The Math Behind The Unluckiest Moments In Melee

[@ChalerYT]

Realizing I should had left in the section were i really solidly established that I understand a lot of what the commenters are saying, that this is wrong math. Cause i thought i got it across enough already. But just gonna establish again that yes this isn't probability in a literal sense. You cant actually put numbers to most events in melee cause decision making isn't odds based ect. but i thought it would be fun to put the idea of it to that and play with the idea much like i have done in the past (hardest character ect) so please dear god dont think I'm talking about it in that sense in the video that's not my intention at all. I just wanted to make a fun video talking about potential probability to put an idea to the reason we see so much unique and interesting stuff in melee all the time hope this makes my intentions clearer and I will be extra clear with my intentions in the future and leave in the clarifications next time since what I think is clear as the maker of the video and what the audience receives is always going to be different If you did enjoy please feel free to support me on my Patreon for just $1 with enough people it will literally change my life. www.patreon.com/ChalerPatreon

[@Savestate]

LMAO the luigi misfire clip is so old, shoutouts to SmAsheville my origins. glad to know i totally lost that set just because of rng and not because i was down 3 stocks x3

[@brynndotcom5703]

SmAsheville my beloved

[@EA_SP0RTCENTER]

I got 7 in a row one time.

[@chatbot3649]

Boone was sicker 💪🤖

[@jonathancangelosi2439]

I think once you start attaching numbers to every choice in a given interaction, you start to lose meaning. Every game in Melee is different. It’s not terribly surprising that probabilities rapidly get small once you factor in every little decision. Even the first 10 frames of any given game in Melee are likely different. The calculation you made for the windmill assumes the three interactions are independent events, which is not true; the windmill follows a deterministic path, so if the first interaction lines up, the other two will also line up with 100% probability. Therefore, the actual probability is 1/180, assuming the windmill interaction has to be frame perfect. There’s also player intent. Players react to what they observe on the screen, so in order to obtain the estimates you gave for landing on Randall, you have to make some unreasonable assumptions, like “the player would have done the same action in all circumstances.”

[@cueshq789]

Im not reading allat

[@dangoxnagisaxushioxlove1367]

hes right tho@@cueshq789

[@jonathancangelosi2439]

@@cueshq789 …I don’t understand why you felt the need to publicly announce your intellectual laziness, but go off I guess.

[@cueshq789]

@@jonathancangelosi2439 igl guess youre “intellectual” that you assumed i had any problem at all. The need to publicly announce anything on SOCIAL MEDIA is the same need that you found necessary to post whatever tf you talking about 😂😂 but go off pimp

[@ChalerYT]

I was aware of this when I made the video so I have updated my pinned comment to address the idea behind this comment ty for the feedback and thank you for watching my stuff :)

[@jahridin]

Kinda disapointed you didn't count every events since the big bang (and potentially before and in PUs) that leads to the two player having these interactions.

[@benjaminh7548]

If you take ANY choice and multiply it by its options and then multiply that by another choice, you'll very quickly reach ridiculous numbers. Option select isn't probability. While these numbers are correct, they're meaningless.

[@Zapdatt]

What I was thinking too. The entire time I was just thinking to myself that he really doesn’t understand probabilities or combinatorics. Peach turnips and Luigi misfires by themselves are ok, the rest of this is just garbage numbers.

[@danielbojkovski723]

Yeah, the probability and math understanding in this video isnt there. But this isnt a math video about melee, its a video about melee moments that involve probability, aka most people dont really care about the specifics of probability theory, they care about funny green plumber and princess down b.

[@benjaminh7548]

​​@@danielbojkovski723 I mean the title of the video is literally "The Math Behind the Unluckiest Moments in Melee" not "Luck in Melee" So no, I think he was trying to do a math video about Melee. Also, he's not calculating probability, that's exactly what I'm trying to point out. He's trying to calculate option select as if all options were possible outcomes. It's fine when he indicates that something is luck-based (turnip, randal, misfire), but the fact that he keeps trying to attach a number to deterministic outcomes that happen as a result of the random occurrences shows a misunderstanding of situational math vs probabilistic math.

