GamePress

PSA about RNG

I've seen enough people attacked and invalidated on various websites over the past few months I've started using sites like these, that it needs to be said.

I constantly see people say "But RNG man" in regards to how shinyness is determined. But we don't even know the way it's calculated in the game master. Any time anyone even remotely suggests there could be other factors at play, I see them shut down with the same, inane, tired answer that they can't even substantiate.

I'm not saying with any certainty there are other factors at play. But we can't say there aren't either. Unless we get a look at the game master, WE CANNOT CONFIRM THAT RNG IS THE ONLY FACTOR. Saying "b-but muh RNG" with absolute certainty is currently unsubstantiated, unprovable, and intellectually dishonest.

And this is coming from someone with 200+ shinies, so don't go accuse me of being someone who's butthurt that they don't get them, which is usually the go-to.

You may now continue about your day. :)

Asked by TheCaptain5 years 4 months ago

Report

Answers

The thing is, taking the other side (that it is not RNG only) seem to create a worse case to prove, and almost certainly carries the burden of proof.

How stupid do some of these sound to you:
"Shinies favor casuals"
"Shinies are locked to the player and each player has better odds with different species"
"Only 2 people can get a shiny in each raid group, so split your teams!"

It is a very sensible assumption to assume it is indeed RNG, unless there is something so obvious that is missing which TSR/this site would have picked up.

The thing is, I've seen enough patterns (again, even as someone with 200+ shinies) that I also believe maybe RNG isn't the only factor at play.

I can't prove it, but you also can't prove it's just simple RNG. Using "but RNG" as a go-to response is void of any actual substance, and until we see the game master, it's misleading.

And I've seen enough people attacked over it by people saying it with literal 100% certainty, that I simply cannot sanction it any further.

The Pokemon Go community would be infinitely less toxic if rather than passing speculation as fact and attacking people for it (implying one is just butthurt because they didn't get a shiny, for example), they simply said "we don't know, man" about things that we legitimately know nothing about.

There may be factors we don't know about that influence RNG in some cases e.g. which biome an egg was collected in, but generally RNG explains every aspect. The mass research projects on Silph Road which give a large enough sample to be statistically pretty good have yet to support some of the wilder theories. Most of the 'I'm not getting my fair share' posts do tend to show a lack of understanding of what random means.

I know what you are saying, and agree...mostly.

There has not and I can't imagine there will ever be a way to test a large enough sample group of accounts that have not logged on for months, to assess improved odds.
This type of study is near impossible to get accurate data for, without accounts created specifically for this purpose. So that is a long term project, if at all.

So we are left with anecdotal evidence. Enough anecdotal evidence can at least be enough to reasonably question factors.

Basically, even TSR would conclude they don't have enough evidence to disprove some/many of these theoretical factors.

That said, once these factors are plugged in, then there will still be a RNG to apply these (possibly variable) odds.

Don't disregard the toxicity created by peddling false information.

It doesn't make sense for the proposed additional interactions to alter shiny rates etc. I implore you to read up some of TSR's research, where they provide their statistical analyses of shiny and other rates.

But it does make sense to me. Again, I can't prove it. But again, until we see the game master, or have an official Niantic employee (ie: not someone outsourced from customer service) testify that it is purely a random number generator, we can't prove that RNG is the only factor either. It is all speculation from both sides without any verifiable data. There is speculation that allows for rough theories, but nothing concrete.

I implore you to consider that the accounts used for these studies would likely be regularly used accounts that log on almost daily, or are brand new accounts created for such a study.

Using that data, can you explain how it disproves that old accounts that log in after a long time away from the game, may possibly have improved shiny rates?

Sure, the data is awesome, and props to TSR teams...but even they would agree it can't be used to dismiss all other theories of variable factors in shiny rates.

