Using Math to Settle Pokémon's Biggest Debate
Вставка
- Опубліковано 19 жов 2024
- Go to mondly.app/adef to get 96% off of lifetime access to 41 languages and start learning today.
Support the show on Patreon: / adefgames
2nd channel that has full VODs of my streams: @adefVODs
My twitch: / adef
Huge thank you to @AbsolBlogsPokemon for helping me out with the research for this video and providing so much insight on this wonderful community! You can watch his streams at: / absoltastic
Thank you to these wonderful folks for also lending their clips:
@partyarlie
@JohnstoneYT
@AjentVee
/ norainthefuture
Intro music by @DonutShoesdonutshoes.wix...
#pokemon #science #maths
Go to mondly.app/adef to get 96% off of lifetime access to 41 languages and start learning today.
Hope you enjoyed this one, gamers! I'm always nervous to do *anything* even tangentially related to **drama** because I truly think youtubers that just cover drama / cheating scandals are the scum of the earth, but I hope that this one brings more positive eyes to shiny hunting than negative ones!!
I enjoyed the ad read being almost entirely in french, I thought that was a creative bit, never seen one like that before
duolingo on top
Using this one as a quick warning that’s gonna be more visible: the lights on the Eiffel Tower are copyrighted. Yes, specifically the lights. Granted that’s never been enforced, ever, but the copyright on the artistic expression of lightbulbs on metal expires in 2055.
Is it accurate to say that your odds of hitting *within a certain time period* increase in a multi-game? Or does formulating an extreme case of this scenario also lead to absurdity? That is, if I played enough games would the math say I *had* to hit within one hour, making the premise unsound? Thanks for the vid!
Man, August the Duck is being FRAMED with this one!
But I agree, covering drama is bad, but CAN be usefull if the ytuber that does it is extremely Objective.
Thank you so much for being the confident Pokemon Math Guy we needed to break this down!!
Omg its absolblogs
thank you absol
matt jumpscare
hi absol it's me metc
KING
Mom, can we have 2/8192?
Mom: no, we have 2/8192 at home
2/8292 at home:
1 - (8191/8192)^2
hehe, 99.99% as good at home, mom might be right here
2/8292 is worse than full odds, glad i dont have that one at home 😂
You are the only one here that understand what's happening and why the video is wrong lol.
@@gabrielbarrantes6946 why? Genuine question
@@gabrielbarrantes6946 how is the video wrong
I was expecting 1 billion lions vs 1 of every pokemon
Same
Yeah me too.
Well its obviously the pokemon
That was already settled by astroid videos
this was solved by Asteroid Videos
I think this distinction is actually more of a language issue than a statistics issue at heart. People who have basic knowledge of statistics would say that playing more than one game at once gives you better odds, but what they're truly saying is that it gives you better odds over time, not per encounter. This is inherently understood when you're talking to other people with similar knowledge levels, but doesn't truly reflect the results and people who aren't experienced with statistics will have their assumptions reinforced.
I think he missed something when he talked about whether time spent shiny hunting was more important or number of encounters. To the average player, time spent is immeasurably more important, but for UA-camrs, number of encounters is more appealing.
100% agreed. We all know what people mean when they say more games = faster encounters = better shiny ''odds''. I really don't understand why the commentary at 4:20 was needed. It's like he's casting unnecessary doubt over the (correct) assumption that more games = more shiny encounters. What afdef actually disagrees with is the definition of the word odds in this context.
All in all, for a video that's supposed to clear things up, this was rather poorly done imo. At multiple points he reiterates that more games don't increase the odds, but that's not at all important in the context of pokemon shiny hunting, where time is easily the most important factor.
It's like what I always say you can know enough to believe you are right but not know that your wrong and make it make sense to others at the same time
@@ScyrousFX you're boosting your encounter rates, not your shiny rates, how the fuck would it not be a full odds hunt?
Yeah, he's arguing semantics. The whole video seems to be in poor taste too. Some full odds shiny hunters are already elitist about it, the last thing we need is more to fuel their egos.
Most people care about the speed at which you find shinies, not so much the total encounter numbers. That's more of a content creator thing they use to give an idea of how long it took. All of which is valid ofc, but not the point of the "multi hunting is method hunting" argument. He either missed the point or ignored it on purpose.
It was in fact, a good bit
i agree!!!
He a real one for that
The sponsorship segment was a fantastic bit.
Yes the accent was pretty good 😂
yes, yes it was, also great transition in particular, using the Masuda method language thing into the language app bit, just great
Bien fait
Nice try
As a French person, I was listening aside and when I heared my language, I went "Wait wtf" and looked back to the video xd
4:02 To answer the question: That was the first ad I didn't skip, because it was so unexpected. So I guess it fulfilled its purpose.
I guess someone beat me to the punch. Danm, I'm late.
If you are mad on the internet about a stranger being more efficient than you, no one can help you.
fr. it actually blows my mind that there can be community drama over how people go about looking for shinies. like, sometime after the release of scarlet and violet there were people getting upset over whether or not someone matches the pokeball colors with the shiny colors (I AM NOT MAKING THIS UP). if you have any sense of elitism or feel superior over the way you look for differently colored pixels on a screen you desperately need to get a life.
@@yepmultiI mean with how easy shiny hunting is in SV I can somewhat see what they mean
@@HuneeBruh literally who cares
@@yepmulticlearly the people that bitch about it care. I think it'd be great if we all could respect other people's opinions and way of doing things without the hostility, but it's inevitable someone gets butthurt over a string of words a stranger types anonymously. Best thing we can do is take it as a joke and move on. Impossible to tell who's serious and who isn't most of the time thru text 😅
Not being in the shiny hunting community at all, be it hunter or content watcher, it seems to me from the few comments shown in the beginning of the video that people are displeased by the way this method of hunting is called rather than the method itself.
Which I honestly get, as the method basically boils down to "spend more money", it probably shouldn't be labeled in the same way as the default one.
A comparison I like is simply rolling dice.
If your goal is to roll a 1 on a 6 sided die then you have a 1 in 6 chance.