[@benjaminh7548]

​@@danielbojkovski723 Also, he recently made a video about how much Melee means to him and others. You're definitely not talking on behalf of chaler or those watching by trying to trivialize Melee with "funny green plumber and princess down b."

[@danielbojkovski723]

@@benjaminh7548 I agree with everything you said! The math is not correct, I just think it doesnt matter. No ones studying for their statistics final by watching this, its all for fun and on Chalers part its good for his bottom line

[@mymyscellany]

the math on the windmill interaction doesn't make sense since the odds of each event aren't independent...

[@hadesthegod9181]

Engagement on this video should be great since people love correcting others about how probability works lol. I on the other hand am just commenting because I love the video.

[@il100374]

This could literally just be a compilation of Mew2King versus Armada. Lmao

[@roundupssbm]

The windmill events were not independent, which is a huge (wrong) assumption, and I doubt there was only 1 frame in the cycle of te windmill that that interaction was possible. Still, very fun video :) keep up the good content!

[@jonathancangelosi2439]

Came here to say this, too. The other two outcomes are guaranteed if the first outcome occurs. So the probability should really just be 1/180 (assuming it’s frame perfect, which it’s likely not).

[@smoothguitarforever]

Spray in CS is not luck based, its the same spray everytime so its inherently skill based

[@Left-sh1db]

That’s all you took from this video?

[@smoothguitarforever]

@@Left-sh1db no

[@NightGhost1400]

your half right, every gun has a set spray but it also has a bullet deviation value that increases the deviation the longer you hold down fire. Meaning that yes spray patterns are mostly predictable but are still inherently random

[@smoothguitarforever]

@@NightGhost1400 no

[@Moj999_]

@@smoothguitarforeveryes.

[@x9x9x9x9x9]

That clip at 8:50 though reminded me of the coolest thing I have ever done in melee and it was not long after I started playing. I dont play fox but was for this game and got an invisible shine landing onto randall, got back on stage with a nair and then an upsmash killing the DK I was playing against and then got another invisible shine onto randall again but on the opposite side. I know this can all be done with good tech skill and knowing randalls timing but this was just pure luck for me as I have maybe done 10 invisible shines ever. But I love that clip.

[@justsumviewing9322]

Actually explored something similar to “Vibe Based Math” during a mathematics module at college. It was referred to as “Critically Applied Probability” or CAP for short. Yeah, a lot of the math in this video was 🧢

[@XinSSBM]

Liked the vid! For the 2nd example at 3:15 instead of doing the vibes based maths you could've added the chances of forced dropping an item when getting hit, which is RNG in and of itself. (in melee the chances are 'damage dealt / 60'. So that would be 51/60 = 85% chance to drop the bomb

[@MrGetownedLP]

THAT LAST CLIP WITH THE CLASSICAL MORNING MUSIC WAS BEAUTIFUL LMAOO

[@Thomas-rw1be]

People are hearing "vibe based maths" and getting mad that we're in the vibe zone. Think of these odds like betting on "what is this player going to do" during a frame-by-frame first watch, and I think we're in something that is pretty close in vibes. Awesome video, ban yoshi's island (joking)

[@wordsayer19]

Yeah, you're definitely taking a lot of liberties with these numbers. It's a pretty slippery slope once you start bringing player intent and input into the mix. You could just about make the numbers say anything you want at that point. I will say I think you give too much weight to the windmill and Randal. Both of them are deterministic, so once you measure the first instance of it, the probabilities for the rest of the clip are 100%. So after the first interaction with Randal, the probabilities are no longer about by how likely Randal is to be in x location, but rather how likely the players are to behave in a way that re-converges with Randal. In the case of the windmill, I think it's safe to say Mang0, after 15+ years of playing Melee, has a solid intuition of how the window moves and how to combo off of it. Not to mention he can simply react to it's position and adjust his combo accordingly. So everything after the safe landing is completely based on Mang0's skill and Plup's DI, not luck. The Marth/Fox/Randal clip is a little trickier, but I think it's clear the Marth is stalling on Randal after the first landing. I don't know if he did it because of Randal, but that was the smartest choice. Waiting to leave the cloud gives it more time to reach the other side of the stage, so if he did get hit in that direction, he would be very likely to have the option to recover there. So if the Fox keeps hitting Marth to opposite sides of the stage, and the Marth keeps stalling, the odds of this scenario go up drastically. If the Fox hits Marth off the same side twice, it would be nearly impossible for the cloud to return in time for Marth to recover. I don't know if the players were thinking about these tactics in the moment, but it certainly is possible to make intelligent decisions to minimize the "luck" in this scenario.

