Drop Rate Analysis and Treasure Hunter

[Enedin]

Pretty sure that about 3 years ago we got information (from SE) that party size increases the possibility to get a drop, and I'm also pretty sure that from my own experiences over the course of playing FF, that it is definently true.

If you've seen the opposite, you're definently playing a game that's not FFXI. Sorry.

Yeah, all this testing has to be done solo.

[Narse]

Uubu,

If we believe them SE already answered this. Once the thf is placed on the hate list for the mob TH is on it.

This is actually not that hard to test and verify, it could be done in a few hours. You can do an alliance with 2 chars, have the thf in one pt and the other pt have some misc job, thief takes first hit and you let the other job finish off the mob. Once we have base TH rates established for one mob very firmly something like this takes almost no time to test. For instance we see that crawlers are ~10% drop rate w/o TH, so go out and do 200-300 kills in this manner (thf can just pull with ranged) and you should be able to establish one way or another.

Thats whats nice about established baselines for a collection of mobs, then testing any other theories about TH superstition becomes a lot easier.

To get a better understanding of when TH effect is placed on the mob and when it's not, you can test with the gilfinder effect too. Kill certain mobs without a THF and see how much gil they drop. Kill it with the Thief and see how much it drops. I think it's safe to assume if a THF successfully places Gilfinder on the mob, he has placed Treasure Hunter on it as well. You can then set up your alliance to have one PT with a THF, and another without a THF, and do your various tests to determine under what exact circumstances Treasure Hunter/Gil Finder are placed onto the mob. Much easier than having to do hundreds of samples under each situation.

[Deadgye]

Just a few things to add:

* I'm 90% certain that /thf th1 is exactly the same as thf/ th1. Off the top of my head I can not think of any job traits that act differently. In fact, I can not think of any job abilities off the top of my head pre-CoP that are different that weren't originally the same.

* Shulula's tests can not be used to confirm or debunk Divisortheory's TH+1 results. Like said before the sample size is just too small for a treasure hunter test. Especially for a test concerning TH3/4 which are, according to DivisorTheory, a 1% increase. The fact that there wasn't even enough kills done to receive a full drop on any set up except the th3 set up is a prime example of why I ignored this test the first time I saw it. The test is also set up in a way to easily read how many items dropped, but not what items dropped. To test against DivisorTheory we would have to rearrange everything into the %drop on the items themselves, not total items dropped.

* Honestly, I think that on drop slots like Strider sword/Kote, and Bahamut staff/lolstaff, th has no effect. I personally think this: TH does nothing but increase the chance for something to drop. On drop slots that have a 100% chance to drop something, th has no effect. Logically how are you supposed to make something drop more than 100%? And if you think SE actually took the time to code which drop is better than the other..

I seriously think that some items have a max drop rate, that TH1/2 are +50% each, that TH3/4 are +1%static, drops like kote/sword, staff/lolstaff, etc are not affected by TH, and that tests/TH are massively skewed on mobs that can drop 2 of the same item.

We need to make a program that records all drop rates on every mob killed, what TH they have(check box) and collects the data until you click the send button at which point all the data is sent to a collective server and the program outputs all the data into a notepad/excel file dump before it wipes itself clean. The only downside is that since we're not entirely sure if number of party members affects drop rates we would have to limit this program to only solo kill data. And unless programmed the right way the users could probably screw it up by sending in data of duo or more kills.

edit: In response to the using gilfinder to test with treasure hunter, does gilfinder even work in parties? I remember a while ago that gilfinder only worked solo, but it could have been changed without me realizing.

[Toksyuryel]

My understanding is that gilfinder works in any situation but only affects the player with the ability.

[Motenten]

Looked over Shulala's data. Took approximated drop rates for wasps from the Wiki: 45% for chips, 17.5% for honey, 7.5% for wings, 2.5% for stingers.

Assuming all drops are fully independantly calculated, worked out the probability of getting 0/1/2/3/4 drops per mob. Result:

0: 40.92%
1: 46.53%
2: 11.69%
3: 0.85%
4: 0.01%

I assume that the drop rates on the wiki are with at least TH1, although they never actually specify, and I don't know if there's any guideline for it. In any case, the theoretical values are not terribly far off from the TH1 and TH2 empirical data from Shulala, and given the low number of samples (and the fact that the original drop rates chosen for my calculations are only estimates), either one is plausible.

In other words, there is nothing in Shulala's data that inherently implies that drop rates are not fully independant per item. Setting them up independantly is pretty simple, so I would say that there needs to be significant evidence that can only be explained by dependant interrelations before pursuing research in that direction. At this point I don't see that evidence, and am thus very skeptical of the "slot" theory for TH's effects.

For reference, assuming that the TH+1 items each add 1% to each item's drop rate, the distribution for number of drops per mob with TH+2 would be:

0: 36.87%
1: 47.24%
2: 14.44%
3: 1.41%
4: 0.04%

In other words, not much of a shift, somewhat comparable to Shulala's results where 0 drops remained fairly stable at 33%.

Enedin's results look very interesting.

[kwontess]

Rather than arguing about what the game actually uses it would be nice if people went and got some data. Most everything at this point is speculation. Lets not see this thread devolve into the other TH thread. The goal should be to collect data and then see. We have a variety of mobs which data is being collected on, but only by a few people. If everyone posting their own theory, their opinions on TH, what they think SE said at one point, etc, etc would just go get 100-200 kills we'd be almost done.

It is hard to say at this point exactly what it is, the reason is the sample sizes are too small and we don't have all the info on some of the tests run. The only way to remedy this is to get people to go collect data. Other issues such as party size, thief in alliance or thief in party and any other piece of odd TH speculation will be much easier to test once we have base rates.

