The Roulette Wheel (probability) question

Red, Black, Odds, Even, 1 to 18, and 19 to 36 all have a 47.37% chance of winning.

If you want to play the odds game, play craps. If I remember my notes from highschool, it has a 51-52% winning chance.

I thought that someone was gonna reply with a factual post that would have obviated mine (which I didn't end up posting). Anyways...

All casino games are tilted in favour of the house.

For your original question, Anthony, of

I'll answer the "red won't come up just once" part. From some brief research on the web, there are typically 37 spots on a roulette wheel with one of them colorless.

Therefore in a given spin, the chances of you LOSING are 19/37, or about 51.3 per cent.

The chances of you losing all seven spins (ie coming up BLACK or NO COLOR) is

(19/37)^7 = 0.00941592820025755468171924197107512

Or just under one per cent.

No, they don't. The table has no memory. The odds of red coming up on any given spin are exactly the same as on every other spin. It doesn't matter if black has come up on the last 1,000 spins, the odds of red coming up on the next spin are no better or worse than before. Just as with flipping a coin. The "odds" (probablities) are 50/50 every time. In a series of 100 flips you would expect to end up with roughly 50 heads and 50 tails, but if you do the experiment you will almost never end up with exactly that result. And if you've flipped the coin 90 times with a 45/45 split there is no reason you can't get 10 "heads" in a row at the end, and the chances of that last flip producing the "tails" result that it "should" are exactly the same as its coming up heads again. The odds don't change. Folks who believe otherwise are the ones who pay the bill for all that neon down in Vegas.

Regards,

Joe

This is one of those systems that looks good on paper but falls apart relatively easily in practice.

Five, six, seven and more consecutive blacks or reds is not at all uncommon, and that's likely all it would take to bust you.

Look at it this way, Anthony...suppose your color misses for six spins in a row. Do you really thing you have the cajones to plop down $640 for a single 50/50 (more or less) wager? Remember, by this point you're already down $630. And, even if you hit this huge bet, what are you up overall after sweating this one out? That's right: $10.

Good luck,

Jon

I thought that bill was paid for by the Hoover Dam.

Joseph laid out the statistical error with the "double" approach. The practical problem is that casinos have table min/max limits; at some point you wouldn't be able to get down your next bet to cover.

Do it with horses and start with a $2 bet

It falls apart because of the unjustified assumption that past results affect future outcomes. You're still not getting the point that the odds don't change based on "x" number of spins. Each spin is an isolated event, independent of all previous and subsequent spins, and any "system" that assumes that certain outcomes have to obtain within a given number of spins is simply wrong.

Regards,

Joe

The way it's talked about here, I would think in one run, seven in a row isn't that big a deal. In the long run, however, it seems like could work theoretically. But, what is large enough a sample is hard to determine.

Thanks, Joseph, for bringing up "memory" (or the lack thereof).

It falls apart because on any given spin the odds are against you winning. Over a sufficiently large number of spins, you'll lose 51% of the time. Over a sufficiently large number of spins, you will trend towards losing.

Spinning seven times in a row without hitting red is highly IMPROBABLE but it is very POSSIBLE. It's that possibility of losing where your wallet grows wings.

Well I'm still not getting it either.

No seriously. I understand that even if I flip a coin 9 times and get 9 heads, it doesn't change the odds on the 10th flip. It's still 50/50. However, can't a series of flips be looked at as a whole different animal together?
What I'm saying is what are the odds of 10 flips all being heads or tails? Is that still 50/50? For some reason that doesn't seem accurate. Help me understand.

Besides, everyone knows that everything in life is 50/50. Either it will happen or it won't. What are the odds of me winning the lottery this weekend, well I either will or won't, so 50/50.

Agreed about the probability of a series. It is very low. A good place to try this would be the Lady Luck casino downtown. I think the minimum spin is 50 cents.

However on your other question. If you have already flipped a coin 9 times and all 9 times it comes up heads, then the probability of the 10th time being heads is still 50/50. But, if you have yet to flip the coin, the probability of coming up with heads 10 times in a row is extremely low.

Exactly. The farther you are into your 10 flips of the same side, the higher the odds become that you will be successful. At the beginning, before you've flipped the coin at all, the odds of flipping the same side 10 times in a row is less than one-tenth of a percent ( about .098%). You can calculate this by:

.50 x .50 x .50 x .50 x .50 x .50 x .50 x .50 x .50 x .50 = .00098 which can also be written as .098%.

After you've flipped it once, the odds rise to just under two-tenths of a percent (about .195%).

.50 x .50 x .50 x .50 x .50 x .50 x .50 x .50 x .50 = .00195 or .195%.