[@drowsemary]

btw shy guy di is only for throws ! not just any attack into the shy guys which if ur going by vibe math marth choosing to grab is another factor

[@drowsemary]

this is cuz when u are thrown your character gets a hitbox in a way just getting sent by an attack does not

[@Niko-bz3df]

"Disgrace him" God damn

[@Taygon45]

I'm liking purely based on the Gravity Rush music. And the fact that this is a hilarious video about something that I actually think about sometimes.

[@JeramiJoseph]

this video was wild to watch. keep up the awesome vids dawg

[@qwertyugzb]

I thought the peach on stadium clip would involve the likelihood of her dropping the bomb, is that guaranteed?

[@qwertyugzb]

also 127/7424 times 1/128 is definitely not 1/22447 that makes no sense? 1/128 times 128/7424 would be 1/7424 so even without checking a calculator it'll be more likely than 1/8000

[@tomrejor4842]

​@@qwertyugzb that's because there is a lot of confusion between the number used and the number presented and the number used, the number that was actually used for bomb probability, was the one on the far right, just as he did for the Turnip, which end up being 2/768 (can be seen at the bottom hidden when presenting the stitch face odds) and when doing the math it ends up with the right number.

[@ChalerYT]

i have learnt this is random after posting this which sucks and as u noted yes i type the wrong number in the video even after multiple re watches stuff still slips through the cracks when trying to catch up on my posting sorry bout that

[@tree2992]

I love your videos, and I like how you analyzed probability in decision-making, not just RNG. But your math is pretty wonky. Next time consult with a stats nerd friend.

[@Melkimund]

I thought the interesting odds lied in the chance of pulling a stitch then a bomb just in time for it to be detonated

[@MrGetownedLP]

12:46 "Did I say how many right yet? I lost zeroes." Xqc viewer so I know

[@ChalerYT]

thank you i laterally couldn't decipher my own words

[@Andy251153]

Wow so shy guys timer is relative to when they are cleared from the screen

[@eizzah8323]

So it's not 0.00000074 % but 0.000074 % Basically 1/1296000 = 0.00000074 in decimal however to get percentage you have to multiply by 100 (0.1=0.1*100=10%) which gets you 0.000074% (or 7.4e-5 % 🫠) 8:25 And so for Marth randall it's 0,00068% For shy guys 0.0024% And in conclusion the luigi link shy guy randall : 0.0000051%

[@ChalerYT]

please no.....

[@jmanfro1]

I think you’re thinking of Val spray. Cs spray always follows a pattern.

[@simanolastname2399]

This goofy statistics makes me want to die lmao

[@kalarse]

I love Vibe Based Match

[@wusstunes]

Chess is luck based. White piece op fr fr!

[@x9x9x9x9x9]

Screw math. Melee is sick though.

[@owenmcgovern8140]

bro is just making up numbers 💀

[@TooMuchDad]

Vibes based math should be a new college major tbh

[@LiquidJeans]

good video

[@CC_-pn2og]

Yooo shout-out to adef. Especially love his d&d series speed runners and dragons. Great mow the lawn to

[@coldnova1037]

I going to say vibe based math makes sences its how roll back net code works or sports betting works a player could bring a gun on to a feild and kill his defender but thats not going to happen its possible but highly inprobible

[@mannysolis9018]

