**1**. Start betting from the 1st try and keep doubling the bet after each miss. Or stop when you succeed.

**2**. Start from the beginning without any wager for as many tries as you like before the 10 , then start betting whenever you feel you are ready.

What would you chose?

**A**. Start betting from the beginning and have the opportunity to win $20 if you succeed in the first 5 bets? (after that you don't have any more money to double the bet. And you lost $620).

**B**. Use your free trials and throw the ball for 6 tries and lose, ( cost $0), then wager in the last 4 tries where success is 100%.

The obvious choice in B.

It's a very poor example to use, and misleading, because in this case there may be dependence between throws, whereas there isn't any dependence between roulette spins. If outcomes are independent the choice of B would give no advantage at all.

No it's not a poor example. I specified 100,000 ball throws.

By then whatever basic skill was to be gained was gained already in probably less than 10,000 tries.

Beyond that and considering the great distance (where skill is not as important as the strength to reach the basket area), all throws can be considered independent. You can have the wrong angle and still get the ball in. You can have the right posture and still miss it. It is the distance that makes each throw an independent event. (even if you were Michael Jordan).

The point is all events have an average of the desired outcome. And neither roulette events can escape the power of averages. Independence is a plus rather than a barrier as far as averages are concerned.

It's no wonder that everything comes to equilibrium given a certain number of spins.

Roulette has 37 numbers, and what happens between spins has to account to an average.

Which is computable.

If a wheel had 150 numbers you'd be seeing 20 and 30 red in a row much more often than when you see it with 37 numbers.

At the same time if you had 2 balls in a bag, one red one black, pulling one at time and replacing it, chances are you will very rarely see 8 black in a row, but in roulette you see it often.

It is therefore the duty of the player to study the average statistical behavior of his particular system.

Then and only then he can gain the much sought after EDGE.

Ok, so what is the best average range of bets for an EC? A dozen?

it depends on the bet:

If the bet is totally random and changeable between spins, a randomly picked EC ( in every spin), should appear in the range of 1-5 spins. Try it and see if you can lose 5 times in a row.

A randomly chosen dozen (in every spin) should have a range between 1 and 8 spins to appear.

Now if the bet is biased (like a specific same EC or a specific same dozen that depends on a trigger, things are different.

It has a lot to do with what that trigger is. So there is no general answer unless you specify the trigger.