The odds keep going higer the closer your streak of 'same sides' gets to 10. After 8 consecutive flips of the same side, your odds of making it to 10 are up to 25%. And after 9 flips of the same side, your odds are now at 50%.

These calculations make the assumption that you are choosing heads or tails BEFORE your first flip of the coin. The odds start out at .195% if you decide to go with whatever side comes up after your first flip.

Or you simply don't have that much to chase the double, whichever is more limiting.

Try this simulator. With the default settings, you often win a little, but occasionally lose a lot. (Let me know if there are any errors. The results run straight down the page if you're using IE, but with a more modern browser, they fill the page horizontally.)

I've been playing a free online game of Roulette. I went from 2000 to 4700 in about 20 minutes with this theory. Then I hit a string of 8 reds in a row and I lost it all.

I had this same brilliant idea while in college, but since I was a CS major I thought that I'd write a simulation and run it overnight on the computers in our lab. While it sounds like a rare probability to get a long string of the same color, in practice you will find that it happens all of the time.

As Mike discovered, the only way you can make this work for you is if you do it for a short period of time. If you play this system for any length of time, you will definitely hit the worst case scenario and lose it all. Of course that doesn't mean that you won't hit the worst case scenario on your first attempt.

Trust me, this is not a new idea, but for some reason the Casinos still manage to stay in business

My way:

1.) Bring a few hundred $$ to the casino.
2.) Go to the roulette wheel and put it all on red or black.
3a.) If you win: pocket half and gamble with your winnings
3b.) If you lose, go home.



and for those who like to play it safer,

1.) Bring a few hundred $$ to the casino.
2.) Go to the roulette wheel and put half on red or black.
3a.) If you win: gamble with your winnings
3b.) If you lose, gmable with what you have left over.

If you are looking for someone that actually did this you found him.

First time I did this I was at the Flamingo. I thought I was a genius when I came uo with my plan of waiting for a roulette wheel to show 4 of the same color in a row. Then I would put $20 on the other color and double my bet until I won. It worked and by the end of my trip, I was up $2000. The only problem was waiting for the tables to show 4 in a row and running, sometimes across the casino, to make the first $20 bet.

On my way home I thought there had to be a glitch somewhere in my thinking because that was way to easy.

Last summer I was at the Mandalay for a bachelor party and got 2 of my friends in on "My System". Again, it worked and the 3 of us where up $1000 in no time. Then the hammer fell. We saw a table that had 5 reds in a row. So we started betting, doubling our bet after each loss. Next thing I knew and 12 reds showing, I had to put down a $2560 bet in order to win my original $20 bet. The big problem was the $2500 limit on these types of bets. My 2 friends were about to kill me when I asked the pit boss if we could bet the $2560. After calling someone upstairs he got the approval and I threw dow $2560. We had a big crowd around the table now and my friends were coaxed into making the bet also. There was absolutely no room on the table to place a bet because everyone wanted to bet on black.

Well you probably guessed the outcome, FUCKIN RED!!! My heart sank because I hate losing money. I watched as red came up 2 more times before black showed up.

I thought this must have been longest streak of one color but the pit boss, who was very cool and compt us a night and told me that this sytem is a loser as is every system people go to Vegas with in the hope of beating the house. He said that if you win using any kind of betting system, take your winnings and never gamble again. That way you can always call yourself a winner.

I’m sure that George Kaplan would write about this more clearly, but a few points to add to the posts—you need to track the money in order to understand the percentages (unless you basically understand the math, in which case the question would not have been asked).

First consider the roulette wheel. In America there are 38 slots, 18 of them red, 18 black and 2 green (0 & 00). Jason is correct if you are playing in Europe where there is only 1 green outcome (0). Assuming that you are playing in the States (as you are in Florida), the probability of losing on any one spin is 20/38 or approximately 54.05%. The probability of losing seven times in a row is:

(20/38)^7 or approximately 1.11%.

Now this seems to be a very rare occurrence—in fact you have only about one chance that it will happen in 100 plays and it should happen only 10 or 11 times on 1,000 plays and of course 111 times in 10,000 plays.

This seems attractive, but consider what will happen to your money while you are playing. For every successful play you win $10. For every unsuccessful play (defined as losing 7 times in a row) you lose $10+$20+$40+$80+$160+$320+$640 or $1,270.

If you play 10,000 times you will win (10,000-111) * $10 or 9,889 * $10 or $98,890—a tidy sum for sure.

However you will lose $1,270 * 111 or $140,970—a net loss of $42,080. Still want to play? In your hypothetical proposition, where you play 100 times you should lose once costing $1,270 and win 99 times leaving you with a net loss of $280.

