Originally Posted by

**recyclersteve**
It is actually quite simple to calculate the odds. The odds of two coins facing the same direction is 1 in 2. You can have heads/heads or tails/tails. You could also have heads/tails or tails/heads.

3 coins- 1 in 4

4 coins- 1 in 8

5 coins- 1 in 16

You keep doubling the number and it will get to be a pretty staggering number in a short while.

This is like the old story where someone says that if you give someone a penny and double it each day for 30 days that you will have over a million dollars in a month.

The tricky part about this exercise is in considering extraneous items that might skew the results. Someone in this thread mentioned dealing with someone with obsessive/compulsive disorder, for instance.

I was pretty clear. You can't calculate the odds that way.

It's been ages ...but if i remember correctly ... they used the example of a coin toss in college stats class to demonstrate the flaw in the reasoning.

Let's look at in a different way.

Get fifty pennies from a roll. Shake em' in a bag. Blindly draw one out and do a random coin toss. There's a 50/50 chance it will land heads up.

Do the same for the remaining 49 and keep track of the results.

Odds are that 25 of them will land heads up and will 25 will land tails up.

There will be some random variation, but with a large enough sample, the variations will average out and you'll hit that exact 50/50 ratio.

Try it for yourself.

The same basic principle should apply to coins randomly tossed into a coin roll. If you're not seeing that 50/50 ratio then there must be some other variable in the mix that's influencing the outcome.

I'm not disagreeing that somebody might have already culled the roll before you got to it. That's one possible explanation among many possible explanations. Truth be told .... you're most likely right on that call if you've been through millions of coins over the years. Experience is a good teacher. You probably know your stuff.

Correctly calculating the odds to a staggering number ? Hummmm ... well ... maybe not so much ?

Don't know about you but i'll happily leave that task to a guy with a PHD in statistical analysis. He could get a federal government grant and spend years of his life researching the penny roll problem.

## Bookmarks