He’s talking about a situation, not where you’re doing live flips, but where you’re reviewing the data of a set of flips that have already happened. He’s saying this reasoning doesn’t pertain to doing live flips.
whether live hyperthetical, past/future hyperthetical, or merely (always) hypothetical (true nihil), an odd # of flips greater than 1 cannot produce 50-50 result
Ah, I get it. An entirely different situation. Because here we know in advance the coming flips. We know their distribution and we know it is weighted towards whatever he said it was.
I was thinking that another way of saying Motor is wrong is to say that the coin does not remember the previous tosses. Nor does the environment. They have not been changed by what went before and have no built up ‘preference’ now for tails (or heads).
Yeah exactly. And imagine if it did. Imagine these three situations:
Situation 1: you flip a single coin 3 times and it comes up heads each time. You wonder what its probability is if coming up heads again on the next flip.
Situation 2: you have 4 different coins in front of you. You flip the first 3, and they each come up heads. You wonder what the probability is that the 4th coin will also come up heads.
Situation 3. A combination of the previous 2. What you do is you “manually weight” the coins with a magic probability. You take the first coin and flip and flip and flip until you get 3 tails in a row, then you set it down on the table. You do the same again for the other 3 coins, so all 4 coins have now all had streaks of 3 tails. Then you go back to coin1. You flip it, it’s heads. You flip coin 2, it’s heads. You flip coin 3, it’s heads. What’s the probability for the 4th coin now being heads, in this world where there’s magical probability memory?
Does the 4th coin want to flip heads because it remembers that it previously flipped 4 tails? Or does it want to flip tails because you’re currently in a head-flipping streak? Where does the magic probability come from? Does it cancel out?
I guess the question is, in this gamblers fallacy world view, how do multiple applicable streaks interact?
This doesn’t really work as an analogy. Coin flips are unlike decks of cards.
If you ask me to pick a card, and I pick a 3, and then you ask me to estimate my probability of randomly selecting a 3 again, of course it will change. The first time it was 4/52, but I’ve literally physically removed a 3 from the deck - it’s not available to select from anymore - so now my probability of selecting a second 3 is 3/51. And if by pure chance I do select a second 3, the next selection has a different probability again, it’s 2/50. When I physically remove 3s from the deck, of course that affects my ability to select more 3s.
Coin flips are entirely unlike that. When you flip a heads, and then you get prepared to flip again, you haven’t physically removed any heads from the coin. The coin has exactly as much heads now as it had on the first flip. And then if you flip a heads again, still, there’s just as much heads on the coin as the first time.
Every flip is a full reset of the coin, physically, whereas every card selection is not a reset of the deck.
The probability is governed by the number of possible options it could flip.
Magic?? Goofy.
Will, however… that can muck up the mathematization
Yeah, the idea that coins remember how they flipped seems like magic to me.
That’s the kind of thing I was getting at. And imagine what would be happening at the subatomic level???
And what if you cheat. Like you do one or two spin tosses. Meaning the coin doesn’t spin many times. Or perhaps you throw it flat, so it’s 100% it comes up heads. You could get good that that, so you can choose, often, how it hits. Does the coin know you are cheating?
And how does the coin remember what is facing up and what is facing down?
You flip a coin and clap it so it is frozen between your hands. One team watching counts that face facing the right hand. Another the left hand. Now you do four tosses onto a table. Which team’s expectations does the coin confirm?
If you have a monkey in the next room tossing does this affect coins in your room?
Or If you toss an excess of heads with one coin and change coins for the next four tosses, are tails still ‘more likely’? Why, why not? What’s the mechanism? Is it me the experiencer who causes it or the coin or some guiding deity of statistics?
1 Like
.
The toss of a coin will always be 50/50… regardless.
…like when a person rolling dice, hits the same combination of results throughout most of the game… so, unpredictable.
1 Like
I interacted to what you wrote and you in return ignored my question to you.
