basic question but need good explanation please

[SpaceAce]

[ QUOTE ]

[18:58] b: some time there is written theory, and application of the theory.


[/ QUOTE ]

That is just awesome.

Do what others have already suggested and bust your friend by flipping a coin. Make him agree to at least X flips ahead of time then punish him for his ignorance. Your friend is wrong and either too stubborn or too dim (both?) to see it.

Does this friend even exist? The silly self-contradicting justifications for his theory seem almost cartoonish.

SpaceAce

[poker12]

[ QUOTE ]


Does this friend even exist? The silly self-contradicting justifications for his theory seem almost cartoonish.

SpaceAce

[/ QUOTE ]

That's funny you say that, I actually did not want to post some of our conversation because I thought that 2p2 would think i was just antagonizing and trolling.

But he does exist and I think I'm going with SplawnDarts post and that "This kind of [censored] cannot be "explained" to someone who doesn't want to learn."

On a side note he is a losing 10nl player.

[PairTheBoard]

Come to think of it. If a fair coin is flipped and comes up heads 6 times in a row I believe that I too can predict the next flip will be a tails. Not only that, but I'll be right fully one half the time. I see. It all makes sense now.


PairTheBoard

[Alexpoker]

I'm also an engineer.
Let's make a little programm under excel.
In colums A : = ENT( AlEA()*2) that give number 0 or 1 (50% / 50%)
Then you searsh the number ot times where there are five times in a row 1. Then you calculate the % that there is 0 after.
Answer : 51%.

Moreover, theory of probability proves that even if you have 100 loss coin flip in a row, the next one is 50/50.

Why ?? If the average is 50/50 and if you have 100 loss coin flip in a row,then you have to be lucky after to get closer the average ??
=> be lucky : NO !!!!!
=> get closer to the average : YES !!!
but not in being lucky 10 times in a row but in the LONG RUN.
If you play 10 coin flips and you lose 9 of them =>
90% of loss
theory => 50% of loss
Then you play 10000 coin flip and you loose half (50000). (you have not be lucky, not unlucky).
Then you loose 50000+9 hands and the number of hands you play is 100000+10
So you lose 50009/100010 = 50,00399 % !!!!!!!!!!!!!!!!!

Sorry for my english which is not very good. i hope some of you have understand.

Alex. Beginner poker player

[Alexpoker]

Just ask this question to your friend.
If you lose coin flip 10 times in a row (he thinks that then he will have more than 50 % chance to win, let's say 60%), then what if his opponent has also lose 10 times in a row ?? The two players have 60 % of chance to win ??

[kevin017]

flip a coin a few times until there's an imbalance, ask him what the next one will be. switch coins, ask him if that one is still more likely. give the coin to someone else, ask them if they flip this coin on there own, is it still more likely.

when you first see him, bring out a quarter and ask if theres any possible way that this flip right now is not 50/50, any way at all. There's no possible way to predict that one is more likely than the other? Then tell him, before you got here, you flipped heads 5 times in a row, so tails should be more likely according to you, which you just swore was impossible. etc etc. you could try this same example in terms of a poker table. Every time you sit down, you think of things as being a blank slate, every hand is just as likely as every other. But someone who has been sitting there for 10 hands already saw that the hand before you sat down flopped aak or whatever. So, the odds of him seeing aak again are lower than yours?

[poker12]

I think I might have to take him up on his bet:


[14:15] b: I think you are having a difficult time moving from the narrow vision that a coin is 50/50. It will always be 50/50! I have never said different. But we know that in time, and the more flips we execute, tails will turn as many times as heads will. The more we flip, say to the millions, that % will get closer and closer to 50%. It doesn't matter that I have been absent for all the coin flips in a life time, I should be able to jump in at any moment, and grab a sample. and within that sample there will be a % that is working its way to 50%. So regardless, if a coin is flipped and shows heads 100 times in a row, I will put my money on tails. It is still 50/50. if it flips head again, I will bet double on the next flip that it will be tails, and so forth, but it is still a 50/50 chance.

