Blackjack: Doubling bet: Why doesn't it work?

[Xema]

@duffer

Tip these guys in a manner consistant with what you’re winning, and they’ll “help out” by “suggesting” where and how many chips you should lay.

How can they know more than you about what bets are advisable?

[Rufus_Xavier]

I’m not saying this is any better, but my theory on betting blackjack is that you want to capitalize on winning streaks and minimize losses during losing streaks. Here’s how I do it:

Bet 2 chips (usually for me it’s 2 $5 chips, but any 2 equal denomination chips will do, as long as your bet exceeds the minimum). If you win, increase your bet by 1 chip. If you lose, bet 2 chips again. Repeat ad nauseum. Here’s an illustration of 5 wins in a row followed by 5 losses in a row:

1. win $10
2. win $15
3. win $20
4. win $25
5. win $30
6. lose $35
7. lose $10
8. lose $10
9. lose $10
10. lose $10

net gain $20

If you had bet flat the whole time, your net gain would be 0. If you bet Martingale, you would have lost $260.

[Xema]

@Rufus Xavier

I’m not saying this is any better, but my theory on betting blackjack is that you want to capitalize on winning streaks and minimize losses during losing streaks. Here’s how I do it:

Bet 2 chips (usually for me it’s 2 $5 chips, but any 2 equal denomination chips will do, as long as your bet exceeds the minimum). If you win, increase your bet by 1 chip. If you lose, bet 2 chips again. Repeat ad nauseum.

It works nicely provided the streaks are 4 wins - 4 losses or longer. With streaks of 3 and 3, you break even. With streaks of 2-2 or 1-1, you lose. Since short “streaks” are more common than long ones, the net result (not surprisingly) is that your odds are not improved by this scheme.

[borschevsky]

@Xema

the net result (not surprisingly) is that your odds are not improved by this scheme.

I think your odds may even be slightly worse in this scheme.

The basis of card counting is that certain sets of cards remaining in the shoe are advantageous to the player. If you’ve won a few hands, it’s likely that you’ve been using up the “good” cards, and that the count is now against you, so you should reduce your bet.

I’ve had the idea before that a crude method of counting is to just count the number of hands won by the players at the table. I doubt that it would be very effective though.

[Munch]

@Xema

How can they know more than you about what bets are advisable?

How can they not? How many people do you see at a craps table who don’t take advantage of odds bets simply because they don’t know about them?

[CurtC]

@jasonh300

Having played a LOT of blackjack in my life, it’s NOT unusual to lose 40 hands in a row. It’s a lot more unusual, but not impossible, to WIN 40 hands in a row.

According to your definition of “unusual.” A particular hand in BJ will be independent of the one before it (streaks don’t exist except in retrospect). If the probability of winning is 49.5%, then the probability of getting a 40-hand losing streak is about 1.4 times out of a trillion. The prob of a 40-hand winning streak is about 0.6 times out of a trillion. You must play a LOT of blackjack.

[cormac262]

@borschevsky

I think your odds may even be slightly worse in this scheme.

The basis of card counting is that certain sets of cards remaining in the shoe are advantageous to the player. If you’ve won a few hands, it’s likely that you’ve been using up the “good” cards, and that the count is now against you, so you should reduce your bet.

I’ve had the idea before that a crude method of counting is to just count the number of hands won by the players at the table. I doubt that it would be very effective though.

A guy I met at a casino showed me this betting scheme (although the way he “reset” after a loss was always to the table minimum - not necessarily 2 chips). And having tried it a number of times, I have actually had very good success with it. There are some other aspects that play into it. Splitting and doubling down (and combinations of both) can get pretty “expensive” when the bet has grown to $25 or more. You can obviously win quite a bit, but when you do end up losing on these, they set you back quite a bit.
I also think you need to have a real good handle on when to surrender (provided the casino allows this practice) when the bet gets large.
Playing at $10 minimum tables, I tend to max out with a bet of $30. So though I’m not consistently following the “add a chip” rule at that point, I figure it helps reduce the losses when a double-down or split hand comes along.

