Mystery Island RNG Pattern - solved with data and stats tests

[sicklewillow]

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.

Don't lose hope because somehow the AC gods have blessed my island hopping last night and was able to get one of my dreamies. Planning to do another island hopping next weekend for my final dreamie. I hope it is as fruitful as the last night.

You will get yours soon!

 

[BluePing]

I’m saving up nmt at the moment and I’ve got just over 30 I think, hopefully I’ll make it to 60 tickets by the time I start going through them ( a few days) just wondering whether you wanted me to post the data on here so you can add it to the data collection. I’ve already started with 5/6 villagers so should be around 65 ish trips

 

[shootingxtar]

I thought it was 1/391 for Raymond and boy were we sadly mistaken LMAO
my bf got godsend lucky then because he found him recently after like 650 tickets and we didn't know it was a 0.12% chance

 

[ctar17]

I did more formal testing to figure out if the game rolled personality after the species roll or not. I found out that the game does NOT roll personality at all.

The tests were conducted by taking a look at all of the species with Uchis, and how many uchis were found in the data. I looked to see if the percentage of Uchis found in villagers with a 8 personalities was 12.5% (1/8) or was it a different number. Using the Chi Square test, I discovered that the actual appearance of the uchis in these species was way under 12.5%, enough to be statistically significant by a lot. I then looked at all of the species with uchis and the Uchi appearances and compared it to the number of uchis total/number of villagers total in species with Uchis. I found that these were not statistically significant. This allows me to formally make the conclusion that the game does NOT roll for a personality at all. It just goes straight from species to villager. The last test's results is consistent with this theory, it makes sense that the percentage of Uchi appearances in species with uchi villagers is not statistically different from the percent of uchi villagers in the species with uchis.

Why did you choose Uchi to test?
When doing my tests to initial figure out if there was a personality roll, in theory each personality would have a 1/8 chance of appearing. 4 out of the 8 species appeared either significantly more or less than 1/8. Cranky, Normal, and Jock all appeared way more, while Uchi appeared way way way less, like half as much as it should have been appearing. Now this could have been due to the fact that not all of the species have an Uchi villager, therefore, after the species roll, sometimes there would be no way to roll an Uchi if a nonUchi species was rolled. So that's why I then just tested the data of species with at least 1 Uchi. I tested species with 7 different personalities, including Uchi, and those with all 8 separately. The 7 personality species test was within the acceptable range, however, the 8 personality species test was way out of the acceptable range, allowing me to conclude there is NOT a personality roll after the initial species one.

Why didn't you test the other personalities too?
Once I found out from the Uchi test there was no personality roll, it didn't matter for the other personalities. The Uchi test alone is enough to disprove the personality roll theory, combined with the fact that 4/8 personalities had a statistical difference in their appearance rate from the 1/8 that would've been had the personality roll been a thing.

What about the different species? Did you see if the species roll first theory applies across the board?
I went back and tested ALL of the species. 30/35 of them fit the test. The ones that didn't were Dog, Duck, Octopus, Kangaroo, and Ostrich with the first 3 of those being just 1 appearance above the acceptable range for the Chi Square test in my sample data. Kangaroo was 1 below the range, and Ostrich was 3 above the range. Since the large majority (almost 85%) of species did fit the test, and it's clear that the personality roll definitely isn't a thing, I conclude that these 5 were just exceptions due to the sample data. They will eventually even out into the acceptable range as more data is collected, that's what the Law of Large Numbers basically states. As sample size grows, it will become more and more representative of the full population.

Tonight or tomorrow, I'm going to finish up my Google Docs with all the information and post the link on this thread. This document should be easier to distribute around and read for anyone who is interested!

If anyone is curious about more things to test for with the mystery island villagers, just drop your idea, and I'll give more thoughts!
Once again, thanks to all who sent data, there is almost 700 island entries so far!

 

[Pfoe]

Thanks for this writeup. I figured this was the case when I did 1100 tickets a couple weeks ago and saw tons of each octopus. It got even more ridiculous when I had Zucker and Marina in my town, which basically forced Octavian into the 1/35 chance and had me seeing him back-to-back multiple times.

To maximize your chances of finding Raymond, it is advised to fill your island with as many cats as possible. If you have 9 cats in your island, the Raymond probability goes from 1-in-805 to 1-in-490.

