 3 times candy corn, or 2 nougat and 2 plastic fangs). Candy corn is also no longer a guaranteed drop, although still extremely likely. And the numbers have changed too, candy corn goes up to probably 15, and I got up to 8 of the materials at once. poke | talk 00:42, 22 October 2014 (UTC)