Mystery Island RNG Pattern - solved with data and stats tests

[ctar17]

Another data update:

I am strongly leading towards an additional hypothesis:
Having or lacking a personality or species does NOT increase or decrease the chances for that one to appear on the mystery islands.

There is also no data to suggest that the game will place recent villagers you meet on the islands are less likely to appear for a bit. I haven't formally conducted tests, but have seen some back-to-back or very close appearances, so I believe that this is the case for now.

 

[Sheba]

I ended up trading Dom, who was my jock starter, to someone on another site briefly and re-adopting him so he could get his normal house. Maybe someone on here would be willing to cycle them back to you if you want their normal houses? Dom even remembered me when I went to pick him up and he was excited we were going to be neighbors again.

I understand your frustration! The only default personality house I like is the peppy house so I'm not as upset about Audie not getting her own custom house, but it's really annoying when you see what they could be.

ETA: I picked up the exact same Dom I traded so that's why he remembered me (and the friendship level was retained). A random amiibo or mystery island copy villager won't remember you, but if that doesn't bother you, definitely consider it.

OMG they remember you? That's awesome! I wanna do this since I worked so hard for my little guys that I formed an attachment before they have even moved in (lol)! To clarify for me as a non-TTer and who hasn't had someone move out yet: I would have to find someone who TTs and wait until my villager is in boxes. Then they would come to my place and get them, I would wait until the next day in my game while they would TT to get my villager into boxes again so that I could come to their place during the next day to get them to move back to me?

As only Drago moved in with me so far and he's still setting up (everything is still in cardboard boxes), I have no idea how the houses will look like yet. If I end up liking them more than the normal layout for the villager, this is of course moot, but I suspect I might not. Thank you so much for helping me!

This thread has a pic of Fauna's interior. It's honestly up to you if it's worth it for an interior layout change. I do know your pain - Wade (who I'd like to keep) has an amazing ice/fish/shell themed interior normally, but he's stuck with the generic lazy interior since I got him from an early-game island tour. Once the adoption glitches are fixed, I may end up letting him go and trying to adopt him again from someone who could scan his amiibo for me.

As for amiibo villagers, they act the same as a game-spawned villager would.

OMG thank you so much! I love her house, it's adorable! And what kind of adoption glitches are there? Ugh I am so out of the loop with how long I've been stuck on island hopping and not really being able to do much in the game...

I hope @ctar17 is not mad with the thread derailing like that? I'll shut up immediately if it's a bother.

Post automatically merged: Apr 18, 2020

Another data update:

I am strongly leading towards an additional hypothesis:
Having or lacking a personality or species does NOT increase or decrease the chances for that one to appear on the mystery islands.

There is also no data to suggest that the game will place recent villagers you meet on the islands are less likely to appear for a bit. I haven't formally conducted tests, but have seen some back-to-back or very close appearances, so I believe that this is the case for now.

I can confirm that outside of the 82 collected tests yesterday, I have had several more in the two weeks I searched for my bunny where I had back-to-back encounters with some villagers. I am 100% certain of Audie (since it was the first time I saw her and then twice after another) and having had it happen at least two times with Pinky because Pinky just loves me. And then there were the ones yesterday, so I'm fairly certain you are right that the game doesn't remember or care about whom you encountered last.

 

[ctar17]

OMG they remember you? That's awesome! I wanna do this since I worked so hard for my little guys that I formed an attachment before they have even moved in (lol)! To clarify for me as a non-TTer and who hasn't had someone move out yet: I would have to find someone who TTs and wait until my villager is in boxes. Then they would come to my place and get them, I would wait until the next day in my game while they would TT to get my villager into boxes again so that I could come to their place during the next day to get them to move back to me?

As only Drago moved in with me so far and he's still setting up (everything is still in cardboard boxes), I have no idea how the houses will look like yet. If I end up liking them more than the normal layout for the villager, this is of course moot, but I suspect I might not. Thank you so much for helping me!

OMG thank you so much! I love her house, it's adorable! And what kind of adoption glitches are there? Ugh I am so out of the loop with how long I've been stuck on island hopping and not really being able to do much in the game...

I hope @ctar17 is not mad with the thread derailing like that? I'll shut up immediately if it's a bother.