[14:20] b: the rest is me making an educated guess on what the next flip may be on past events.
[14:22] b: what creates the past event is the human interaction with the coin being flipped, or the cards being shuffled

[MegumiAmano]

Why do you not have all of his money already?

[SpaceAce]

[ QUOTE ]
I think I might have to take him up on his bet:


[14:15] b: I think you are having a difficult time moving from the narrow vision that a coin is 50/50. It will always be 50/50!

[/ QUOTE ]

Come on, you're making this up. No, I guess you're probably not making it up.

Every time I determine to be done with this silly argument, your friend manages to say something stupid enough to compel me to reply. The probability of a given coinflip outcome cannot be both 50/50 AND weighted toward one outcome over the other. How can your friend seriously say "It will always be 50/50!" and then go on to say that he can predict coinflips because the coin will be biased toward one outcome over the other? Unless your friend has has some understanding of nth-dimensional reality that the rest of us lack, there is no way for the probability of the coin landing heads to be both 50/50 and 52/48 or 60/40 or whatever.

This, though, is truly the central jewel in your friend's crown of ignorance:

[ QUOTE ]

It doesn't matter that I have been absent for all the coin flips in a life time, I should be able to jump in at any moment, and grab a sample. and within that sample there will be a % that is working its way to 50%. So regardless, if a coin is flipped and shows heads 100 times in a row, I will put my money on tails. It is still 50/50. if it flips head again, I will bet double on the next flip that it will be tails, and so forth, but it is still a 50/50 chance.

[14:20] b: the rest is me making an educated guess on what the next flip may be on past events.
[14:22] b: what creates the past event is the human interaction with the coin being flipped, or the cards being shuffled

[/ QUOTE ]

Here, your friend is claiming that the million or billion flips the coin experienced in its lifetime are of absolutely no significance but that the ten flips he himself "jumps in" and witnesses are. How absurd is that? Your friend truly believes that you can predict ("make an educated guess") a future coinflip based on the coin's previous flips but he also believes that only the flips he has personally witnessed matter in determining what will happen next. Does your friend, perchance, have a healthier than average ego? Your friend then throws in a little Martingale for good measure. He ends the bulk of his message by acknowledging that 50/50 nature of the proposition and then goes on to once again claim he can make an "educated guess" at what is more likely to happen, the 50% event or the other 50% event. Huh?

Finally, do I even have to point out how completely and utterly his final statement has nothing to do with anything discussed? It doesn't matter what created the past events because those past events have no effect on future outcomes.

SpaceAce

[PairTheBoard]

[ QUOTE ]
But we know that in time, and the more flips we execute, tails will turn as many times as heads will. The more we flip, say to the millions, that % will get closer and closer to 50%.

[/ QUOTE ]

I think this is the crux of his misconception. The second sentence does not imply the first. The number of heads and tails does not have to even out for the Percent to work closer to 50%. In fact, the Number of heads and tails can actually get MORE out of balance while the Percent is working toward 50%. The reason is that the Standard Deviation increases in proportion to the square root of the number of flips.

For example, you could start out with 100 heads in a row. That's 0% tails. After another 9,900 flips you might be an additional 300 flips out of balance with a total of 5200 heads and 4800 tails. But now you have 48% tails. The percent has worked its way toward 50% even while you happen to have again flipped more heads than tails. After a total of 1,000,000 flips you might be even more out of balance in the number of flips with an enormous 2000 more heads than tails for a total of 501,000 heads and 499,000 tails. Yet the Percent has worked its way even closer to 50% with 49.9% Tails.

The fact is that after an initial 100 heads in a row you can indeed expect the Percent to grow closer to 50% after another 100 flips, 10,000 flips, and 1,000,000 flips. But you are just as likely to see this happen with more heads as with more tails. It's not the evening out of the number of flips that does it. It's the slow growth of the standard deviation and the swamping of the original 100 by the large number of flips when forming the percentage.

PairTheBoard