If you have only 1 die and roll it 10 times, then your odds for getting 10 1's is identical to a person who rolled 2 dice 5 times.
In the end you rolled 10 dice either way. The person with 2 dice could just do it faster.
Edit: Changing the desired number from 6 to 1 for a better example.
Using two copies of the game at the same time improves speed of encounters/rolls, but not the probably for any roll to be a 1.
Using a D4 instead of a D6 though would not improve the speed of rolls/encounters, but does improve the probability of any roll being a 1.
I think all D&D players can confirm that rolling 6 dice does not guarantee rolling a 6.
Ok but that's a bit silly, like, if you roll more dice everytime then your odds per dice isn't increased but your odds to get a 6 in a given period of time is increased.
The argument would make sense if there was a cost associated with each dice roll and that cost was what you compared your odds to (like say you have to pay a fee for each roll, so rolling multiple at once doesn't help you make more money in your gambling endeavours).
In this case the only cost is time so that's what you should be comparing the odds with.
@@Laezar1 There's a reason I specified rolling only 10 dice total each time. The goal with this example of probability is exactly that YES, you're likely to take less personal TIME to get to the final results, but it does not improve the possibility of getting that desired 6.
Like the video says. Each copy of Firered and a Gameboy you have will generally improve the speed of which you'll find a shiny, but it does not improve the odds. Adding more dice does the same thing, it improves the speed not because your odds are better but because you're getting the results faster.
If you were shiny hunting with a limit of 10 encounters then it doesn't matter if you are using 1 copy of the game or 10, your odds are the same but your speed is not.
Compared to if you hatched 5 masuda method eggs 1 at a time to hatching 5 random eggs all at once. Hatching the random eggs would be faster to do, but getting a shiny from the masuda eggs would be more likely with a slower speed.
@@Laezar1 "your odds per dice isn't increased but your odds to get a 6 in a given period of time is increased"
That's precisely what this video is about, so the analogy is apt.
@@starfishman1000this reasoning is flawed, because we are analyzing a stopping time, the case you are using as example is something different... If you look for the first 1 you have more probabilities using two dices than just one in each toss.
TLDW version:
A method reduces the expected number of encounters.
Multi-game reduces the expected time, but it doesn't change the expected number of encounters.
I wonder why we care about the number of encounters, time is much more important to me.
Real though
Probably because we dont have a standard unit time measure for each encounter
Because its easier for people to mentally gauge how long a single encounter takes; it’s easier for people to make the mental approximated association than interconvert
probably because people lose track of time. These hunts can take weeks..
If people decide to sit down and eat or shiny hunt really sedately, or their games load slow or fast, their times will be different. "Took me 100 hours" can mean very different things to different people but we know immediately that 10 encounters or 100,000 encounters is something special/crazy.
How would I measure my time. Like, I am currently doing full odds hunt around 9000 encounters deep, some of the hunt has been done with full focus on the encounters, and it has been fairly fast, but most of the time I've been hanging with friends or family, doing other stuff, which significantly slows down the encounters, like maybe up to 3 times slower. Is the time spent on both instances valued equally, or should I add modifiers to adjust the time when I am more distracted. Or maybe it's the overall time, in which case I am around 9 months deep.
I am not saying the time does not matter, but it does get murky really fast. Encounter numbers are easy, I can easily check my counter and tell you the exact number of encounters is 9033, and sure there might be a slight mistakes there as well, encounters when I didn't remember to increase the counter, but the error margin there is within couple of % at most, but with time depending on how you count it could be 20 hours or could be 6500 hours
As a lot of commenters here have already pointed out, I also do think its disingenuous extrapolating the point to "per encounter" or "odds". If the point of 'increasing the odds' is to accelerate the process, aka shortcuts, playing multiple games also accelerate the process, therefore making you accomplish the goal (finding a shiny) faster.
btw great sponsor segment
very well said at the end, and i think it extrapolates even further into stuff like RNG manipulation... i see so much discourse over what counts as "pure" and "proper" shiny hunting and in reality, we do this thing for fun! do it how you want and do what makes you happy. using your knowledge of the game to hunt for the thing you want is pretty rad.
Yes, but using a system to make the hunt easier and then tell everyone else you didn't use said system is not fair on everyone else. The important part is people being honest in the acheivements.
hi shayy! :DDD
If you do 8192 rolls with a 1/8192 chance each, the probability of at least one roll being shiny is not 100%. But there *is* a probability of at least one roll being shiny, and you can calculate it. 1 - (1 - 1 / 8192)^8192 ≈ 0.632, or a 63.2% chance.
@@comexk Math fact! That value is approximately 1-(1/e) and the probability gets closer and closer to that as the numbers in the place of 8192 grow.
He made a video explaining that a few months ago. To be honest the last bit explaining that the counter adds every encounter individually summarizes why it’s not method hunting, everything else he goes on much more detail on that video, but we always love to see more from our fav pokemon math guy :)
Thank you! This is the point that I wanted to make but didn't want to figure out the math to make the point
And the average number of shinys you'll get per 'round' of encounters is: 1. It's not 100% you'll get a shiny, but you'll also often get multiple.
@andijacobsen9148 the average would round up to one, if you had to round to the nearest whole number. But computing a mathematical average would give you 0.63 shinies per round of encounters
DID THE LIONS WIN?!?!??
Oh…
Me (a person from the Detroit area) on January 29th, 2024:
L COWGIRLS W LIONS RAHHHHHH
Forward down the field
The lions don’t have a ghost of a chance I don’t understand why anyone thinks it’s even debatable.
As a French person I have to say you did a great job for the sponsorship, you spoke really well :)
Im sorry to hear that
merci!
My condolences
@@HuneeBruh haha, for a foreigner he did a great job I understood what he was saying ;)
gamesop employee's dialogue got a laugh outta me
glad someone paused for it lmao
for some reason i feel like you missed a letter, but i wonder which...
maybe gameshop or gameslop? perhaps gamesoap or gamesoup.