Nope not too mad at all! Happy to help tbh. It also allows the thread to move back up on the site for more to see and find out about the RNG mechanics. I would suggest looking at other threads for your questions first though, as there have been tons about the villager houses and adoption glitches. It would be nice to maybe bring the discussion a little bit back to the mystery islands though so I can answer anyone else's questions they may have about the stats or data. This helps me get more ideas of tests to perform to discover more mechanics and also eventually make an FAQ section. I plan to make a more formal Google Docs with all the info in one place that can be spread around more easily.

 

[Aliya]

OMG they remember you? That's awesome! I wanna do this since I worked so hard for my little guys that I formed an attachment before they have even moved in (lol)! To clarify for me as a non-TTer and who hasn't had someone move out yet: I would have to find someone who TTs and wait until my villager is in boxes. Then they would come to my place and get them, I would wait until the next day in my game while they would TT to get my villager into boxes again so that I could come to their place during the next day to get them to move back to me?

Dom remembered me since it was the exact same Dom who moved out of my island! I only TTed to cycle out villagers in the beginning to get my dreamies and to help others find theirs, but all of my villagers were natural move outs. I no longer TT, but I find that it takes a while to get villagers to move out in this game.

Yes, exactly. The villager plot will open up the day after they move out and you will have to wait until they are able to get your villager back in boxes. You will be able to ask them to move back to your island afterwards, just like any other villager trade. They will then have their normal house and not the default house.

Sorry @ctar17 !

 

[trashpedia]

Thank you for the thread! If this is the case, then that’s kinda annoying. A 1/1000 chance for Raymond? Might as well save my NMT for something else.

Since I have no smug villagers, I think I’ll avoid island hopping when someone moves out next time because as you said, the game is likely to push in a villager with a missing personality.

 

[Sheba]

It would be nice to maybe bring the discussion a little bit back to the mystery islands though so I can answer anyone else's questions they may have about the stats or data.

Got it, thanks!
I actually have a question, but don't know if it's even valid because I am not that far in the game yet and don't have that function unlocked. If it's stupid/impossible, don't mind me: I would like to know if villagers at your campsite count as "residents". Let's say you're hunting for you-know-who and want to raise your chances of getting him, hence you want to remove as many other cats from the pool. And let's say you have Merry's amiibo and invite her to your campsite. Does that remove her from the Mystery Island spawnpool? I have no idea how long villagers stay at the campsite and if there's a difference between amiibo-invited ones and "naturally" appearing ones, but I wondered if the odds can be ever so slightly pushed for those island-hunting for a dreamie.

Also, not sure how related: How soon is island hunting possible again? I know you can have ten villagers max and I've heard that you have to let whoever the first campsite visitor is move into your town, but when do villagers start to move out (only after the 10th has moved in or could you stay at six and get someone to move out)? And does that influence matters somehow? I would guess not if it's really just a species roll and not a personality roll (since you will not have all personalities with six villagers only) but I wondered if maybe there's another "barrier" in place when you don't have ten villagers yet, just like there is the "Lazy/Normal/Peppy"-thing going on for the first three. As in...the game forced you to get Jock/Uchi/Normal/Lazy/Peppy/Smug at the first six. If you can go down to five, does the game still think you're in that "set-personality" stage of the island so that hunting a sixth, new villager will only generate the personality of the initial six that you're missing now?

Dom remembered me since it was the exact same Dom who moved out of my island! I only TTed to cycle out villagers in the beginning to get my dreamies and to help others find theirs, but all of my villagers were natural move outs. I no longer TT, but I find that it takes a while to get villagers to move out in this game.

Yes, exactly. The villager plot will open up the day after they move out and you will have to wait until they are able to get your villager back in boxes. You will be able to ask them to move back to your island afterwards, just like any other villager trade. They will then have their normal house and not the default house.

Not writing about the house thing anymore (I might PM you if I fail to figure this out when the time actually comes for me and I end up confused, if that's okay!), but in general...do I understand correctly that after the initial three houses, I have only one day to invite someone back in or someone random moves into the then-vacant lot? Because obviously, that would make island hunting an absolute pain if it has to be done in basically 24 hours and the game is so unkind and does just roll for species and not for replacing personality.

