Coin Flipping (extra footage)

แชร์
ฝัง
  • เผยแพร่เมื่อ 29 ธ.ค. 2024
  • A bit extra from our interview with Persi Diaconis.
    Main video: • How random is a coin t...
    Second video (catch or drop): • Should you catch a tos...
    Super bowl coin tosses: • 14 Super Bowl Coin Tos...
    Dynamical Bias in the Coin Toss: bit.ly/Dynamica...
    NUMBERPHILE
    Website: www.numberphile...
    Numberphile on Facebook: / numberphile
    Numberphile tweets: / numberphile
    Numberphile is supported by the Mathematical Sciences Research Institute (MSRI): bit.ly/MSRINumb...
    Videos by Brady Haran
    Support us on Patreon: / numberphile
    Brady's videos subreddit: / bradyharan
    A run-down of Brady's channels: www.bradyharan.com
    Sign up for (occasional) emails: eepurl.com/YdjL9

ความคิดเห็น • 57

  • @Sakanakao
    @Sakanakao 9 ปีที่แล้ว +14

    Watched this whole series of Persi Diaconis videos, very interesting topic! Thank you!

  • @cemerson
    @cemerson 10 ปีที่แล้ว +7

    When I flip a coin, after catching it I turn my hand over onto the opposite wrist and then uncover it by taking my hand away, and I think a lot of people do that. Presumably that 'reverses' the bias as it were, so you'd need to know in advance what the 'catching and showing' technique the coin tosser is using surely?

  • @gdasailor4634
    @gdasailor4634 8 ปีที่แล้ว +1

    An additional randomizing feature is how (at what height) you catch the coin or how high you release it if you are going to allow it to fall to the ground. A "well tossed" coin is essentially a 50/50 odds test regardless of any fore-knowledge of the initial condition.

  • @garethdean6382
    @garethdean6382 10 ปีที่แล้ว +17

    Only a mathematician needs extra footage when it comes to coin flipping. I await a video on dripping taps. (Could relate to the geometry of spheres or the predictability of drips.)

    • @anothermoth
      @anothermoth 10 ปีที่แล้ว +7

      Actually dipping taps can be pretty interesting. At very low flow they are regular, but if you increase flow rate so that the next drip is due before the oscillations from the last have died away they can be chaotic, and then settle down to regular again once flow rate has increased more.

    • @garethdean6382
      @garethdean6382 10 ปีที่แล้ว +3

      anothermoth
      Indeed, and that's the point. There's probably some real interesting math buried there. And drips were used in water clocks, but exactly how regular are they?

    • @saemj
      @saemj 10 ปีที่แล้ว +3

      You sound like the kind of person who would enjoy 60symbols video on paint drying :D

    • @garethdean6382
      @garethdean6382 10 ปีที่แล้ว +2

      Saemj Baumgartner
      I'd like to see one on coffee stains drying, the mathematics and physics behind how it forms a ring (and not, as you'd expect an evenly colored stain) is fascinating.

    • @DeathBringer769
      @DeathBringer769 6 ปีที่แล้ว

      +Saemj Baumgartner
      If you paint houses/apartment buildings/etc for your living then the science/math/engineering/etc of paint drying (and thereby the implications of how its made affecting that property) could be very fascinating or even potentially useful for you... I can tell you from experience paint doesn't always dry the same speed depending on what typed you used, the temperature and also how humid it is out (obvious), how much paint you used, what the condition of the surface you painted was, what the actual material IS that the surface is constructed from, etc. It might seem boring and obvious to a lot of people at first glance, but there could be interesting science or math there waiting to be found for those willing to put in a little intellectual or academic effort ;)

  • @jethrojangles9541
    @jethrojangles9541 10 ปีที่แล้ว +1

    Love the "You just watched it!!!" on the link to this video at the end. Hey, you linked it, buddy!

  • @brianpso
    @brianpso 10 ปีที่แล้ว +1

    I went all in that annotation, and I lol'd very hard when I saw the "You just watched it!" on there. Damn Brady, don't play with my feelings man.

  • @EebstertheGreat
    @EebstertheGreat 7 ปีที่แล้ว +2

    You can't "load" a coin, but you can surely warp a coin to bias its outcome. Imagine a coin so badly warped that it has been bent into the shape of a thimble. Clearly it is more likely to land with the convex side down than with the concave side down.

  • @oafkad
    @oafkad 10 ปีที่แล้ว

    The Extra Footage stuff is always like a present to me.

  • @IceMetalPunk
    @IceMetalPunk 10 ปีที่แล้ว +3

    So air resistance is probably, but unprovedly, negligible, but weighting on either side of the coin is provably noncontributing? That's really weird...but good to know!

