Misdirection sponsored by Camelot
Well Derren Brown is predicting the lottery numbers. Live. In only an hour. This man is brilliant, he reminds us that he is a master of misdirection. I think he misdirects us on a higher plane than we think. He always reminds us that what we see isn't necessarily what is actually happen. No doubt in an hour we will think the numbers written on paper in a sealed envelope remained in that envelope for the entirity. There are 13,983,816 possible combinations to play. There is no way he will "predict" the numbers.
This is an interesting one and has inspired me and my sciency buddies to get into a nerdy debate. Statistics is horibbly complex. I mean when you forget about finding means, medians and modes the actual nature of statistics gets horrible. Right so the chance of winning the lottery is 1/13,983,816. Great. Easy. Also every combination has an equal chance of winning. So you can chose 1, 2, 3, 4, 5, 6 or 1, 5, 13, 20, 37, 43. Both have an equal chance of winning. (Interestingly it's better to choose random numbers. Thousands of people pick sequential numbers. This means if they win, the winnings will be split with more people and you will win less).
The dificulty comes when you play the lottery the week after you win. What numbers do you pick? Surely you can't win twice with the same numbers? That must be impossible! Surely the odds will be (1/13,983,816)2? This comes out as a chance of one in 2 x 1014. If a weighted coin gave a "tails" only one time in every 2 x 1014 throws and asuming you threw the coin every second, you would be tossing the coin for 32.7 million years before you got a tails. Wow pretty improbable right?
Well here's the head fuck. The numbers are independant of one another so the next week each combination has exactly the same probablility as the next. That's right, there is still a 1 in 13,983,816 probability of the same combination you have picked. There's no misdirection like the misdirection found in statistics.
So if you win the lottery with one combination one week, would you play those same numbers the week after? Even if you know the odds of winning with your combination is exactely the same the second time around? There's no misdirection like the misdirection found in statistics.
This is an interesting one and has inspired me and my sciency buddies to get into a nerdy debate. Statistics is horibbly complex. I mean when you forget about finding means, medians and modes the actual nature of statistics gets horrible. Right so the chance of winning the lottery is 1/13,983,816. Great. Easy. Also every combination has an equal chance of winning. So you can chose 1, 2, 3, 4, 5, 6 or 1, 5, 13, 20, 37, 43. Both have an equal chance of winning. (Interestingly it's better to choose random numbers. Thousands of people pick sequential numbers. This means if they win, the winnings will be split with more people and you will win less).
The dificulty comes when you play the lottery the week after you win. What numbers do you pick? Surely you can't win twice with the same numbers? That must be impossible! Surely the odds will be (1/13,983,816)2? This comes out as a chance of one in 2 x 1014. If a weighted coin gave a "tails" only one time in every 2 x 1014 throws and asuming you threw the coin every second, you would be tossing the coin for 32.7 million years before you got a tails. Wow pretty improbable right?
Well here's the head fuck. The numbers are independant of one another so the next week each combination has exactly the same probablility as the next. That's right, there is still a 1 in 13,983,816 probability of the same combination you have picked. There's no misdirection like the misdirection found in statistics.
So if you win the lottery with one combination one week, would you play those same numbers the week after? Even if you know the odds of winning with your combination is exactely the same the second time around? There's no misdirection like the misdirection found in statistics.
Comments