As to why it appears to work has more to do with “streakiness” than pure odds. For the most part (and this is only casual observation - no real statistics to back this up), blackjack does have a tendency to be streaky - on the order of 4 to 5 hand streaks (winners or losers). And this strategy takes advantage of this tendency (notice I didn’t say “fact” ;-).
Perhaps the best measure that I’ve noticed is when I compare either how well I’m doing (in terms of gain) or how much longer I last at a table compared to the other gamblers at a table. There is kind of a knee-jerk tendency when you lose to increase your bet, and I see this often. Sometimes it pays off, but since there is this streaky tendency, more times it doesn’t. Using this “add a chip and reset” strategy I have noticed that I generally outlast the “I can’t lose this many hands in a row” types of bettors, and I have had very good success at walking away with more than I started. (of course, you still need to be able to walk away at the “right time” or at all).

So though it may not seem to give you much of an edge, my experience has been that it does seem to work.

[mazinger_z]

@Xema

Outside of special (and presumably unusual) promotions like you describe, how can there be a “professional” blackjack player? Previous threads have described how counting cards is theoretically advantageous but doesn’t work in practice (basically, because it’s easy for casinos to detect and to frustrate). So how can anyone (except a casino owner) make money playing blackjack?

There are still professional blackjack players, they have to move around often, because they do not want to be known. As a semi-pro myself (working my way through law school), they have the act of looking “lucky” down pat. They also don’t bet as much, and are most likely red chip bettors (think 10k-15k bankroll for the week on avg, though aquaintances and I have calculated that a lifetime bankroll of 300k is more common.) Green puts too much at risk and draws too much attention. Black takes too much work, e.g. the casino expects you to be flamboyantly wealthy, not a professional gambler, meaning you like to spend money. If you can pull it off, it pays to bet black because often the high rollers get better odds (less shoes, house hits S17, multiple splits, double anything).

Still the pros are few and far between, and they definitely do not operate like the MIT crew (see the real-life fiction account in Bringing Down the House, also a National Geographic Channel special). That MIT crew was all about profit maximization and they were bringing HUGE cash. The average week for a pro is something like 1.5k-2k hit a week against the casino. The MIT crew was doing 100x’s that.

[David_Simmons]

@Cisco

The only context I’ve ever heard this strategy in is that it doesn’t work. What’s wrong with it?

Say I hop on down to the casino with $500 in my pocket. The first time I cross $100 and win I’m going to leave the table. I bet $1 on the first hand, lose. $2 on the next hand, lose, $4 on the next hand, lose. Say I keep losing until the 8th bet - $128 - and I win. I walk away from the table with an extra dollar and two free beers in my gut.

What’s the catch? Is it that casinos don’t allow this strategy or only that you risk losing that last bet, thus losing everything?

In probablity terms playing any casino game is a random walk with restraining and absorbing barriers. This means that you start someplace on a line, say to the right of zero. Zero is when you are out of money and that is the absorbing barrier. Further to the right of your position on the line is the house limit for any one bet and that is the restraining barrier.

Now each hand if you win you take a step to the left, getting further from zero, and if you lose you take a step to the right. Sooner or later you will have a long series of losses and you will either hit zero, or the house limit. In the first case you’re out of the game and in the second case you are not allowed to bet enough to get your expected return and you system fails because its success is predicated on getting the expected return for each bet.

Ah, you say, you are also likely to get a long series of wins which will put you further from zero. It doesn’t matter because sooner or later a long loss run, which is equally likely, will get you.

In addition to the above, the house tilts the odds just slightly in its favor. And there is no way you can beat the odds by spending more money, i.e. making bigger bets.

[SenorBeef]

@Munch

How can they not? How many people do you see at a craps table who don’t take advantage of odds bets simply because they don’t know about them?

Odds bets aren’t advantageous, they’re neutral.

[SenorBeef]

@Xema

Outside of special (and presumably unusual) promotions like you describe, how can there be a “professional” blackjack player? Previous threads have described how counting cards is theoretically advantageous but doesn’t work in practice (basically, because it’s easy for casinos to detect and to frustrate). So how can anyone (except a casino owner) make money playing blackjack?

It’s pretty difficult to do nowadays, but basically, if you’re just grinding out $10 an hour, the casino isn’t going to care enough to go through the effort of booting you. So, as far as I understand, if you stick to the low limit stuff, and don’t push your edge too far, you can grind out a low wage without anyone really caring.