  • @BoogerBarn
    @BoogerBarn 10 ปีที่แล้ว

    I would assume that not only would the barometric pressure play a factor...but also the humidity...as it could have impact on the angular momentum.

  • @vinniciusg
    @vinniciusg 10 ปีที่แล้ว

    2:35 Artistic picture haha

  • @Faxter313
    @Faxter313 10 ปีที่แล้ว +3

    "You just watched it!!!!"

  • @tsunghan_yu
    @tsunghan_yu 5 ปีที่แล้ว

    what is third order effect?

  • @letdomf
    @letdomf 5 ปีที่แล้ว +1

    I can't believe asymmetry (head tale) of a coin doesn't affect the outcome of a flip.

    • @paulolameiras5298
      @paulolameiras5298 4 ปีที่แล้ว

      Everything affects the outcome of a coin flip. Being the structure of the actual object being used or tested and/or even the process that we adopt in order to fulfill are test. That's why I always say that it is almost impossible for man or anything that is man made or man originated to be truly random.

  • @palmomki
    @palmomki 10 ปีที่แล้ว

    Yeah, I just watched it, I was thinking the same thing. Thank you.

  • @davidsweeney111
    @davidsweeney111 10 ปีที่แล้ว +1

    good god I thought this was Richard Feynman, must be the beer ....

  • @vinniciusg
    @vinniciusg 10 ปีที่แล้ว +1

    I completely understand the point of professor Diaconis about air resistance. The idea of solving Navier-Stokes equations for the complex motion of a coin seems somehow absurd.

  • @naamloos246_
    @naamloos246_ 10 ปีที่แล้ว

    i knew there was a extra footage video

  • @B3nnub1rd
    @B3nnub1rd 10 ปีที่แล้ว

    Ooh! Underwater. That's intriguing...

  • @AnnaDamm
    @AnnaDamm 10 ปีที่แล้ว +1

    What about when you do not change the weight on one side, but change the aerodynamics, by file off the edges on one side? make them more round on the one side, that should change the odds, no?

    • @chillsahoy2640
      @chillsahoy2640 10 ปีที่แล้ว

      That's what I was wondering. While I may not understand the mathematical reasoning behind it, I can believe that if a two-metal coin shows no preference to land on the lead side, then weight plays no part in the flipping process. But the heads and tails patterns, or a completely smooth side, might have an effect due to aerodynamics. I'd be curious to see if anyone has tested this or constructed mathematical models.

    • @Melomathics
      @Melomathics 10 ปีที่แล้ว +2

      Yep, by 42.

    • @ragir
      @ragir 10 ปีที่แล้ว

      It seems that the way the coin is flipped makes this not the case - if you look from one side, both sides of the coin are visible for the same (well, almost) amount of time, thus making those differences cance eachother out.

    • @tryfinally
      @tryfinally 10 ปีที่แล้ว

      Aerodynamics already plays a role in the process, but because the coin is fairly heavy, the effect of the coin's slight asymmetry is negligible. If you altered the coin's shape, the effect would surely be greater, but obviously, it's difficult to predict by how much.
      Naturally, if you had a coin made of wood, or even better, a coin made of paper, then aerodynamics would make a much bigger difference. You can easily imagine that a paper coin would prefer to fall like a leaf instead of maintaining its angular velocity.

  • @U014B
    @U014B 9 ปีที่แล้ว

    Special appearance by Audrey the Adorable Chihuahua.
    Also, what about coin tosses in a vacuum?

  • @gillbates21
    @gillbates21 2 ปีที่แล้ว

    if there was a competition to see who can flip heads the most times in a row and the winner would have a score of 50+, would it change your mind that coinflip is really random? y/n?

  • @BlinkplusAvA44
    @BlinkplusAvA44 10 ปีที่แล้ว +1

    So this means that, Cellphones cases should have enough mass to change the center of gravity, so this way the cellphone will tend to land on the case side is a hoax?
    I would really like an answer it is pretty interesting.

    • @tryfinally
      @tryfinally 10 ปีที่แล้ว +1

      If you throw an object in a way that makes it spin around its center of mass, you don't see it slow down and speed up its rotation. It just rotates at a (roughly) constant speed. And if it does, why would it be more likely to land on one side and not the other? The effect that alters the coin flip chance probably has the greatest influence here as well; your phone has a slightly larger chance to land in a certain position depending on how you were holding it when you dropped it. Don't drop your phone.