 

[Heartcore]

Wow, this makes me feel even better about finding Raymond on a mystery island. I'll have to continue to cherish that little guy. Lol.

 

[ctar17]

Got it, thanks!
I actually have a question, but don't know if it's even valid because I am not that far in the game yet and don't have that function unlocked. If it's stupid/impossible, don't mind me: I would like to know if villagers at your campsite count as "residents". Let's say you're hunting for you-know-who and want to raise your chances of getting him, hence you want to remove as many other cats from the pool. And let's say you have Merry's amiibo and invite her to your campsite. Does that remove her from the Mystery Island spawnpool? I have no idea how long villagers stay at the campsite and if there's a difference between amiibo-invited ones and "naturally" appearing ones, but I wondered if the odds can be ever so slightly pushed for those island-hunting for a dreamie.

Also, not sure how related: How soon is island hunting possible again? I know you can have ten villagers max and I've heard that you have to let whoever the first campsite visitor is move into your town, but when do villagers start to move out (only after the 10th has moved in or could you stay at six and get someone to move out)? And does that influence matters somehow? I would guess not if it's really just a species roll and not a personality roll (since you will not have all personalities with six villagers only) but I wondered if maybe there's another "barrier" in place when you don't have ten villagers yet, just like there is the "Lazy/Normal/Peppy"-thing going on for the first three. As in...the game forced you to get Jock/Uchi/Normal/Lazy/Peppy/Smug at the first six. If you can go down to five, does the game still think you're in that "set-personality" stage of the island so that hunting a sixth, new villager will only generate the personality of the initial six that you're missing now?


Not writing about the house thing anymore (I might PM you if I fail to figure this out when the time actually comes for me and I end up confused, if that's okay!), but in general...do I understand correctly that after the initial three houses, I have only one day to invite someone back in or someone random moves into the then-vacant lot? Because obviously, that would make island hunting an absolute pain if it has to be done in basically 24 hours and the game is so unkind and does just roll for species and not for replacing personality.

I do not have the data for mystery islands hunting while there is a campsite visitor, but my gut tells me that you won't find the campsite visitor on mystery islands while they are in your campsite. Once they leave your campsite, I would guess that they go back into the mystery island pool. The random campsite visitors stay for 1 day besides the first one. Island hunting will be possible for you again after you invite your first campsite villager and lay down an empty plot. You must have an empty plot free (not SOLD) to find villagers on the islands. Keep in mind that your very first visitor to the campsite must be invited for the game to progress. They will NOT leave after a day. Also the first campsite villager seems to almost always be of the smug personality.

 

[PajamaCat]

Wow, thank you for the post! Last time I tried to find a villager I wanted, I kept getting eagles which I thought was odd but now it makes so much more sense. I literally saw Amelia twice within five islands! Guess I'll be saving my NMT's for island improvement stuff since it seems the chances of finding a specific villager are so low.

 

[ctar17]

Wow, thank you for the post! Last time I tried to find a villager I wanted, I kept getting eagles which I thought was odd but now it makes so much more sense. I literally saw Amelia twice within five islands! Guess I'll be saving my NMT's for island improvement stuff since it seems the chances of finding a specific villager are so low.

Yeah if you're looking for 1 specific villager, especially one in a species with a lot of villagers, the odds are very low. However, if you're looking for a specific octopus or another species with a low villager count, the odds are higher than average (still less than 1% though, octopuses clock in at 0.9%)

 

[Bk1234]

Just some more of my data (maybe it will help?)
Note: These are NMT Island encounters from AFTER I invited my first campsite villager.

1. Doc
2. Benedict
3. Graham
4. Elmer
5. Sherb (invited)

4 out of the 5 were Lazy.

 

[DubiousDelphine]

I should just cherish raymond. he appeared in my campsite as the 6th villager after all.

Spoiler: pic too big

 

[Bk1234]

@ctar17 I hope I find Caroline on a NMT island, it would remind me of your amazing contributions to the community.

 

[ctar17]

@ctar17 I hope I find Caroline on a NMT island, it would remind me of your amazing contributions to the community.