😂
lmao
you're not more likely to roll a 1 on any given dice, but if you roll a lot of dice at once, chances are you'll get a 1 before someone rolling just one dice repeatedly does
(not disagreeing with you, trying to summarize as best i can)
Essentially, playing multiple slots just means you have multiple base spins at once, while method hunting increases your chance on ONE machine
cool video but unfortunately i heard french
Tbh I think the french language gets a bad rep because of its... "people"
funny enough i’m never happy to hear my own mother tongue either but mad props to adef for the funny bit
As a french person, I can tell you that your french sponsor was on point, I was NOT prepared
merci!!
did not expect to see my face at 10:46 LOOOOL. what a fun video! idk any math but a shiny hunt, no mater the odds, should be A JOURNEY!!! YES!!!
🙌
respect the commitment for the ad read bro
The only reason this debate has annoyed me is because I’ve often seen full odds hunters act high and mighty about not using methods, while running loads of systems at once. It’s true that it doesn’t mathematically increase each game’s chance of a shiny, but we need to ask ourselves why do people use methods in the first place? I think it’s as simple as “it takes an average less amount of real life time”, and guess what other thing you could do for that result that doesn’t include in game methods. Ultimately, I just wish people would hunt however they wanted, and didn’t assign different values to full odds and method hunted shinies. It’s all about having fun in the end
Yeah, why do people brag about not using method hunts? Your essentially bragging you wasted more time looking for a shiny, when you could have spent less. You could even argue that farming for Herba Mystica and getting shinies in less than a minute is more impressive considering the more active effort compared to just pressing a few buttons every once in a while while watching TV. I just like people being happy they found the shiny they like, doesn't matter if they spent 3 minutes or 3 months.
@@TheDragonQueen-uh4lmI’m pretty it’s that people don’t have that choice. Not everyone can buy several consoles to hunt while you have folks out there claiming they’re so dedicated to full odds shiny hunting, going through that grind, only to have 15 devices hunting at once. It’s demoralizing to people who want to hunt using their old ds from 2010
@@michaelvaldes3572 yeah, that's defo a factor too. I always find it crazy to see clips of people hunting in old games with 6+ DS's. Just looks exhausting having to keep up with running away and encountering constantly.
Congratulations on becoming Captain Obvious..
Yeah to me using multiple systems is no different than using speed up if you’re hunting on an emulator, and you’d definitely get shit on for trying to boast about hunting full odds but admit you used speed up.
9:09 same thing goes for Horde, Doubles, and HGSS starter hunts. You are not making the bullseye any bigger. You are just throwing 5, 2, or 3 darts at once/in faster succession
Imagine speaking French (cool video)
Bro how is ur comment from 20 hours ago???
how did you do this 20 hours ago
@@tobymyles9764 it's a french tech
Je suis la jeune fille!
@@tobymyles9764 the release time of the video is when it was made public. Bébé Poucre W probably is a patron so they have access to the unlisted video days before the rest of us.
I would assume, before watching most of the video, that it functions like advantage in dnd. Two simultaneous encounters are technically more likely to have a shiny than one. The odds of an individual roll isn’t increased, naturally.
This is a lovely comparison! I think I would only caution that advantage is a touch different than multi-game hunting in that advantage does actually increase odds, because while the dice themselves are independent our TREATMENT of them is not; we are checking for the higher result and ignoring the lower, treating two dice as a single result! This is because (generally), the point of advantage isn't to find a specific number, but to find a *higher* number-- rolling a 1 and a 2 still nets you a better roll than with otherwise, even though both are bad! This is also why if somebody rolls with advantage by accident, they can't just pick one of the dice and claim it as their intended roll-- advantage means those dice are paired, and so they have to scrap them both and start completely over with a single roll.
With shiny hunting though, you're not looking for any higher value but for just one single value-- that sweet, sweet 8192. It does not matter what else is on that die, whether its 1 or 8191-- all we want is the one outcome. Rolling 2, 5, or 100 dice at once doesn't change the odds that you get it, because each die is a single individual check for a value. You just save yourself the time of picking up the die, throwing it, waiting for it to land, and repeating it 100 times.
I was not in the slightest familiar with this argument, but this video feels very incomplete without talking about the main factor that would actually affect how this feels in practice, time spent dedicated to the hunt. Increasing encounter rate (raising number of trials) or increasing odds (raising probability of success) both decrease the average expected time it should take to find a shiny, but in different ways, and it's not easy to compare them intuitively. Obviously this depends heavily on the time required per encounter and thus many different in-game factors, but I think giving an example with all other things being equal would have been both more educational to see how the math is different for raising encounter rate & odds each, and to ground the video more in actual shiny hunting practice bc as is it feels pretty abstract and only relevant to this very idiosyncratic argument over terminology
Quite Patrick you're scaring him!
Best ad segment I've seen in a while
I didn’t even realize because I’m shiny hunting while watching this
I want to start this saying I have no horse in this race; I just watched a neat math video about pokemon :)
At 7:40 you bring up that the Shiny Charm is effectively increasing your odds to 3/8192 by performing three "rolls" on an individual encounter. Isn't this functionally identical to having 3 copies of a game and getting 3 encounters? The shiny charm is just doing the work of having a "multigame" for you instead of making you press more buttons on more game boys, right?
For the dartboard analogy, it seems that the Shiny Charm would be allowing you to throw 3 darts at once, instead of either increasing the bullseye size, or increasing the speed of throwing. I guess my question is whether 3 darts at once is functionally different than changing the speed of thrown darts to match that 3 dart output?
Totally agree! Glad someone else brought this up! :)
Yeah, I don't see any evidence that the base odds per shiny roll actually change. As you say, each shiny roll (whether occurring in the background or as an individual encounter) is always 1/fullodds. Shiny charm just does three consecutive full-odds shiny rolls behind the scenes, whereas 3x multi-hunting does the same number of full-odds shiny rolls in-front-of-the-scenes. The only difference is that the player experiences each roll as an encounter in the multi-hunting scenario and every 3 rolls as an encounter with charm, but in the end, you'd expect to get the shiny after the same number of full-odds rolls.