I already acknowledged that both of us agree that a single flip has a 50-50 chance of landing on heads or tails. The chances before the flip are always wrong, because it is either 100% that the coin lands on heads, or it is 0% that it lands on heads. If you bet on heads then you either win (100%) or you lose (0%). It is impossible for the coin to land “half heads and half tails.” It either does or doesn’t. The “50-50” comes from the fact that there are 2 possible outcomes, and 1 flip with 2 outcomes has a 50-50 chance for either outcome.
In MULTIPLE FLIPS you are betting on HOW MANY FLIPS will be heads and how many will be tails.
So I ask you once more, out of 100 flips, how many will be heads and how many will be tails? What is your answer? Care to interact, or are you going to ignore the question once again?
BEFORE the toss the coin has a 50% chance of landing on heads. After the toss the coin is either 100% heads or 0% heads. So the “50-50” before the toss was wrong, it landed on heads, 100%.
You’re talking about after-the-fact/the act/the toss.
…so the probability of 50/50 causing the resultant effect of 100/0, are 2 separate entities/occurrences/states/phases of time of past present future, so the former creating the latter state over a period of lapsed time… capisce! 
.
So the “50-50” before the toss was wrong, it landed on heads, 100%.
See above^
Right. Before the flip you claim there is a 50% chance the coin lands on heads. Then you flip and find out you were wrong, it is 100% heads. So you predicted 50% but it ended up being 100%. Your prediction was wrong. And you were wrong about it being 50% for tails, it ended up to be 0% tails.
.
Before: probability = static… during: transition of state = motion, of the yet unknown resultant end state… after: result = in-keeping with the flow of time/change.
You seem to be putting the cart before the horse, here… why?
Do you want to find out if your predictions were right or wrong? You predicted 50% heads and 50% tails. Now flip the coin and find out if your prediction was right or wrong. I bet your prediction of “50% heads-50% tails” is wrong 100% of the flips.
Probability is not prediction… again, why the conflation of two separate concepts?
Probability is not about right or wrong… probability is exactly what it says on the tin…
One way of thinking of my second scenario is as a “deck of flips”, because it behaves like a deck of cards. The deck starts 50-50 heads and tails, but a run of heads changes what flips are left in the deck.
But that’s only the case when we have a finite set of flips about which we know a global property, e.g. that the whole deck is 50-50 heads and tails.
This is different from just knowing that it’s a fair coin. I can see how the fair coin appears to have this kind of known global property, but for a fair coin it’s only a statistical property that over time the set will tend to be equally split. With the ‘deck of flips’, the probability starts out 50-50, but after each ‘card’ is pulled the remaining deck is a different deck, with a different probability of producing heads or tails. When a head is flipped, the remaining deck has one fewer head, so the change of the next flip being tails increases. With a fair coin, the probability stays 50-50, there’s no ‘deck’ that’s changing with each flip.
@Motor_Daddy, I think this gets at you 100 coin flips question: before we flip the coin 100 times, we can only say what the probability is of possible outcomes, we can’t say for sure what the actual result would be. Results around 50-50 are most likely, but they aren’t certain; a run of heads could be a fluke that leads to a 55-45 split in the final tally.
I am going to flip a coin. What do you predict the coin will land on, heads or tails? So you whip out your calculator and figure that for 1 flip there are 2 possible outcomes, heads or tails. You divide 1 divided by two and you get 0.5 (50%). So you claim the coin has a 50% chance of landing on heads and 50% chance it lands on tails. That is your prediction as you stated earlier, that the coin has a 50-50 chance for every flip. But you are wrong 100% of the flips, because for every flip it is either 100% or 0% that it lands on heads, never “50-50” as you predicted before the flip.
Yeah sure, if it’s a preset set of flips that you already have explicit statistics about, then absolutely.
The base scenario is about an yet-to-be-flipped flip, of course.
There are 3 phases to probability, not 2.