ㅤ How do I rocket jump in Team Fortress 2? Rocket jumping is performed by looking behind your character, usually next to a wall (you can rocket jump off a lot more props and stuff), then shooting the rocket, and at the same time pressing jump and crouch. Rocket jumping is one of Soldiers most important, versatile, and skillful moves in TF2. The only thing that rivals the power of rocket jumping's speed is sticking jumping. How do I stickjump in Team Fortress 2? Sticky jumping is performed by doing the same actions as a rocket jump but in a slightly less complicated fashion. First, you have to place down a sticky and wait for the Aug timer of 0.9 seconds (this is the Aug timer for the original sticky bomb launcher, not including the Scottish resistance or the Quickie bomb launcher). After the Aug timer is finished, you can simply go behind the stickybomb(s) and press mouse 2 (the right mouse button) to detonate the sticky, then jump and crouch. Sticky jumping is also one of Demoman's most important skills in TF2. Both of these ways are the simplest ways of rocket/sticky jumping, but there are many others. Why do I have to crouch/not hold W when rocket/sticky jumping? The reason you have to crouch while rocket/sticky jumping is because of this property called the bounding box. This property controls the weight and mass of every character in TF2, and how it works is that it checks to see a character's weight/mass is by checking where the box is in relation to your character, so while crouching, your collision box gets smaller, making your character lighter and making you jump further. This also applies to other enemies bullets/projectiles, because you weigh less while crouching, which causes you to go further, which can let you surf other people's rockets and bullets. And The reason you can't hold w while rocket jumping is simply because if you do, then you won't be able to strafe (turn) while rocket jumping. What is C-taping, and how do I perform it? C-tapping is a TF2 exploit that lets you get more height from any exposive jump by holding crouch for a short duration and then releasing it, and while you are in the middle of your unducking animation, you press jump and fire a rocket. This works because when you C-tap you are closer to the ground. The reason C-taping brings you closer to the ground is because when you uncrouch before you are ever fully crouched, the game does a check to make sure you have a place to stand up in, but if you jump, then the game assumes that you are trying to uncrouch in the air, which is usually instant. But because the game thinks you are uncrouching, it checks where your bounding box is extended to make sure you have room to uncrouch, and while mid air this check brings you closer to the ground, making you closer to your rocket and making you go higher than normal. What is a wall bug, and how do I do it? Wall bugs are possible due to the weirdness of how TF2’s walls work. When you rotate your character, your bounding box does not move with it, so when you are facing a diagonal wall, your bounding box can hit the clip of the wall, causing a wall bug. To do a wall bug, you will need to perform a rocket jump up to a diagonal wall. You will want to hold W and continuously press A and D until you are stuck. If you get stuck, continue to hold in the direction you are moving in; for example, if you get stuck while holding A, continue holding A because if not, you will fall off. The same applies if you get stuck holding D, then continue holding D, and make sure you dont move to slowly or too fast because if you move too slowly, then you will have fewer chances to hit a wall bug, and if you move to fast, then you wont have enough time to react to which straf key you have to keep holding. The reason that this is even possible is because there are two systems, one checking that you are colliding and one constantly checking if you are getting stuck. If you move side to side on a flat wall, then only one coordinate changes, but if you move side to side on a diagonal wall, then the game changes two coordinates. And the two numbers (cordinates) changing rapidly cause the collusion check and the stuck check to not work correctly, and zero’s all of your momentum's velocity, keeping you trapped because you're not actually stuck; its just the game hovering over you because it doesn't know what to do, causing you to stick to the wall.

[@gengarrrx3]

OCTOPUS ❤

[@Zophuh]

WE SO BACK

[@gehteuchnichtsan7911]

The most unscientific math ever haha. You cant just take numbers you like and multiply them. There are so many more possibilities that could have happen every frame. Its ridiculous to calculate them in a game like melee

[@DaaSush]

The math Behind the windmill Situation ist just wrong. 1/4 Times 1/32600 Equals 1/129600 not 1/1296000. And dezimal numbers and percent arent the same, vor example 0.1 Equals 10% not 0,1%. So the actual percantage is 0,00077%, which is different to your calculations by Factor 1000. And as others Pointed out, the first Event cannot happen if theres is no windmill, so you cant multiply them.