His dartboard analogy WOULD make sense if it were actually changing the base odds of each shiny roll to 3/8192, but I don't see any evidence that that's what's happening. All the documentation I can find on Bulbapedia says that the odds are *approximately* 3/8192, because the base odds per roll are not changing. Rather, the game is just performing a full odds roll three times. So the odds work out to be roughly 3/8192 per *encounter* but it's not exactly 3/8192.
Well yeh, the dart board analogy is not actually correct. Whats actually occuring if your throwing 3 darts at the same time and counting as one throw, whereas playing on three games you throw 3 darts in the time it take someone else to throw "1" dart.
@@XenithShadow But these are functionally the same thing right? I guess that's what my point is, that the only difference is how you count it. The point as far as the actual math is concerned is they're identical, right? if people want to ascribe a title of "full-odds" to a specific way of hunting, that's a separate issue of clarifying what "full-odds" vs "method" means.
@@ROCKMANDRONIAN As some one else explained in a different comment, using 3 consoles is actually even better than a shiny charms for total shinies encountered as while extremely unlikely it is possible to simultaneously encounter 2 shines while using 3 consoles whereas you would discard one of those shinies when using a shiny charms as you only encounter one of the pokemon.
But yeh aside from that super unlikely edge case in both scenarios 3 rolls are occuring to see if a shiny is found just that in the shiny charm case those roll happen behind the scenes.
Does the Shiny Charm actually work that way, though? The number on screen says 3/8192, which is NOT the same as three trials of 1/8192. That distinction is a major thesis of this video.
THANK YOU FOR YOUR SERVICE!! This topic has plagued the online shiny hunting space for so so so long, and I will now forever send folks to this video 🎉
It doesn't change the odds at all. It increases the encounters per unit of time, but at the full odds.
So glad somebody actually explained to those who don't understand probability that results don't affect each other. Even though the odds are there, the denomination isn't changed. You can always hit the exact same spot on that "dart board".
This is a very good video! It really does help convey the information in an understanding, yet concise bite sized chunk. It also just makes me happy to have another adef video lol
I feel like a much simpler way to put this is that it's the same as speeding up the game
The methods that increase your odds are like buying multiple tickets for the same lottery.
Playing multiple games is like buying tickets for multiple different lotteries.
Really, the easy answer is just that they get their shinies faster, not because the odds change, but because they're getting encounters faster. This is a weird argument to be having, but hey, Absol feature! I'm glad.
Can't believe people believe this you can mathematically prove the odds increase. Why is people arguing they are the same?
@@gabrielbarrantes6946 So you didn't watch the video then?
@@NightKev I did and posted a comment showing the computation he didn't show in the video, proving the odds indeed increase.
multigame hunts effectively increases the encounter rate. the odds themselves stay the same, but you are rolling the dice more often
Masters in Statistics here. I was skeptical coming in but you ended up explaining it well, and I 100% agree.
Honestly im surprised people assumed the odds get bumped up if you use multiple games. I thought it was common knowledge that, thats not the case and its just giving you more opportunities to find one because you see more than you would with just one.
while I think the math is all super interesting, I actually disagree with this video and think it kind of misses the point. I don’t think this debate has ever been a question of whether multiple games increase the odds, but rather if it decreases the effort required. and even then, I don’t think the arguement has ever really been in good faith.
the main thing I always come back to is Horde Hunting in Generation 6, which lets you see five Pokemon at once in a single encounter, each with their own chance to be shiny. this isn’t really any different from doing single encounters on 5 copies of the game, and yet pretty much everyone agrees that Horde Hunting is a method hunt. seeing multiple Pokemon at once obviously does increase your odds of finding a shiny within a set amount of time, so then why are these two methods, which are functionally exactly the same, treated so differently?
in my opinion, it’s because it’s never been about odds or even effort, but that this whole debate is arbitrary. shiny Pokemon have no inherent value, only the value you give to them based on your own experiences, and people trying to argue that certain shinies are not truly full odds, or cheated, or at all less valuable because of how they were hunted, is just a way for people to try and invalidate other people’s personal victories. I don’t think explaining the math will change anyone’s mind, especially when you ignore that time and effort are the most important factors in many people’s eyes, not just the statistical probability of each encounter.
Yep. This video completely missed the main point of the debate which revolved around time and effort to get the shiny. It was NEVER about shiny chance per encounter. This guy probably has a "holier than thou" attitude to anyone who does not full odds/multi-console full odds hunt.
@@OsirusIrdia no, he was pretty respectful about everything, just mistaken on what the main point people were arguing about was. just because you disagree with someone, doesn't mean you should make assumptions about what else they believe so you can get angrier
@@OsirusIrdia You can disagree with someone without attacking their character
@@OsirusIrdia he has an entire segment at the end dedicated to talking about how it doesnt matter what method you use, i dont think you understood the aim of this video
yeah id certainly agree that it ignores the argument people are trying to make about it cutting down time and not actually being about the maths, but then again from the start he frames it as being about the maths and correcting people who get the maths wrong, not being about correcting the argument itself, so in a way i guess were all missing the point lol
the dart board segment is actually the greatest explanation of shiny hunting i have ever heard
I'm pretty sure the argument about odds for multi game hunting isn't "per encounter", it's "per unit of time". If you roll a dice once every minute, you have a 1/6th chance for at least one six per minute. If you roll six dice, you have 2/3rds chance for at least one six per minute.
wish i had 8192 pokémon games
This is your best video yet. This was a really good one
5:22 you get twice as many encounters while it is true that it doesn't increase your odds in a given encounter it does double the amount of encounter lets say you get 5 encounters per min with one copy and 10 with two at just 7 hrs your odds of having gotten a shiny are 17.5% more than only using one copy if you don't consider increasing the odds that you have gotten a shiny in a given amount of time a method than you should redefine method to include it also boost of having gotten a shiny in a given time isn't minimal until over 100 hrs
It's so surprising to me that this is even a debate - I am absolutely terrible at math and cannot calculate probability at all, but it just seems obvious to me that multi game hunting is not a "method hunt," because as you say, each game is still at 1/8192 odds and are not related.