It's very simple. Ask around or just read some related topics from others here or on other sites and you will figure out that these "patterns" that you have found are absolutely not universal. The same goes for the alleged patterns that other players have allegedly found. It happens in absolutely every game in which any degree of RNG is involved and these pointless discussions arise all the time because humans have been evolutionarily designed to seek patterns, even when there are none (look into pareidolia for a real life example of this). You may take these logical reprimands as hostile or toxic, but in reality it is even more harmful to introduce any kind of speculation about the hidden game mechanics into the zeigeist without a modicum of proof, hence why so many game veterans shun these assumptions on the spot when absolutely no data is presented to support it; whereas the totality of all metadata gathered ever since the game was introduced has led to the general conclusion that there are no predetermined measurable factors that would skew the assumption that a random number generator or a combination of them is the driving engine behind all of the randomized events taking place specifically for each player. Anyone with a differing theory has the responsibility of providing irrefutable proof that would show that these are definetely not random and, had it happen, the community would have accepted those findings already because it would mean that these "patterns" are replicable, so it would be easy to simulate them and predict results with great accuracy. After 3 years and millions of dedicated players, many with deep inclinations towards the thorough study of the game and its mechanics, this has no yet happened and the prevalent RNG theory has not been close to being replaced with an alternative one based on individual patterns. It is simply baseless. We may not "know" what the exact mechanisms are, but we do know how to analyze data and discriminate coincidences unlike the vast majority of players that you are championing for. What you and all of them have been experiencing is confirmation bias, and just little research on the subject would alter your view on this topic. And if it doesn't, well good luck to you disproving the majority consensus, I hope you have a great argument in your favor because that would make you a legend on the SilphRoad and GP, and millions of us would be very grateful that someone has finally cracked the code. Until then, just understand that speculation can be treated as fact if it is the most likely option 99% of the time, which is the case here.
Happy holidays and don't be so uptight ; )

It cannot be treated as fact. We have rough theories but nothing substantiative or definitive.

Just to correct a simple misconception you seem to be having: We have full access to the game master, it's a file loaded to every phone that has pogo installed. That's where a lot of the data that you find on TSR and here comes from. But the game master has absolutely nothing to do with shiny rates, IV randomization and many other things, those are determined server side.

What you want to say is "until we see what happens on Niantic's server".

The odds may change, but RNG will stay the same.

Sure, there may be factors in play that we are not totally aware of that affect the odds in some way (what you are saying), but once the factors are considered, for a possible variation in odds, THEN 'RNG' will use these factors to determine outcome.

Again, all I'm saying is that we don't know how shininess is determined. It could be strictly RNG. There could be other factors. We don't know. All I'm saying is using "Its just RNG" as a way to "disprove" someone claiming there could be other factors is speculative, kind of toxic, and frankly just as unsubstantiated as the point it's trying to rebut.

I totally agree with the sentiment.

I was just being pedantic...pointing out that variable odds (theoretical or real) doesn't stop RNG coming in to play...
...and vice versa, saying 'it's RNG' means nothing and is irrelevant to a discussion about factors that may influence the odds.

First, and this is from a very technical way of speaking, it's not RNG,its PRNG (pseudo-random), so yes, there is or are other factors in the equation: anything that has to do with the seed of the RNG algorithm is that factor you mention, so yes, technically is not "pure RNG".
Second, while I agree at some degree with what you say, you have to admit that someone saying "but RNG" as its sole argument is as silly as someone stating "it's not RNG and you cannot disprove me unless you work at Niantic". In the end, I think we will indeed never know "the truth", and I can only hope that all those people mad at Niantic because they're somehow being "unfairly unfavored" understand that this game has never been about showing and proving that you're best player that someone else; it is all about social interactions.

And ultimately, if there's indeed some other factor that affect the way things are distributed and obtained in-game, and that factor is ever publicly known, I assure you we will have DESTRUCTION, DEVASTATION AND DEMOLITION, so perhaps we're better right as we are.

All we can really say is 'there is no substantive evidence that anything other than RNG affects different shiny rates between players' until someone provides significant statistical analysis to prove otherwise.

No, it is very clear that each account has shinies that they can see more frequently while others cannot be seen at all from them. I am very sure that there is a lock placed on some pokemon on various accounts.

Same goes with raid boss ivs.

Where's you proof? I can and will absolutely disprove every assumption of yours that you can use from personal experience by providing a counter argument against it from my own personal experience. That is the nature of declaring that it is "very clear" that something like shiny rate encounters or IVS is locked to each account. My account has no such lock and I can prove it. But maybe you'll then claim that it is just some accounts do not have any locks or some bs like that.
You are just either insurmountably unlucky or, most likely, an indisputable pessimist with zero tolerance for failure and terribly skewed priorities.

