Data from this project is being compiled here.
I'm interested in the drop rates for various missions. Some of these we know from Japanese sources, and some we don't.
I propose that those of us who are interested start recording the drops we can receive. We can combine our data to estimate the actual drop rates. (It doesn't hurt to try, right?)
The current idea is:
- Drop results can be reported as a comment on this post.
- Before starting the mission, decide whether or not you will record the result and report it here. (This may sound stupid, but it is important to fight sampling bias.) If you came here just to report a string of good/bad/average luck, thanks for your interest, but unfortunately we can't use your data. (But you can still help in the future by following this rule from now on.) (Exceptions may be made for large pre-existing data sets that have followed sound practices.)
- Please only decide to record if you're pretty sure you have a shot at all drops -- preferably you can 3-star the mission. If you miss an enemy that has a drop (or if you're not sure), the result must be discarded. (Theoretically, this could be a small source of bias, but I don't see much that can be done about it.)
- Please report results for all drops on the mission, not just the ones you are interested in.
- Please note any units in your team that have an effect on drop rates including the percentage value of the effect.
- Please note the date of the last maintenance.
- Some missions have one or more guaranteed drops on the first clear. Until it has been established otherwise for the mission in question, please report the first clear results separately. (I'm not sure how this works if the drop is missed on the first clear -- does it carry over? Best be safe and assume so until it is established otherwise.)
- Some missions have special drops that are only available under certain conditions. For example, platinum unit drops in Year End Gold Rush are only available after earning 3 stars. In such cases, please make sure I can tell how many runs had a chance at the special drop (e.g. by saying all but one run was with 3 stars).
- Commission sortie is fine to use. (But you don't have to use it if you don't want to.)
- For missions where the same thing can drop more than once, please record the data in such a way that we can see how often the drop was 0x vs 1x vs 2x vs 3x etc.
- I will add the data to Talk:Drop Rates and calculate estimates of the true drop rate.
Proposed reporting format (still very flexible, but this should be acceptable):
Mission: Phalanx 2 Cuterie: +5% affection gifts Last maintenance: 2016-04-12 Total clears: 8 Flower: 8 Gem: 7 Ruby x1: 5, x2: 1 Gold Spirit: 1 Platinum Spirit : 2
Another good format would be to list the results of each run individually.
Since I think it is a good idea to track the maintenance date with the data, it would make sense to report once a week, except for event missions where after a week it may be too late.
I will accept data about all missions. I'm sure less popular missions will not get many (or any) results, but whatever. Event missions may not be around long enough to get good data, but it doesn't hurt to try.
Discussion and comments are welcome.