9:02 oh no, I missed the dart bord and got missing no
Multi-game hunting is just like shinies in Legends Arceus or Scarlet/Violet. The base odds are the same (roughly), but you are still more likely to find one just because of the sheer number of encounters you generate during your playthrough
To be honest I think everything you said here is very intuitive and doesn’t really counter the commenters’ points? You aren’t changing the odds of any given encounter but you’re speeding up the process of finding a shiny. Somewhere baked into the impressiveness of the original 1/8192 odds is the idea of how long an encounter takes, and how many encounters you’re likely to have in a typical game or day. I think it’s totally reasonable to consider speeding up that process an advantage, and maybe even be mad about the term “full odds” being applied if you’re into getting mad about that kind of thing.
As someone who watches Arlie, I thought it was really cute when Adef used footage from one of her Shiny hunting streams
Going against my new policy here of Zero Math, Only Turtle. But i think i can make an exception because it might lead to a shiny turtle
[Captain America voice] I understood that reference.
So if so did the math right (it’s like 2 AM for me right now) if you have 8192 games and had an encounter simultaneously on each one, the chance of there being at least 1 Shiny is about 63.2%
sounds about right - the expected number of shinies is going to be exactly one (expected shiny count is the actual trivial addition of odds), so you should see no shinies less than half the time but not that much less, and you should see two shinies somewhat often and more than two not so often, etc.
We need an Adef, Ackolade, ShepskyDad collab
oh hell yeah they all rule
The crux is time vs attempts.
Time is obviously the actually important factor here. Why? Well our lives have finite time, but theoretically near infinite attempt possibilities. Time is our valuable resource here.
And even when attempts are what is touted in the title of the video, the *reason* it sounds impressive is because it implies a very high expected "time to catch". Because we all know each attempt will take roughly X time. So its like "wow, hes got a big journey ahead of him!" But then it turns out, hes driving a car, that distance, rather than running it. Yup, same distance, but much less.... Impressive? Or tedious.... Or whatever. Its not the same, is the point.
And Double game hunting does roughly halve the expected time to catch, so *functionally* within the real world (not within the game data, but within the real life time you spend sitting there pressing A) yea you increased your odds. And i think its definitely reasonable to lump this slightly distinct but ultimately mostly the same method, in the "method" category.
Now, using methods is just smart.... Like why would anyone WANT to sit there and mash A for a really long time?.... Idfk. But regardless i believe multi game hunts should probably be in the method category.
After all, would you really consider a 1000 game simultaneous emulation hunt the same as a single game hunt? The category seems innately different at that point. 1000 game hunts would be over much much faster, and require computer setups and method stuff. Vs the raw game hunt is just like if you were a kid doing it as the devs intended or... Whatever im rambling now.
I found it interesting that you brought up the "impressiveness" of the shiny. That's because when I started to shiny hunt I was like "wow, these shiny hunters are amazing, look how much time they spent! They are so lucky!". However, after catching a few shinies myself, I started to realize that, sooner or later, eventually the shiny would appear. And that made the whole process a little less exciting. I still like to shiny hunt sometimes, but it's definitely not as impressive as I thought it were at the beginning.
@@matheusbecker2351 yea, and i believe that perceived impressiveness is why ppl go for raw shiny hunts. And for that exact same reason, multi game hunts should not count as raw, because they are alot faster and alot less tedious and less "impressive".
Personally I don't understand the appeal to any of this.... Idk lol XD
i had no idea this was an argument even going on, but i guess i don't really have my finger on the pulse of the shiny hunting community right now. good to know that i seem to have a decent understanding of these things despite being bad at numbers. good video! :)
You don't even need to play pokemon to shiny hunt technically speaking. You could use a random number generator program/app/website and set the list of numbers to range from 1 to 8192 then pick a number within that range as your target and then you just hit the "generate number" button untill it hits the number you chose as your target.
Now i obviously can't say that the way pokemon does it's number generation when generating encounters and checking for shinies is the exact same as the way a random number generating tool generates its numbers but in essence they should be similar enough.
I was genuinely so happy when I saw adef post.
8:00 TL;DR: Performing 3 rolls to check for shininess of each encounter is not the same as tripling your odds.
The shiny charm performs a set of 3 rolls per encounter to check for shininess. Each roll generates a new personality value, and 8 (or 16 in gen 6 and beyond) out of the 65536 possibilities correspond to a shiny Pokémon. I am not certain if the rolls are done with replacement or not. If they are done with replacement (i.e., the same personality value can be rolled twice in a set of 3), it is effectively tripling the number of encounters (but stopping in the middle of the last set of 3 if the first or second roll in that set corresponds to a shiny), which leads to lower than triple odds. (This scenario would make hunting with the shiny charm essentially like hunting with 3 consoles at once, except that a maximum of 1 shiny can pop up between the 3 games on each set of encounters.) If they are done without replacement (i.e., a personality value that is rolled cannot be rolled again in the same set of 3), the shiny odds would be ever so slightly higher, but still lower than triple odds.
For 1/8192 base odds (gen 5):
-3x Odds: 0.0366211% (3/8192)
-Shiny Charm (with replacement): 0.0366166% (201302017/549755813888)
-Shiny Charm (without replacement): 0.0366172% (306736569/837684797440)
-Ratio of odds (3x/SC with r/SC without r): 1/0.999878/0.999893
For 1/4096 base odds (gen 6 and beyond):
-3x Odds: 0.0732422% (3/4096)
-Shiny Charm (with replacement): 0.0732243% (50319361/68719476736)
-Shiny Charm (without replacement): 0.0732254% (3608225/4927557632)
-Ratio of odds (3x/SC with r/SC without r): 1/0.999756/0.999771
While the difference is essentially negligible in this case, the values would be much more distinct if the odds weren't so low in the first place, so it's important to understand the concept.
It doesn't change probability it just changes frequency of attempts. Pretty simple.
I think the conclusion is technically correct, but not for the reason stated, as there's a subtle flaw in the comparison made: The shiny charm (and similar) don't provide a flat increase in the odds. I'm not 100% sure this is true, though it is based on what I consider to be credible sources, but I welcome being corrected.
