It's Good To Be The House

Last night while I played in a FR, just to pass the time so I wouldn't donk out, I read this golden thread entitled

is it just me, or is roulette easy to make money on?


Well, the thread is about a member (who is now banned, by the way,) who makes a post revealing his super system for winning at roulette. You can read the thread for more hilarious details, but it boils down to the poster not understanding odds. So I read the entire thread, invested some time in it while I played, got to the end and there was no climactic ending. I felt like I had wasted my time. Throughout the day today, I thought about the thread here and there, and then it hit me. One of the main points against the original poster was that the 0 and 00 made his system, and actually all of roulette, a losing proposition. He was betting the outside 12 numbered groups, which pay 3-1. So if you bet on one group, you have 36 total numbers and 12 are outs, 36-12 equals 3-1 right? Well yes, but the total numbers that can hit isn't 36 but 38 making your odds of hitting 38-12 which is 3.17-1 which is what? Ah, yes -Ev. This was a nice light bulb for me. I've understood pot odds, and getting your money in with the best of it, but this thread really put it home for me. I've always dreamed, "man, it would be nice to own a casino", sure it would, offering bets where the payouts are shorter than that of the odds of winning. It puts odds in a nice mental framework, I basically want to offer and accept bets where I have the edge...I want to be the house.

If you want to win at roulette and have a lot of time on your hands.

Bet 1$ on red, if black comes up
bet 2$ on red, if black comes up
bet 4$ on red, if black comes up
bet 8$ on red, if black comes up
bet 16$ on red, if black comes up
bet 32$ on red.

Continue that process then as soon as a red appear, you will have made a profit. Using that system, if 26 balacks appear in a row for example, but the next is a red, you have made an overall profit across the last 27 spins.

You just need time and a lot of money in case of streaks of black spins

And also a casino that doesn't cap roulette betting

Which is why poker is the only thing I'll play in the casino. Apparently there's one even money option on the craps table, but who wants to go to all that effort just to break even?

seems like a winning strategy
but if u lose like 20 spins u have to bet 1048576$ on the 21st spin to make a 1$ profit.... going to make huge money this way....woohhoooo, thanks for the idea

They actually refer to this in the thread...I've tried it on Black Jack went on a six or seven losing streak lost my role.



"Chuck its called the Martingale System, like all systems to beat casino games its flawed and you'll end up losing eventually.

Heres a good site to check out about roulette and many many other casino games

Wizard of Odds: The last word on gambling strategy

(incidently, that site is the one that lead me to this site)"

Right, while you make a profit the heavy majority of the time, the one time you do lose, you lose a bazillion dollars. I tried a simulation on my computer over 1 million field bets in craps (basically a 55-45 bet). If there were no betting cap, this strategy would work. The problem is once you reach the cap you have lost so much money trying to win a couple bucks that it's not worth it
Agreed on poker, at least when I turn 21. In craps, there is an even money option only after you've put a money tilted in the house's favor, meaning that you can never JUST play even money bets.

Exactly. I think you could probably make this work inside the cap if you were just betting on red and black in roulette, and starting with the absolute minimum bet. But there's no way you'd win enough money to be bothered doing it.

Oh - and as to Sub's post, I think there's a further complication in applying this theory at the blackjack table: human 'error'.

Applied to roulette or craps you're playing pure odds. Applied to blackjack, you have to factor in decisions on hitting or standng. You're exploiting the tiniest of edges to start with in this theory: introduce judgement calls into the equation, and it's probably a losing proposition. There are other systems for blackjack that will probably work better for you.

You could have a large probability of winning a small amount, but this proposition, no matter how small, would still be -ev. Even if you have a 1% to win $1 and 99% to lose $150, you'll most likely make $1 but it is not a good bet.

You've lost me on that one - the odds of red hitting are... what about 48%? It all depends on the betting cap, but if it pays 2:1, start by betting $1 and then doube each time until you win, you'd have to get uber-unlucky not to make the $1 before you hit a cap.

The catch is that once you've made your $1 profit, you've gotta go back to the beginning and bet just that $1 again (which is now your winnings... so you're starting the next round for "free"). If you don't do that, then it's a losing proposition.

Of course, most people subscribe to a different theory: wait until the ball is already in motion, then put your bets on the numbers that nobody's chosen so far. Preferably, place your bets right as the croupier is saying "No more bets". Or immediately afterwards: then pout and complain when your bet isn't accepted. But that way, you'll be betting on the numbers that the ball is more likely to land on, since the house rigs the wheel to land on the numbers that nobody is betting on.

A friend of mine is a roulette croupier, and people seriously believe that. Like, lots of people.

I'm sorry, but the point of a house edge is just that, a house edge. If you continually make 48%-to-win bets, you can't possibly do it profitably (unless you can do it infinitely. No matter how it's set up, you will most likely win $1 before you hit the cap, but your loss will be worse than your wins in the long term. Basically I'm saying that if the min bet is $1, you stand to win $1 every time. So say the cap is $1,024 (trust me on this, makes it easier). The amount bet on the nth roll is 2^(n-1) in this example ($1, $2, $4, $8, ect.). It will take 11 straight losses to hit the cap. So the odds of losing 11 in a row is 0.52^11 = 1 in 1330. So 1329 times in 1330 we win a buck, that's $1329. But that 1 in 1330, we lose 1024+512+256+128+64+32+16+8+4+2+1 = $2047. We lost $718 over the 1330 rolls and the ev for this game is -$0.54. That's right, although we win $1 the heavy majority of the time, we are actually losing 54 cents every time we try this scheme. Still sound like a good idea? Try the math with a $1,000,000 limit. Trust me, it works the same way. There may be like a 1 in a million chance of busting, but when the bust costs you a trillion and the win nets you $1, it's a bad bet.

This scheme is like a reverse lottery, although the odds are still on the side of the house, even though the roles are reversed (we're wagering a ton of money to win only a bit).

Well the theory relies on having an infinite bankroll, no betting cap, and running infinite trials. By that logic it's profitable, but if any one of those conditions isn't met then it becomes a losing system, and of course no one's ever going to meet all three of those.

Fun for theoretical discussion at least

Hence the "uber-unlucky" - but you're right on the math, taking a 48% bet at 2:1 indefinitely will indeed be a losing proposition.

Of course, in this system you don't take the bet indefinitely: you stop once you win a bet and bank your $1 profit (and then start over with that $1, should you so choose).

But should you go on a massive cooler and the ball lands on black 11 times in a row... yes, you stand to lose a lot of money. And there's no way a table that allows a $1 bet would allow a $1024 bet anyway, so chances are you hit the ceiling often enough to make this an even greater losing proposition. With a $50 ceiling, for example, you hit it after six spins.

I've in no way advocated this system, BTW - I'm just acknowledging that in theory, it's about as reliable and probably a way as you're going to find to grind a very small profit out of a casino table game. It's definitely got enough flaws for me not to bother using it

This is not true though. You grind a poker game long-term for small profit and you make at least a small profit. You grind roulette a long time you end up losing an average of worse than half the bet in the example I used (which you admitted is even optomistic).

Of course: "as reliable as you're going to find" doesn't actually equal "guaranteed profitable!"