[@KOxArtist2]

One problem I have with your video and Adef's is that you don't put the astronomical odds into context. given the number of hours of melee that's been played on stream, how many times would we expect to see rare events happen? For example, a stitch is 1/58. What happens if peach pulls 50 turnips every match? (57/58)^50 = 40% chance to see a stitch in a match, and so, envelope calculation, 40% * 1/58 = .6% chance to see a double (not so far off from 1/58 now). What both your videos are missing are the context of all the unextraordinary events -- making the number more realistic would help me appreciate the rarities more. There's so much more to probability than just multiplying independent events. For example, how many peach games have been streamed live? How many would a viewer need to watch in order to have a 50% chance of witnessing a double stitch? (1-.6%)^x = .5, solve for x. After watching 115 games, a viewer would have a 50% chance to see a double stitch. I think this "100+ peach games" idea is a really powerful idea that can give your viewers real context on how unlikely these events are. And then you can continue with accounting for peach dittos. Maybe you can learn some python, or R. I saw your last video, and I think you can do better than Adef by expanding on his idea with melee's raw statistical content.

[@KOxArtist2]

And honestly, stats formulas and python shouldn't even be that hard to implement with chatgpt as an assistant

[@KOxArtist2]

Like what if you used python to track shy guy locations and say something like "I watched 100 matches and shy guys are in this spot of the map .5% of the time"

[@tomrejor4842]

I will say no matter the odds of everything, Armada "disgrace him" clip on Mango is super impressive even with just raw number because it's a unique situation in which for the punish to work as it did it needed to pull a doubl stitch, Amanda just went for it being like "yeah I'll get the double" and got it making the punish awesome. But I completely agree with you on the putting things in perspective. The difference of "likeliness that it happens to me right now" and "likeliness that it happened to whomever whenever during the game lifespan" are 2 completely different stats

[@thebigdug]

there is a typo in the video and i will not tell you where :)

[@chicha400]

There is a typo in your comment and I will not tell you where :)

[@Sidereon]

Is this the new standard having a dude stand in front of the footage playing instead of just the voice for ego, I hate it.

[@meamking]

Maths

[@Readesu]

Hi.

[@pixelbushbebe4619]

usually a big fan of ur stuff but i think you missed the mark here, you clearly dont understand probability and apply it weirdly all over this video. Keep doing melee stuff its great but please do not talk about statistics ever again

[@Fedivdicher]

This video is hilarious to me cause I remember I was up 1 on a peach player, wasn’t very close, next game he pulls not 1 not 2 but 3 stitches and a bomb, it’s seems like item play is the only thing he’s good at cause hes good at cause it’s the only way he could take stocks, next game fucker pulls 2 more stitches and a beam sword and 3 stocks me cause that is the only time I’ve ever seen a beam sword In tournament. At the time I thought I was very unlucky and his item play was just on point, but looking back I should have checked the console cause it seemed like item play was the ONLY thing he knew, for reference this was a grimey local in a grimey region, there was no real over site and half the consoles were running 20XX with the ugliest skins imaginable. After the Ohio pichu incident nothing is off the table

[@cyclone8411]

I don't wanna completely dog on you any more than the other comments already have, but this video could have been a really fascinating dive into some really cool math and stats, but instead it was full of arbitration and pointless calculations based on vibes that do nothing but bloat your numbers. i'm sorry but this video is really just pointless and uninteresting. I read the pinned comment and do think you should have been more clear, but more than that I think this videos title is also pretty misleading. I would have clicked off once you explained you weren't gonna be doing real math for most of this. again I want to clarify that i'm not trying to come across as harsh, I truly enjoy your content and will continue watching of course, but this video was disappointing in many many ways, even if you had left clarification in I think I would have been disappointed in what this set out to do

[@ChalerYT]

Sometimes you miss and this one seems to be one. sorry you didn't enjoy the vid will make sure the next one bangs to make up for it

[@cyclone8411]

@@ChalerYT 100% good on ya for taking it on the chin and keeping on trucking though