Explanation:
Let's use the example you gave, using the shiny charm in a game with base 1/8192 odds. You claim this increases the odds to 3/8192. Best I can tell, this is only approximately correct: The game actually performs three full odds rolls, and if any are shiny, then you get a shiny. The odds of that are (1/8192) * 1 * 1 + (8191/8192) * (1/8192) * 1 + (8191/8192) * (8191/8192) * (1/8192), since it doesn't matter if any of the rolls after a success are shiny or not.
Contrast this with three full-odds encounters, say from three independent games, with no shiny charm. I agree, these are independent, so each one has 1/8192 chances of being shiny. The combined odds of getting at least one shiny is 1 - (8191/8192)^3, which is... identical to the earlier calculation. But wait, that's at least one shiny! In particular, all three could be shiny, so these cases are clearly different.
While that's certainly true, let's consider the most common case where someone stops after getting a single shiny. Now, for the multi-game approach, there's always the chance of getting multiple shinies on that last set of attempts, and that's always going to make them different. For the moment, I'm going to ignore that small chance of multiple shinies on the last set. Under these conditions, each method is going to end with a shiny after a certain number of attempts (probably, of course - it could never end). For each shiny charm encounter, the odds of getting a shiny are identical to getting at least one shiny with a three-game hunt, which means the odds of getting no shinies is 1-P(at least one shiny), which is identical for both. Thus, every set of three full-odds encounters which does not find a shiny is identical to a single shiny charm encounter which does not find a shiny. This is why I ignored the chance of multiple shinies on the last set of attempts: Every attempt up to that point, which I believe is what most people care about, is identical. So long as you count the total number of rolls rather than encounters.
Now, that's definitely not true if, as you assert in this video, the shiny charm actually increases the odds to 3/8192, as opposed to being three 1/8192 rolls. I haven't seen the source code, so I'm not 100% sure which is correct, and many places are inconsistent about which is the case. However, taking a look at how Anubis explains it (This is a SV source, but I don't think anyone's claiming the shiny charm has changed functionality since its inception: www.reddit.com/r/PokeLeaks/comments/yxtbj6/datamine_official_shiny_rates_ver_101_via_anubis/#lightbox), rolls are consistently used to explain effects like the shiny charm. Notably, 3/4096 (since this is gen 9) never shows up in his description for the shiny charm, but rather 1/1365.67. This doesn't prove who is correct, as both methods equate to that with very little difference/error between 3 rolls of 1/4096 and 3/4096, and it makes sense for the final column to be an easily digestible 1/x probability. Yet I do think it's telling that so much effort is spent on "rolls" rather than just stating increased odds.
The comparison in this video is definitely true for the gen 8 Dynamax Adventures, which apparently have exactly 1/300 base odds. Interestingly, the shiny charm still acts on this, giving two additional rolls, or perhaps flat increasing the odds if I'm wrong, to (approximately) 1/100. So comparing a single DA encounter is never going to be comparable to any integer number of full odds encounters at 1/4096 (gen 8), but if the shiny charm does indeed induce multiple rolls, then the above arguments are still valid for multi-game hunts with full-odds DA encounters vs single-game shiny-charm-odds DA encounters.
Edit: To be clear, I agree with the takeaway message, as I couldn't care less how people hunt or how they choose to count it.
Also, I did have an earlier comment that I deleted, as I don't feel like it gave the video enough credit, nor was it quite as thorough.
Edit 2: I saw some people mentioning horde hunts, and that's a great point. I think most people would agree a horde hunt is a method hunt, as it's not a single encounter at full odds. Instead, it's 5 encounters at once, at full odds or with shiny charm odds. This is exactly identical to 5 games at once performing single encounters, either at full odds or with shiny charm respectively, without the edge-case difference, and regardless of whether the shiny charm is a flat increase or uses additional rolls (in the case of all using shiny charm).
YES! So happy someone else brought this up! This needs to be up higher, haha. :)
I 100% agree with you, all evidence seems to point to the fact that base odds per shiny roll do not change, even if the number of rolls per encounter do change. Barring those edge cases you described (e.g., multiple shinies on the last set), both scenarios should be mathematically identical.
LOVED the dart board analogy!
Also, I enjoyed the French bit.
Love the way you explain complicated ideas!
6:01 as soon as you made the 8192 analogy & showed the 8192/8192, then asked if that was a 100% chance that one of the copies would have a Shiny on the first try, i feel like everything clicked & i thought "well, of course not, each game has only had 1 individual cycle, there's not only no chance of it happening, but having 8192 different sets of odds stacked together feels like it would be way higher." Am i wrong? I paused the video at 6:01 & typed this, i'll finish the video & find out
Just finished watching the video, I thought it was really good! I've never looked into shiny hunting streams really but I love your videos on the math behind ideas in pokemon hunts. Thank you, Mr. Pokemon math guy
stop
Your videos always make me wish I followed through on doing a math minor in university
Wait, this isn’t the news?
I think the "encounters spent or hours spent" point is a great point into why the argument exists. Ask any full odds hunter why they hunt on multiple systems and they'll tell you something along the lines of "To get through encounters faster". Valuing the "encounters spent" more. You could hunt on 30 games doing 1 encounter every 20s on each game and find a shiny in 45 minutes, you'll probably say you found the shiny in 4000 encounters instead of 45 minutes.
Where as someone might use Masuda method hatching 1 egg every minute on a single system and gets a shiny on the 500th egg. It's a smaller number but it still took 4x longer than the multi hunter. To them the increased commitment is more important than the final encounter number
It's why some people who hunt full odds are against methods, even if they only use 1 system so that the time spent is identical, the total encounter number is smaller so it feels less special.
And why some method hunters are against multi systems, because they don't care about the final number, they just see it as cutting down time, which they value more
But that's just my insight on the matter. Now I'll leave before I get cooked alive for expressing my opinion on the internet
Oh, is a "method hunt" or "changed odds" like a thing people use to downplay success of some shiny hunts?
My first reaction is that it'd obviously make shiny hunting faster.