Awwwww I love you guys! Caroline is my name so it's only fair for me to love her lol. Maybe I'll find her eventually too.
Meanwhile the people in the Facebook groups have been pushing their own theories that are contradictory to what the evidence shows... even when I provide my data, they don't believe it rip.
At least that doesn't happen here so thank you everyone for listening to reason and evidence!

 

[ctar17]

Hey everyone,

If you're now a bit disheartened about the chances of getting a certain business cat villager because it's 1 in a 1000 on the mystery islands pretty much, Calysis has discovered a trick that doesn't involve the mystery islands at all. It has a higher chance than using NMTs.
Here is the thread for it! It involves open plots and random move ins. Check it out!

 

[Lozz]

Hello all,

There is lots of talk about the specific pattern that is used for villagers on mystery islands. I would like to present my findings in this thread and open it up for possible statistics discussions to confirm this theory even more.

First, I'd like to thank Selkie for sending me their data to perform statistical tests on and Sheba and ForbiddenSecrets for bringing up the theory in the first place. The theory is that the game first rolls for a species. Once the species is chosen, it then rolls for a specific villager in that chosen species. The game does NOT roll for a specific villager right off the bat.

Here is the thread where this theory was initially proposed:

[Spreadsheet] Mystery Tour Villager "RNG" (Peppys only. Spoiler: Species gets rolled first, thus odds are not equal)

EDIT: It's clear now! TL;DR: Game rolls species first, then villager if there's more than one in the species, making the odds uneven - less possible villagers in a species mean higher chances for those villagers appearing. Thank you to ForbiddenSecrets for helping me realize this! --- So, as I...

www.belltreeforums.com

The way the tests were done are pretty straightforward. People have been noticing the octopuses appearing a lot more than they should as there's only 3 of them. Therefore, I tested to see if the number of octopuses that actually appeared in the sample was statistically significant than the number of octopuses that we would expect to appear in the sample.

Earlier today, before this theory was brought up, I performed a Chi Square test on Selkie's data to check if it was completely random or pseudo-random or if there was a pattern. Selkie's data has a sample size of 344. The expected chance of an octopus appearing using the old theory of the game randomly rolls a villager from the pool of 391 is 3/391. In the data we would expect one to have about 2.56 appearances in the 344. The data contained 12. Here is the Chi Square test for this model:
View attachment 243901
As you can see, the Chi Square value is larger (much larger) than the p value of 0.05 for a Chi Square test with 1 degree of freedom. This concludes that the observed and expected are indeed statistically different. Therefore, the game does NOT just choose a villager at random out of the 391.

After seeing Sheba's thread and ForbiddenSecret's reply with the species first theory, I decided to go back to Selkie's data and perform another Chi Square test using the parameters that the game chooses a species first at random then a villager in that species. Using this, the expected chance of an octopus is 1/35 not 3/391. In the sample of 344, we would expect the octopuses to have 9.54 appearances. Here is the new Chi Square test:
View attachment 243902
As you can see, the Chi Square value is less than the p value of 0.05 for a 1 degree of freedom test. This means that we cannot reject the null hypothesis of "The game randomly rolls a species of villager first" I conclude that the theory that the species is rolled first then one is selected in that chosen species is basically correct.

What is a Chi Square test?
Basically it tests whether 2 groups of data are statistically different. A common use of it is to test whether your observed set of data is statistically different (not due to random chance) from the expected data.

UPDATE: I have now tested (with the help of more data from TBT users) to see if this theory applies evenly across the board, only 2/35 species didn't uphold the theory, but that is not enough to disprove it all together, and I can attribute it to the nature of RNG. I conclude that it does! The chance for a specific species to be rolled is the same for every species.

UPDATE 2: While I have not formally conducted tests on if the game rolls for personality after species or just rolls straight for a villager yet (still working on this), my initial analysis of the new data provided suggest that this is NOT the case. The game does NOT roll for personality at all (villager 1-5 are locked personalities). I'm 95%+ sure this is the case, but can't outright solidly confirm that there is no personality roll. Nor does lacking a personality on your island increase or decrease the chances for that personality to appear on mystery islands. This also applies to what species you have as well, it won't increase or decrease the chances for a species to appear if you already have 1 or more on your island.