There is a rule of thumb: "the simplest explanation is usually the best". I think pure RNG is the simplest explanation. There is also a human tendency to see patterns in random data that don't really exist. So, I think RNG skeptics need some very credible evidence to overcome those two objections, before I'll believe in "other factors".

Why is the onus on the skeptics? We have no idea how the game functions. That is all I am saying.

My stance (that we don't know if RNG is the sole factor) can be substantiated, as we don't know that until we see the algorithms. Those saying "its just RNG" can't substantiate anything. There is rough theoretical data and that's all. Nothing definitive.

What does it mean RNG??

Random number generator

But we can, indeed! People already run statistical analysis for most (if not all) of pogo features and they already ruled out all the variables that might influence. Such as weather, distance, time of the day, the moon phase, etc.

So, there are plenty of studies that substantiate the answer. RNG is indeed the right answer here.

No it's not. It's theoretical. Strictly a hypothesis.

We don't know how shininess is determined. That is all I am saying. And again, I can substantiate my claim as we have not seen the relevant code. You cannot.

Theoretical data is not hard data.

Additional factors are unlikely simply because it takes effort to program them in to the game. It'd be incredibly inefficient for companies to waste resources on creating convoluted mechanics if they can't reasonably assume that those mechanics create a lot of value. This means that most of the conspiracy theories people come up with about shiny rates are probably just RNG. Money is the largest factor governing these things.

That does still leave a few where Niantic might have a an incentive to alter the shiny rates, like returning players. But any theory needs to pass the simple check of "why would Niantic go through the effort of programming this?" That still leaves bugs and bad use of pseudo-RNG that could explain some unintended patterns though.

It's not that it's 100% guaranteed to be about RNG, it's just that RNG is the overwhelmingly most likely answer to anything. There's no intellectual dishonesty in this. The burden of proof is on the person saying there is a pattern, and if they can't demonstrate proof, there is no reason to believe in their hypothesis.

Also, a note about statistics: Sample size is everything. 200+ shinies is a tiny sample size, practically irrelevant. You need tens of thousands of data points to have any shot at seeing patterns that can't be attributed to "just RNG".

Again: theoretical data is not hard data.

Until we see the relevant code, saying "its just RNG" with 100% certainty is not substantiative. Thats all I'm saying.

Honestly did you read my reply at all? Where did I say that RNG is 100% the answer? You claim that bringing up RNG is intellectually dishonest, and I heavily disagree, because RNG is the most plausible answer every time somebody comes up with a theory based on their own anecdotal evidence with a pitifully small sample size. Which is the majority of theories that people post on this site.

I don't understand what you mean by theoretical or hard data. There is no such thing as theoretical data, it's not data if it's theoretical. Data is data and it either exists or it doesn't. Currently TSR researchers have some data on things governed by (presumably) some sort of RNG in the game and they've made educated guesses about what that data means by using statistical analysis.

I don't know if you took any statistics/econometrics class in college but my professor gave me this wonderful story:

Thinking back to world war 2, the Americans tested new armor for their planes to be used in air raids against Japan. After the raid was complete, half of the planes returned.

The general saw the damage inflicted to the remaining aircraft. The engineer rushed over, and exclaimed, "Woah! Look at those holes everywhere in the center of the fuselage! Should we reinforce the armour there?"

The general replied, "No. Reinforce the places where there are no bullet holes".

"But Why?"

"That's the reason why we lost the other half of the planes".

TL;DR

There is a huge amount of data in TSR. There is also a huge amount of data that is not there. This type of study is very prone to selection bias and myself and many others have not participated.

Latest Content

Typing Review: An Overview of Poison Types for Raids in Pokemon Go
2024-04-27

Rivals Week & Go Battle Weekend
2024-04-25

Necrozma Confirmed for Pokemon Go Fest 2024
2024-04-11

Bug Out 2024
2024-04-08

Mega Heracross Raid Day Guide
2024-04-05

Mega Heracross Raid Day
2024-04-05

An Excellent Opportunity - Good and Great Throws count as Excellent fo…
2024-03-31

Typing Review: An Overview of Normal Types for Raids in Pokemon Go
2024-03-24