I'd have looked at it as in the time that each game would get an encounter we would instead get n encounters, so rather than 2/8192 we have 1/8192 + 1/8192 * 8191/8192, which is rather close to 2. Or more broadly a summation of i=1 to n of (1/8192)(8191/8192)^(i-1).
We cannot add the probabilities since they are not disjoint (you could get shinies in multiple games).
I know that this is also simplifying things (not all games will get encounters at the same time, so they will desync. This might mean that you do not like the thought of blocking a set of 1 encounter in each game to view the probability of there being at least 1 shiny in that encounter block; however, the odds of a shiny appearing in an encounter block is greater than a single encounter.
However, this is obviously equivalent to the probability of doing that many encounters in a single game equal to the block size n.
This also ignores the internal mechanism for the games to determine their randomness and create shinies.
Omg thank you so much for making this! the amount of times I’ve had to explain this to people is insane, I’m so glad we have a Pokémon math guy now who can explain things super clearly like this
I understand the video’s point that multi-game shiny hunting doesn't change the 1/8192 odds per encounter, but that's never been a point of discussion for me personally. By playing multiple copies of the game, you're still significantly increasing the number of encounters. Your 8192 copies of FireRed example doesn't guarantee a shiny, but you’ll likely find one within mere minutes. So while the odds per encounter don’t change, the higher encounter rate does have an obvious benefit.
I feel like it was kinda unnecessary to call that into question - atleast that's how I perceived some of your commentary. Feels like we're splitting hairs over the true definition of method hunting, rather than looking at the net results. In the end, the result is what matters, no? Someone playing on 10 copies of the game is getting more shiny encounters within the same time frame. Thus, I can only conclude it's a valid and effective way of obtaining shiny pokemon more quickly.
I don't think it's unreasonable for people to say faster shinies = better odds. Yes, it's not the correct definition when it's the rate of rolls being increased, but we all understand what's being said. Just my two cents I guess.
I read the title wrong and thought this was a Breaking Bad collab.
I think it's a bit obvious that it doesn't increase the odds per encounter though? when people say it increases the odds it just means it increases the odds to find a pokemon in a given amount of time, it's still an increase in odds just not for the encounter metric.
You can say that multi-game hunting increases the odds of getting a shiny within a certain time frame, right? Increasing the number of encounters within a time frame should increase the odds of getting a shiny within that time frame
I thought this was going to be about RNG Manipulation as soon as you mentioned the biggest controversy in shiny hunting. I was expecting im a blisey to show up
As a French person the "segment sponsorisé" is really fun and honestly your french is great
The French wasn't bad!
Came for the math, stayed for the message of love and community
Here’s a more simple analogy: flipping two coins does not mean you have a 2/2 or 100% to get heads. You will lose *a lot* bets that way.
75% winning if your win condition is winning on whatever coin (you win if you get a shiny on whatever game), so odds indeed increased. You guys really don't understand probability at all.
@@gabrielbarrantes6946 running multiple games does not change the shiny odds, they are still 1/8192, it just increases the number of encounters, thus speeding up the process. The probability of finding a shiny after 10000 encounters in a single game is 70.5%, while the probability of finding a shiny after 10000 encounters with 2 simultaneous games is also 70.5%, it just took you half the time to get to 10000 encounters while running 2 games. Running 2 games is the same as running 1 game in 2x speed, it does not change the probability of finding a shiny within a certain amount of encounters, it only changes the number of encounters per unit of time.
REMEMBER THAT EVEN IF THEY ARE USING MULTIPLE SYSTEMS, JUST BECAUSE IT TOOK LESS TIME DOESNT MEAN THEY DIDNT DO IT IN LESS ENCOUNTERS
time is what matters the most. people bolster about "full odds" but certain non-full odds hunts will take longer on average than full odds.
While yes time matters.. should it really matter? Would you be mad that getting a shiny ekans with random encounters took 2 weeks with 1 game or 2 days with 5 games, or would you just be happy you even just got a shiny?
And time is such a variant. Like technically yes, you *should* get mons faster with multiple games. But that doesnt mean you always will. I could have gotten a shiny ekans in 2 days using 1 game while i could have gotten it in 2 weeks with 5 games.
@@EliasMorals Yes, a hunt with 10000 consoles can take a year. But ON AVERAGE, u will get the shiny in less time than with 1 console. What matters here is not the variance (1 hour or 2 weeks), but the expected amount of time it takes to get the shiny. And even when a multi console hunt takes longer than a single console one, MOST OF THE TIME the multi console hunt will take less time. This is mapped pretty well to binomial distributions.
@@EliasMorals Think of it like this:
A shiny celebi hunt in crystal on one copy takes 72 hours on average. 3 days of hunting. That same hunt on 3 consoles only takes 24 hours on average, just 1 day. You just saved 2 days worth of time by hunting on 3 consoles instead of 1. Time matters the most, the only reason we care about encounters is because it is linked to the time it takes to perform a hunt. That is why some hunts are considered "harder" than others. A shiny manaphy hunt, is just 1/8192, but each encounter takes a lot longer than shiny hunting a 1/8192 wild encounter.
@EliasMorals if time doesn't matter, wouldn't it make more sense just to hack it in? Time is vital to the value of a shiny; the patience it takes to keep resetting until you get a shiny pokemon is what gives shiny pokemon their value, it's a reflection of the work you put in. Counter intuitively, full-odds hunters don't want shinies, they want the challenge, and the shiny is their badge of honor to show for conquering the challenge. Multi-hunting eliminates/mitigates the endurance part of the challenge, and I think that's why it doesn't sit well with some people.
this is dumb. no one is saying it “increases your odds at a shiny” they’re just saying it takes less total time invested to find 1 shiny if you’re using multiple games at the same time. obviously.
If you are using multiple games, you aren’t increasing your odds, you are just increasing the rate of encounters
In other words, you aren’t shiny hunting better, you are shiny hunting _faster_
Please make "e plays Pokemon" as a competitor to "π plays Pokemon" with better inputs :)
I really don't like when people make stupid comments acting like they know what they are talking about. Everybody knows that φ would be clearly superior.
