Drop Data - Whether Targeting Effects Drop Rates

Okay, it always seems like a point of discussion... whether or not targeting a specific opponent yields better drop rates.  The thinking is that the drops are calculated/determined on a separate "roll" or RNG upon the defeat of each opponent.   Other possibilities include the drops being determined at battle load or even at mission load, in which case your order of attack doesn't matter at all.  Not being a app decompiler, I have no idea what is really there in terms of how it is coded.  So the best I can do is collect data and see what can be drawn from it.

I decided to collect data using the Kuro Event (Elite).  I tested on stage 1, stage 3, and stage 5.  For stage 1, I did my first round of data recording by targeting the 3-star guy in the back.  To compare, I later targeted one of the 2-star guys in the front to see if that would make any difference.  For stage 3, I recorded only when it was the baby turtles, as it gave me 5 units of equal rarity and it seemed I got more baby turtles showing up than big turtles.  For stage 5, unless Django snuck in, it was always 5 units of the equal rarity.  Anyway, here is the data:



As you can see, out of 75 attempts each for stage 1, there was minimal difference between targetting the 3* guy vs the 2* guy.  I am assuming that the 3* guy is of higher rarity.  Not on the chart is how often the 3* guy dropped when attacked a 2* first.  It was 3 times - not meaningfully different.  

For the other stages where all 5 units had equal rarity, there are many ways to look at the data.  But it seems fairly clear to me that there is no statistical benefit for targeting a particular unit.  If the drops are calculated at each defeat, then a drop on the first should happen most often out of the 5.  But this is not the case.  If you take away the times the turtles drop nothing at all (count only rounds with drops), the first turtle hit dropped something (turtle or book) 20.7% of the time.  Basically 1/5.  For the 5 2-star guys, it was a bit higher at 25%.  I'm not trained in statistics and this is still a small sample.  If you take out the "drop nothings" for both of these cases, the rate of drop for the first target is around 23%... not significantly better than 1/5, and probably not significant given the small sample size. Maybe someone who actually studies stats knows how to look at this. But based on my small test, I just don't see targeting as improving chances for drops of a specific unit.

I don't know. I thought I read some days ago that it surely is predetermined ... But I am not sure where it was and wether it was correct.

Greetings

I believe that all drops, for the entire mission set, are determined upon mission load. I could be wrong, but the fact that the game only contacts the server at mission load and mission complete (or if you die) indicates to me that any drops are predetermined.

You can put the phone into airplane mode after starting the mission and still proceed as normal.

Unfortunately that sample is too small o give a definite answer, but people reported quitting mid-battle to reset the stages and they got the exact same drops. So drops should be totall predetermined.

Lol Oreo, you would go that far to research about the drop rates xD honestly, I don't even understand the spreadsheet lol but it's ok. TLDR it is predetermined, not affected with target

Thanks Oreo! That was, what i meant.

Greetings.

Yeah, I totally admit this is small sample size. I've been of the opinion that it is all predetermined for a while. But I figured, I'm doing the runs anyway. So might as well write down a few things since so many people still say that targeting matters. Maybe someone will look at it and use this as one more voice in the "it doesn't matter" camp.

Thanks for the info. I find this very interesting.  Now I have to test this myself.  So far I find a different story.  Here is an example, gold run on extra island during the weekend.  I always target the final boss and he drop gold at least 50% of the time, and if that's predetermined I find it kind of strange.  Going to keep track of it on coming weekend. Maybe 20 runs with him targeted, and 20 runs targeting trash first.

I hope the answer is predetermined

Kudos to you TwoPiece. As already mentioned, the sample is a little small to give any distinct data, but it is a good representation of what we should be expecting.

Personally when I was farming Arlong ch 15 a while ago, I didn't notice any difference whether I finished off Arlong first or the other two, but I can't say for sure since I ended up only getting one Arlong poster and none for Hatchi and Stringray-chan

Well, it certainly is something to ponder upon. Unless another theory comes up which contradicts yours and seems more legit, I'll stick to this one for now

I hope the answer is predetermined so we dont have to bother changing targets and understand the math lol...I guess I'm the only one here who doesn't really get the spreadsheet

I am pretty sure it's predetermined, data is only sent and received upon entering the dungeon. I have not tried this, but if anyone have packet sniffing software, they can figure out the drops rugh away. Come to think of it, this is helpful if you need to gem certain dungeons (30 stamina mihawk?)

If you need to fruit a 30 stamina dungeon you just shouldn't do it since he will came one or two months later

Can confirm the drop is fixed when the level begins. I farmed for rainbow penguin today for at least 30 some tries. (yes I gemed because I want to evolve my vivi asap) I select rainbow penguin as my first target everytime, and it only dropped when I got tired of doing so.

Now I will never bother doing that again.