If you are playing in Europe you won’t lose quite so much as you lose only 94 times in 10,000 plays. So you win 9,906 * $10 or $99,060. You lose 94 * $1,270 or $119,380 for a net loss of $20,320.

Of course in the limited 100 plays you will probably lose the same in either location--$280.

If you want to spend $280 for a night’s entertainment (assuming this is your definition of entertainment), go right ahead. Just don’t do it with the expectation of making money.

Lew sums it up well. Basically, for every person who "wins" using this system, there are more people who will lose, and the house will come out ahead. And over time, you will lose more than you make no matter what. If you use the 7 spin limit, what you have is a high probability of winning a little bit and a low probability of losing a lot. You might "win" enough to think it's a good idea, but it only takes one big loss to make a handful of small wins seem unimportant.

As to the idea of waiting for a table to show a number of one color before betting the other - purely coincidence. As has been pointed out, the probability of an unrigged wheel coming up red is exactly the same whether the past 100 spins were all black, all red, or any other combination.

If you are gambling on games of chance to try to make money, then you're doing it for the wrong reason (IMO). If you're doing it because it's fun (even when you do end up losing), that makes a lot more sense.

FYI,

The betting system is called Martingale (which encompasses a number of "double up after losses" systems) and, of course, will work fine as long as you have an infinite amount of money and the casino will assume an unlimited bet. I've done this at times, when I feel like it, but I will tell you, when you're pushing $1,600 on the betting square to essentially win back $25.00, and you know you have, at best, a 50/50 shot, it doesn't seem too logical.

BTW, there are other versions of this on a roulette wheel covering 2 out of 3, "2to1" payoff bets (such as first third & second third), that are even safer, but pay off slower (you win half your bet). The odds of 7 straight losses are .00031 or .031%. However, it's slow as shit and more boring than watching paint dry.

Steve

In fact: you cannot change your over-all expectation by playing any "system". Of course you can give away money, like tipping the croupier or burning it, but as long as you play regular roulette, your chances will always be the same. You cannot improve them, or even lower them (e.g. by playing "stupidly"). That's 36/37 in Europe and 36/38 in the Greedy-Las-Vegas US of A. Whatever you do.

What those "systems" are doing is: trading bigger chances to a lower gain for smaller chances of higher loss. But the total expectation always remains the same.

Want an honest huge gain? Most efficient method is: enter the casino with every dollar you own. Place all of it on one of the two colours. Three seconds later you either have nothing left or twice your wealth. Leave the casino. Never play again.
Have no money? Lend a million. Buy a gun and 1 bullet. Do the above. In one case you can give back the million to the lender, be almost a millionaire and also sell the gun again.

If there were no limits, a Martingale (or inverse Martingale - even more scarier for the Bank) may seem safe. But the fact is: you either win a little, or else go on losing - which you only view differently because you think you will win it back, eventually. But even if there wasn't a maximum to your bet: there's a limit in time and value. The series cannot take more than your lifetime and the stakes can never be more than all the money in the world. So the unfavourable, devastating (for you) streak does exist. Making it's chances smaller also increases the height of the bet involved. The expectation stays the same.

The Monte Carlo Casino has - since it's opening in the days of Blaisse Pascal - never had an evening they didn't close with a (significant) profit. Popular stories of the contrary notwithstanding.


Cees

Lew,

That explanation makes a heckuva lot of sense.

Cees:


I think I saw that in a movie once . Those crazy Europeans....

"The best way to double your money in a casino is to take it out of your pocket, fold it once and put it back in your pocket."

Joe

I like to play when I am in a casino and generally I come out ahead. I limit my loses and generally stop if I go through the amount.
I watch the wheel and most of the time see some sort of pattern after a dozen or so spins. If I see complete randomness I try another wheel.
I understand that it has no memory and should be completely random but I tend to see some pattern and generally I walk away ahead which means it works for me

Joseph, .


Oh, Anthony, it's rather easy to explain why your chances will not increase or decrease by any betting system.

There are many systems, based on complicated combined bets (e.g. 2 on black, 3 on red), or on clever series (like yours) or on combinations thereof. But it essentially blurs what's going on (the casinos love that).

Whatever you do, you can split those actions into single bets. $10 now - if you lose $20 or else $10: it's just one bet of $10 and another one of $10 or $20. Never mind that you link them logically together by some decision of your own. THEY ARE ALL SEPARATE BETS, some at the same time, some not. To get the feeling: just imagine someone else, not you, made that second bet, and another person a third one in "your" series.

Bets, chips and roulette tables have no knowledge of other bets, chips or tables, nor of themselves at another instance of time.

And each of those bets carry the simple expectation of 36/37 (or 36/38). So the sum of all those bets (comprising your "system") carry exactly the same chance as a total. It's as simple as that.


Cees