[Ayr]

My understanding is that gilfinder works in any situation but only affects the player with the ability.

I'm thought that Gilfinder affected the party? Can someone verify, because I know years ago I used to duo Ancient Gobbue with a friend. We used to always go Bst/whm and Bst/thf, so we would both get TH1 and some extra gil (taking 10k each to 15k each) and it worked for us for about.... 300-odd kills.

Iunno, it was a long time ago, and Beastmaster usually = alcohol, so my memories of my Boyahda Tree excursions are usually a bit fuzzy, lol.

[Wizerd]

I'm thought that Gilfinder affected the party? Can someone verify, because I know years ago I used to duo Ancient Gobbue with a friend. We used to always go Bst/whm and Bst/thf, so we would both get TH1 and some extra gil (taking 10k each to 15k each) and it worked for us for about.... 300-odd kills.

Iunno, it was a long time ago, and Beastmaster usually = alcohol, so my memories of my Boyahda Tree excursions are usually a bit fuzzy, lol.

Everyone in the party gets the increased gil from my observations.

[Enedin]

Enedin's results look very interesting.

Thanks mate.

Now, just to make sure, I'll post my findings again, the tables are there and I explained how to look at em. I'll post a new one, a little more explaining. If any of you doesn't get it now, I'll gladly explain it again.

http://users.telenet.be/jandebruyn/THTEST0.png

- It seems that either... Families of monsters have a different drop rate correlation (50/50 or 113/37 so far), or that's due to OUTDOOR/INDOOR (I'd REALLY like some testing on this part). TH works, Shulula and me clearly showed that, I just randomly found out that there's a difference, but whether that's due to family, type or area is unknown as of now.

- Full moon does NOT increase drop rates.

[Shalafi]

It could be that TH & TH2 are multiplied separately to create the overall TH added to the mob.

i.e.
(TH + AA + TK) * TH2 = Increased chance applied to the mob

as opposed to:

TH + TH2 + TK + AA = Increased chance applied to the mob

[Cutriss]

We need to make a program that records all drop rates on every mob killed, what TH they have(check box) and collects the data until you click the send button at which point all the data is sent to a collective server and the program outputs all the data into a notepad/excel file dump before it wipes itself clean. The only downside is that since we're not entirely sure if number of party members affects drop rates we would have to limit this program to only solo kill data. And unless programmed the right way the users could probably screw it up by sending in data of duo or more kills.

^ This.

XI needs Thottbot.

[fantasticdan]

Maybe a little off topic, but has anyone ever done a drop rate test of player kill vs. pet kill?

[Motenten]

Test as requested in OP:

Crawlers and Carrion Crows in West Sarutabaruta [S] over 90 minutes. Mnk/Rng, TH0.

Carrion Crow (Killed 100 times)
       10 bird egg                     [Drop Rate:  10.00 %]
       18 bird feather                 [Drop Rate:  18.00 %]

Crawler (Killed 173 times)
       17 spool of silk thread         [Drop Rate:   9.83 %]

Should note that silk thread drop rate mostly varied between 8% and 9%, only pushing 10% near the start and near the end, so is probably closer to 9% overall.

[Dawin]

Should note that silk thread drop rate mostly varied between 8% and 9%, only pushing 10% near the start and near the end, so is probably closer to 9% overall.

This isn't relevant. The most statistically meaningful data is the one with the biggest sample size, which will always be your final result. If this was the only data to go by, we would have to assume that 10% was more likely than 9%.

[kwontess]

More TH2 crawlers

15/100
21/100

That should bring the total to 900. I'll try and get the last 100 tomorrow, this slowly gets more and more boring.

[kwontess]

I decided to go ahead and finish off TH2 crawlers tonight. Final data set is

19/100

[carmancilla]

A lot of good info on this thread and extensive tests too^^

My question is can some items in a monster's loot pool be unaffected by TH at all? I ask this in reference to diorites (lolyuck). From countless hours of farming there are times when you could not get a single one for hours with any TH and I've seen alot of times getting one or even two in less than an hour without.

[Motenten]

Total for my run with TH0 (cumulative with previous post; do not add them together) covering another 5 game days (dark-wind). Remainder is as Mnk/Blm; West Sarutabaruta [S], killing crawlers and crows at the south of the map.

Crawler (Killed 677 times)
       80 spool of silk thread         [Drop Rate:  11.82 %]

Combined with Toksyuryel's total of 54/500, overall total is 134/1177 = 11.4%


Kwontess's TH2 total:
197 / 1000 == 19.7%


Almost forgot:

Data on birds, for reference:

Carrion Crow (Killed 384 times)
       31 bird egg                     [Drop Rate:   8.07 %]
       66 bird feather                 [Drop Rate:  17.19 %]

[Toksyuryel]

TH0 crawlers today: 12/100

sorry for the long delay, starting to get sidetracked by the rest of the game lol

[pchan]

Test as requested in OP:

Crawlers and Carrion Crows in West Sarutabaruta [S] over 90 minutes. Mnk/Rng, TH0.

Carrion Crow (Killed 100 times)
       10 bird egg                     [Drop Rate:  10.00 %]
       18 bird feather                 [Drop Rate:  18.00 %]

Crawler (Killed 173 times)
       17 spool of silk thread         [Drop Rate:   9.83 %]

Should note that silk thread drop rate mostly varied between 8% and 9%, only pushing 10% near the start and near the end, so is probably closer to 9% overall.

You have an error of +/- 3%ish with such a small sample size.