    • @BlinkplusAvA44
      @BlinkplusAvA44 10 ปีที่แล้ว

      apk Yeah that is what I was thinking.
      Thank you very much for clearing it up! Really helped

    • @Nibr24
      @Nibr24 10 ปีที่แล้ว +1

      In my experience though such heavy objects like a phone don't spin at all if you drop them, so never watch its back and it will most likely land on the case. And obviously, don't toss your phone.;)

    • @tryfinally
      @tryfinally 10 ปีที่แล้ว +1

      Alonsinho Gomez Darts (and arrows) have two sides: one side is heavy, and has a large momentum. The other side is very light and has a relatively small momentum. Because of this, the rear side of the dart is affected by aerodynamics much more strongly. As a result, when you throw a dart, the tip will "pull" the rest of the dart with it, while the rear will be pulled back by the drag it produces. This effect stabilizes the dart.
      This is different than a falling phone or coin because darts are specially constructed to be affected by aerodynamics. Meanwhile, phones and coins are mostly symmetrical and unaffected by those forces.

    • @BlinkplusAvA44
      @BlinkplusAvA44 10 ปีที่แล้ว

      apk Thank you!
      But can you build a case that can make a phone more aerodynamic friendly?
      For example a case with a form of a P could potentially change its center of mass towards the semi circumference of the P
      Do I explain Myself?

  • @culwin
    @culwin 10 ปีที่แล้ว +4

    What if I use a Bitcoin

  • @fumitakahayashi
    @fumitakahayashi 9 ปีที่แล้ว

    what's up with the slo-mo coin-flippers thumbnail?

  • @BX138
    @BX138 6 ปีที่แล้ว

    Do they have a caveman flipping those coins in the slo-mo footage? Sheesh!!!

  • @hingedelephant
    @hingedelephant 8 ปีที่แล้ว +1

    Every comment has a hand that just flipped a coin. 👍🏻

  • @pdrgallo
    @pdrgallo 8 ปีที่แล้ว

    pretty sure he could make super tight cristal

  • @SoftBreadSoft
    @SoftBreadSoft 6 ปีที่แล้ว

    It's only fair if the person that is flipping hasn't practiced coin flipping. It's no more random than your ability to get a basketball in a hoop. Unless there's considerable wind.

  • @teeede
    @teeede 10 ปีที่แล้ว

    1. Spend some time finding as many of these minor probabilistic facts
    2. Use them in your favour
    3. ????
    4. PROFIT!!!

    • @Yllipolly
      @Yllipolly 10 ปีที่แล้ว

      You need to be a very patient gambler, if this is your advantage play.

    • @teeede
      @teeede 10 ปีที่แล้ว

      You mean in coin flipping? Well, not if you bet real high :)

  • @KosovoJeSrceSrbije89
    @KosovoJeSrceSrbije89 10 ปีที่แล้ว +1

    So the most likely combination that you will get after 100 coin flips if you start with heads for example is (1)heads (2)heads ... (100)heads.

  • @dante224real1
    @dante224real1 10 ปีที่แล้ว +2

    i think a better probability experiment to do with kids is organize a shiny pokemon hunt. explain to them the odds of finding shiny pokemon in the wild in various games and give them a week to find a shiny pokemon in some way. it would give them a clearer idea of probability 1/8000~, 1/2000~, and 1/1000~ in various time scales depending on their chosen methodologies.
    it would be a fun little homework experiment and would get the kids to play with numbers and probabilities in a way that would reward them in something they would get to keep for as long as they like.

    • @erichernandez6102
      @erichernandez6102 9 ปีที่แล้ว

      *enables "all Pokemon are shiny cheat"* Done, the probability is 1.

    • @dante224real1
      @dante224real1 9 ปีที่แล้ว

      they might not, but in XY its much easier. 1/800ish in the safari and breeding through the masuda method. i think they could get one per group if it was set up in groups hunting. obviously we could plan out what game they would use so each group would have an equal chance. or they could try chaining in platinum

  • @paulolameiras5298
    @paulolameiras5298 4 ปีที่แล้ว

    If use try using different coins you should reach the natural and obvious conclusion that some coins are more biased than others. So the results are never the same. So I don't see the usefulness of doing all these experiments using a single coin.One thing is to prove that a certain coin is fair another thing is to prove that there are processes that you could use that could get you fair results using any coin. Either way it is possible to have a structurally unfair coin and adopt a certain process that takes that into account and is able to create fair outcomes. Although a 50% - 50% outcome is not a natural or a more likeable outcome when dealing with "fairness" or "randomness" because that is not the way it works. The results are never that accurate, because if they were, randomness would be more predictable than we think. A true natural random generator will originate very slight differences between outcomes, but the differences should exist.

  • @Walther9mmultra
    @Walther9mmultra 10 ปีที่แล้ว

    I hope they know that coin they have is 90% silver.

  • @lucashoffses9019
    @lucashoffses9019 10 ปีที่แล้ว +1

    1st comment!

  • @grekiki
    @grekiki 10 ปีที่แล้ว +1

    First