 

[ctar17]

Thanks for this writeup. I figured this was the case when I did 1100 tickets a couple weeks ago and saw tons of each octopus. It got even more ridiculous when I had Zucker and Marina in my town, which basically forced Octavian into the 1/35 chance and had me seeing him back-to-back multiple times.

To maximize your chances of finding Raymond, it is advised to fill your island with as many cats as possible. If you have 9 cats in your island, the Raymond probability goes from 1-in-805 to 1-in-490.

Yeah the Octopuses really make you think!

Still not too good for Raymond ... The best methods remain the campsite trick and the random move in being forced to smug. Both are done by lacking just a smug in your town. My friend did the campsite hunt last night and found that when lacking a personality, about 50% of campsite spawns turned out to be smug for him. He also got NO repeats in his hunt, and the 27th smug turned out to be Raymond. Took him half a day in real time though, and by the end of it, he had TTed all the way to 2022 in his game lol. I think he only had about 50+ entries though, so it would be interesting to have gotten more data...

His strategy was just be missing a smug. TT until a campsiter, if it's not Raymond, TT 6 days ahead and then day-by-day until a new campsiter, then rinse and repeat. And more importantly, spent only a couple NMTs at the beginning to make sure the slot was filled by a non Smug villager he thought he could later later. (eventually settled on Aurora after about 15 NMTs spent)

 

[ctar17]

ALRIGHT STOP THE PRESSES!

The dataminers have confirmed my theory! Yay us! Thank you TBT for helping this got off the ground enough to be datamined

credit to Aibo for showing me this and all who helped with data collection/coming up with the theory!

 

[sicklewillow]

Congratulations! I mentioned it in the other thread that it was confirmed through data mining but I didn't link the discord discussion in it.

All the data gathering have really paid off!

 

[Saga]

Yay, confirmation!

Now I realize exactly how ridiculously lucky I was to find Tangy on my second NMT... And how impossibly lucky I'll have to be to also find Lolly. At this rate I may end up just buying her on the forums...

I guess the good news it that I have a pretty high chance of finding Zucker since I already have one of the other octopi!

 

[Hyperslurpie]

Not sure if you're still collecting data but I took about 14 trips (which is a small sample size I know haha) but can provide you the data if you need!

 

[cactuscactus]

Okay, I'm going to start this out by saying that I have never taken a statistics class. However I am currently in three university biology classes, and I have performed numerous Chi Square tests in order to determine if I can or cannot reject the null hypothesis based off of the p value in these classes. To my knowledge, and based off everything I've researched and what my professors have told me, when the p value is less than 0.05 in a Chi Square test with 1 degree of freedom that means you can reject the null hypothesis. And when the p value is greater than 0.05 in a test with 1 degree of freedom you can't reject it. So I'm a little confused to your analysis given that you said the exact opposite based off of your p values. Again, I have not taken a stats class so I am not an authority on this by any means, but I'm just confused given everything I've researched and learned has told me the opposite in regards to p values.

I also googled it just to make sure I wasn't talking nonsense and the first website I saw said the same thing. Apologies in advance if my reasoning is wrong, I'm just an animal crossing fan trying to pass my biology classes.

 

[Pfoe]

Okay, I'm going to start this out by saying that I have never taken a statistics class. However I am currently in three university biology classes, and I have performed numerous Chi Square tests in order to determine if I can or cannot reject the null hypothesis based off of the p value in these classes. To my knowledge, and based off everything I've researched and what my professors have told me, when the p value is less than 0.05 in a Chi Square test with 1 degree of freedom that means you can reject the null hypothesis. And when the p value is greater than 0.05 in a test with 1 degree of freedom you can't reject it. So I'm a little confused to your analysis given that you said the exact opposite based off of your p values. Again, I have not taken a stats class so I am not an authority on this by any means, but I'm just confused given everything I've researched and learned has told me the opposite in regards to p values.

I also googled it just to make sure I wasn't talking nonsense and the first website I saw said the same thing. Apologies in advance if my reasoning is wrong, I'm just an animal crossing fan trying to pass my biology classes.

That's a good observation. In this case, the OP has mislabelled the number 3.841 as a p-value, but it's actually just the threshold that produces a p-value of exactly 0.05 (i.e. let X be chi-sq with DF 1, then P(X<3.841)=0.05).

The first test yielded a chi-squared value of 35.01 which is much greater than 3.841 (i.e. p-value much lower than 0.05), while the second test yielded 0.65 which is much lower than 3.841 (i.e. p-value much higher than 0.05).