Thanks again to Selkie, Sheba, and ForbiddenSecrets for providing data and/or the theory!

Happy hunting!

Now the big question that a lot of you are wondering...
What does this mean for Raymond hunting?
Well, I'll tell you. The cat is even more elusive than we originally thought! It means that the chance to find Raymond on mystery islands is very low. It's lower than 1/391 because there are 20+ cats in this game. In fact, I have calculated the chance to find Raymond on a mystery island to be about 0.12% Basically 1 in a 1000. Good luck!
Other has pointed out a theory that lacking a smug will cause random move ins, i.e. letting the plot fill up, to be smug. Sounds like this is a much better bet, especially if you're willing to TT to force out the smug to try again!

This user collected all their data as well in this thread: https://www.belltreeforums.com/threads/459-nmt-island-encounters-heres-the-data-i-collected.508602/

Potentially you could use this and see if there is any statistical variance with a larger sample size? They've got a google doc with all the raw data.

 

[ctar17]

This user collected all their data as well in this thread: https://www.belltreeforums.com/threads/459-nmt-island-encounters-heres-the-data-i-collected.508602/

Potentially you could use this and see if there is any statistical variance with a larger sample size? They've got a google doc with all the raw data.

Yeah I'll just add it to the data I've been using if I ever get it. I've already got at least 600 plus and adding more data in hasn't really changed anything so far, so it's safe to say that these theories are most likely correct. The only reason I'm not saying these are confirmed is because I did not mine the data directly and thus cannot find the true formula. But statistics laws (the Law of Large Numbers) states that with a large enough sample size, it is representative of the overall population. I don't expect adding more data in at this point to change anything as the sample size is large enough already. It would help with more complex tests however.

 

[Sheba]

I do not have the data for mystery islands hunting while there is a campsite visitor, but my gut tells me that you won't find the campsite visitor on mystery islands while they are in your campsite. Once they leave your campsite, I would guess that they go back into the mystery island pool. The random campsite visitors stay for 1 day besides the first one. Island hunting will be possible for you again after you invite your first campsite villager and lay down an empty plot. You must have an empty plot free (not SOLD) to find villagers on the islands. Keep in mind that your very first visitor to the campsite must be invited for the game to progress. They will NOT leave after a day. Also the first campsite villager seems to almost always be of the smug personality.

That's what my gut tells me, too. Hope that if someone ever meets an animal that's in the campsite on an island tour as well, they will let us know. And dang, sounds like island hunting will be so time-sensitive then. Ah well... Dotty moved in today, so it'll hopefully be only a few more days until I can get new data for you. (One for Fauna to move in, one for resident services to be upgraded, one for building the campsite, one for inviting the campsite villager and building the next plot, maybe? So four?)
Also, just because you seem to be good at math and I can't figure it out because percentages will be the death of me: What's the chance of encountering Annalisa? There's seven anteaters in the game. I'm in my mid-thirties and I am too stupid for basic math, which is embarrassing, but then again, I sucked at it in school, too...

 

[sicklewillow]

You just have to do 1/35 times 1/7= 0.41%. You have to get pass the first roll though. You have to pray real hard that it roll to anteaters species first.

That's what my gut tells me, too. Hope that if someone ever meets an animal that's in the campsite on an island tour as well, they will let us know. And dang, sounds like island hunting will be so time-sensitive then. Ah well... Dotty moved in today, so it'll hopefully be only a few more days until I can get new data for you. (One for Fauna to move in, one for resident services to be upgraded, one for building the campsite, one for inviting the campsite villager and building the next plot, maybe? So four?)
Also, just because you seem to be good at math and I can't figure it out because percentages will be the death of me: What's the chance of encountering Annalisa? There's seven anteaters in the game. I'm in my mid-thirties and I am too stupid for basic math, which is embarrassing, but then again, I sucked at it in school, too...

 

[Sheba]

You just have to do 1/35 times 1/7= 0.41%. You have to get pass the first roll though. You have to pray real hard that it roll to anteaters species first.

Oooh I see, thank you so much! And that's...almost half a percent, haha. Still not exactly the best of odds, but hey, it could be worse. Not checking the math of my other dreamies, though - that's bound to be much worse and demotivate me.