Well the ad certainly worked i was very confused why you were suddenly speaking french and tabbed back over to youtube
I imagine a normal person cares more about the speed vs the number of encounters. People just put the number in video thumbnails because it's eye catching
I think a lot of "droprate"-type probability misunderstanding/confusion comes from "chance of event" vs "chance of seeing event happen after x amount of time/trials". When people are going for droprate, they're repeating the event multiple times and people will often not be precise in what they mean.
I actually disagree with the dart board analogy. I get where you're coming from, but 3 rolls/checks within one game operates no differently to 3 games doing one check. So if you're using a method that makes the odds 3/8192, i.e the game checks for shininess 3 times against the regular 1/8192 odds with every encounter, then it doesn't matter if it's happening within one game, it's doing the exact same thing as you doing 3 encounters yourself by using 3 games to do an encounter each. The odds are and always will be 1/8192, just the method checks more times within one game, and the multi game setup has more checks because there's more games. In other words, both the method and the multi game setup are giving you faster checks/dart throws, therefore reducing the time of finding a shiny (in most cases). There is no functional difference between the two whatsoever, so if you hunt with multiple games, expect similar results to methods. Whether that affects a shiny's value to you is subjective, people can hunt how they like.
EDIT: The time where the larger bullseye analogy would be accurate is if the base odds for the method were changed from 8192 to something else. If the base odds were halved to 4096, i.e every pokemon has a 1/4096 chance of being shiny on every encounter, that's different from having two checks/rolls/dart throws against 8192. From 6th gen onwards, the base odds were programed to be 4096, so 6th gen and above have a bigger bullseye. In 4th gen and below, the methods simply increase the shiny checks, they don't change the base odds.
I was hoping someone else would bring this up; I agree with you 100%! :)
Yeah, I don't see any evidence that the base odds per shiny roll actually change. As you say, each shiny roll (whether occurring in the background or as an individual encounter) is always 1/fullodds. Shiny charm just does three consecutive full-odds shiny rolls behind the scenes, whereas 3x multi-hunting does the same number of full-odds shiny rolls in-front-of-the-scenes. The only difference is that the player experiences each roll as an encounter in the multi-hunting scenario and every 3 rolls as an encounter with charm, but in the end, you'd expect to get the shiny after the same number of full-odds rolls.
Glad someone else brought this up! :)
I feel like its a simpler thing for the idiots that the video is directed at to understand though
"Was this a good bit? We were so worried when we were writing it. But now I'm glad! It's better than some previous bits if you ask me."
"GET ON WITH IT."
You're 100% right about all the math, but I think where the more educated part of the community lands is that full odds does not solely mean 1/8192. It is a fan made term and has an ambiguous definition, and the behaviors exerted by the full odds community implies that it means more than just the chances itself. I'll be making a video soon enough and I hope you'll entertain the ideas 😉
There's other philosophical arguments as well, such as is the odds the individual odds of a game, or is it YOUR, the players odds at any given time. It's just not as simple as "the chances don't affect each other," which nobody outside of UA-cam comments will really argue against.
As a French person, I'm shocked that you nailed French tbh. Listened to the entirety of the bit and didn't even have to look at the subtitles!
Thank you for your hard work once again (basically thinking this every time I watch one of your videos). I hope that you get the recognition you deserve for your fun methods, and thanks again for this creativity that you always show.
The example with the shiny charm (7:40) was very poorly chosen imo. Let's say that the shiny charm increased the odds to a guaranteed shiny (8192/8192), is that the same as the game doing 8192 and choosing a shiny among them? No, it is not. It might be of course be that all of those 8192 are non-shiny.
Well-said!
Exactly, this guy doesn't understand probability at all... The main topic of the video is also wrong, odd do indeed increase because running simultaneous experiments is not the same as running them in sequence, the probability space shows that the odds increase.
If the Shiny Charm raised odds to 8192 in a 1/8192 game, the way it’s coded in game would make it 63.212% per encounter
Shiny Charm is not a multiplier, it’s an extra encounter generator. That standard 3/8192 is just three 1/8192 where, if it lands a shiny, it picks that encounter over the non-Shiny ones
@@skarmtsk I don't actually know how the shiny charm works, but that's not really relevant. My point is that if the game actually works by generating 3 encounters and choosing a shiny if there is one, then that's not the same as changing the odds from 1/8192 to 3/8192. The odds to find a shiny in that case would be 1-(1-1/8192)^3. This is very close to 3/8192 but it is not the same. These two numbers are close enough that I'd normally think it's not a big deal to approximate the former by the latter, but I think it's a very poor choice to do so in a video that'd supposed to explain probability properly to all the people who don't understand it.
@@ekigonzalezgarcia I know that’s the case, them being similar but not the same. Your point though doesn’t work on the merit of the prior two minutes of the video disproving how it can’t be cumulative like people assume
Once it gets to the 3/8192 comment, it’s supposed to be saying “the Shiny Charm makes Shinies roughly three times more common, turning 1/8192 into practically 3/8192 and this is how it works”. Was it potentially poor wording? Most likely, yes but adef didn’t say what you’re claiming him to have said
This was a nice video! I already knew the math behind this, but it was still interesting to hear how you explained it and also the culture and terminology with shiny hunting, something I hadn't really known much of anything about before. Also that sponsor segment was an *incredible* bit.
So it's not a method hunt, but it's still basically the same effect as one, got it
I'll say this: the problem with people who use multiple games isnt that they are "cheating" as a shiny hunter, they are disgusting consoomers who encourage bad buisness practices.
If the professors at my university explained math with Pokemon related stuff I might actually have finished my master degree
It was a great bit - I laughed out loud when you did most of your ad read in french. Full marks.
Vraiment nice le sponsored segment. J'apprecie beaucoup!
Best ad bit I’ve seen in years, funny, well timed, showed off the product well and actually took effort.
Loved the add will watch again! To be honest great video as always and I appreciate the effort for the advert read. I’ve watched more UA-cam than I care to admit but this is definitely one of the best advertisement reads I’ve seen. Keep up the great work ❤
your videos are so fun to watch as someone taking a statistics class this semester and also attempting to complete a living dex
The most creative and engaging ad I’ve seen, genius