 

[chainosaur]

about 0.12% Basically 1 in a 1000.

My god, really puts into perspective how lucky I was to find Raymond in only about a dozen tickets.

 

[ctar17]

That's a good observation. In this case, the OP has mislabelled the number 3.841 as a p-value, but it's actually just the threshold that produces a p-value of exactly 0.05 (i.e. let X be chi-sq with DF 1, then P(X<3.841)=0.05).

The first test yielded a chi-squared value of 35.01 which is much greater than 3.841 (i.e. p-value much lower than 0.05), while the second test yielded 0.65 which is much lower than 3.841 (i.e. p-value much higher than 0.05).

Okay, I'm going to start this out by saying that I have never taken a statistics class. However I am currently in three university biology classes, and I have performed numerous Chi Square tests in order to determine if I can or cannot reject the null hypothesis based off of the p value in these classes. To my knowledge, and based off everything I've researched and what my professors have told me, when the p value is less than 0.05 in a Chi Square test with 1 degree of freedom that means you can reject the null hypothesis. And when the p value is greater than 0.05 in a test with 1 degree of freedom you can't reject it. So I'm a little confused to your analysis given that you said the exact opposite based off of your p values. Again, I have not taken a stats class so I am not an authority on this by any means, but I'm just confused given everything I've researched and learned has told me the opposite in regards to p values.

I also googled it just to make sure I wasn't talking nonsense and the first website I saw said the same thing. Apologies in advance if my reasoning is wrong, I'm just an animal crossing fan trying to pass my biology classes.

Yeah both of y'all get the gist of what I'm saying. I just completed my stats class and we were taught that's how to do it. It does appear to be a mislabeling like Pfoe stated, and that's all it is. I did go back and correct the language from "p-value of 0.05" to "value that produces a p-value of 0.05" so it's fixed now and the conclusions of the analysis do not change at all.

Basically for 3.841 and the p-value of 0.05, this is what it technically means:

The Chi Square values meaning:
Above 3.841 (which produces a p-value of 0.05) - means that there is less than a 5% chance for the results to be due to natural random chance. Which means we reject the null hypothesis.
Below 3.841 (again, produces a p-value of 0.05) - means that the chance the results occurred due to natural random chance is above 5%. Which means we cannot reject the null hypothesis.

In my case, 35.01 is much greater than 3.841 and yields a p-value of less than 0.00001. So we reject that null hypothesis.
0.65 is much less than 3.841 and yields a p-value of 0.42. So we cannot reject that null hypothesis (the species-roll first one).

The post is quite simplified compared to an actual stats class (I just finished my university statistics for economics majors class today), but above is the technical meaning of the p-values and Chi square values used in this analysis.

Does this clear it up? Basically had word order switched lol...

 

[Lokidoki]

So if you run into a octopi villager you despawned Raymond... lol

 

[ChrisNewLeaf]

So that means the best time to search for a peppy/lazy/normal dreamie would be with the first 3 plots near the start of the game? Since those are the only ones that are personality locked... I can already see myself spending the first days searching for Beau like crazy Since there are only two lazy deers that means I have better chances than trying to find Bob

If you do this though they will have the default houses and not their Unique houses!!

 

[Kyneria]

If you do this though they will have the default houses and not their Unique houses!!

But since we can ''upgrade'' those houses with the help of friends/ cycling users and more, I don't think it's a bad idea! Though it requires some patience

 

[ChrisNewLeaf]

But since we can ''upgrade'' those houses with the help of friends/ cycling users and more, I don't think it's a bad idea! Though it requires some patience

You mean giving the villager to a friend then getting them back? I didn't know that worked, but still sounds like a hassle! Good to know though

 

[Kyneria]

You mean giving the villager to a friend then getting them back? I didn't know that worked, but still sounds like a hassle! Good to know though

Yep, it works! And the villager remembers you apparently, which means no ''memory resets'' like if you just invite them later through amiibo or find them again in a mistery island. There was a thread about this with some photos for proof, if I remember right the ''experiment'' used a Bam for it

 

[rudythefighter]

Sorry for the bump. Is there any evidence to dismiss the claim that NMTs hold data from the purchaser?

Example: A friend gifts me 10 NMTs which they purchased on their island. Raymond is one of their villagers. Since they have Raymond already, I cannot find Raymond with any of the NMTs I received from their gift.