[Munch]

@SenorBeef

Odds bets aren’t advantageous, they’re neutral.

Yes, but NOT taking odds bets considerably lowers your odds, don’t they?

[Left_Hand_of_Dorkness]

Is this an accurate analogy for the Martingale system?

I pull out seven pennies, and offer you a bet. If I flip them all, and they all come up heads, then you owe me $130. Otherwise, I’ll give you a dollar.

Almost all the time, I’ll end up giving you a dollar. But just often enough you’ll give me $130 that I’ll come out ahead–since the odds of all seven coins coming up heads is 1/128.

Or we can make it eight coins, and you owe me $260 when they all come up heads. Or whatever. As long as your rare payment to me is just slightly greater than the aggregate of what I’ve ended up paying to you, I’ll come out ahead.

Heh. In college I came up with the Martingale system, and for a period of a couple days was convinced I’d come up with a way to revolutionize gambling. Fortunately, I wasn’t foolish enough to put my money where my mouth was.

Daniel

[Munch]

@Left Hand of Dorkness

I pull out seven pennies, and offer you a bet. If I flip them all, and they all come up heads, then you owe me $130. Otherwise, I’ll give you a dollar.

Close, but it has to be such a penny that comes up head’s slightly more often than tails.

[muttrox]

@jasonh300

I also want to add that someone said that the game is never in the player’s favor. This isn’t true. Occasionally, casinos will run a promotion where the game shifts a percent or two into the player’s favor. The most common is to pay 2:1 on blackjacks instead of 1.5:1 (a $5 bet wins $10 on a 2-card 21 instead of $7.50). When the casino does this, playing perfect basic strategy and flat-betting $5 a hand will yield the player an average of $5 extra every 20 hands. If you can play 100 hands an hour (depends on how busy the casino is and how often they shuffle), you can make $25 an hour playing $5 bets…in my experience, they limit it to $5 bets.

Your conclusions doesn’t follow from this example. Yes, the casino has made the odds slightly better than “standard”. It does not follow that the odds are in the players favor. Are the odds actually better than 50% with a 2:1 blackjack payoff? Maybe, but that needs a cite to demonstrate.

Otherwise, it seems like they have tricked you into playing hands as fast you can, and losing your money that much quicker. Sure, you “make” $25 an hour this way, but you’ve also lost a great deal through the normal cost of playing.

[Left_Hand_of_Dorkness]

@Munch

Close, but it has to be such a penny that comes up head’s slightly more often than tails.

Yeah, that’d be better–but to account for house odds, I had your payment when I won be $130, instead of the $128 that would even it out. I figured that’d make for a simpler explanation than going into slightly weighted coins.

Daniel

[Giles]

@Left Hand of Dorkness

I pull out seven pennies, and offer you a bet. If I flip them all, and they all come up heads, then you owe me $130. Otherwise, I’ll give you a dollar.

I only deal in round figures. Can we make it 10 coins? – all heads, and I pay you $1000; otherwise, you pay me $1. That looks fair to me.

[mazinger_z]

@muttrox

Your conclusions doesn’t follow from this example. Yes, the casino has made the odds slightly better than “standard”. It does not follow that the odds are in the players favor. Are the odds actually better than 50% with a 2:1 blackjack payoff? Maybe, but that needs a cite to demonstrate.

Otherwise, it seems like they have tricked you into playing hands as fast you can, and losing your money that much quicker. Sure, you “make” $25 an hour this way, but you’ve also lost a great deal through the normal cost of playing.

Actually, I was doing the math on the back of this envelope, but now I have to run to a meeting; however, I found this site: Blackjack Odds - Advantage Calculator

Paying 2 to 1 is probably the best advantage given to a player, making 6 deck shoes, no resplitting As worthwhile to play.

Btw, 6 to 5 pay off on the natural is probably the most evil thing that casinos have done, even on 1 deck. (Note to all: do not play this game!)

[Left_Hand_of_Dorkness]

@Giles

I only deal in round figures. Can we make it 10 coins? – all heads, and I pay you $1000; otherwise, you pay me $1. That looks fair to me.

Make your payment to me $2,000, and we got a deal :).

Daniel

[Left_Hand_of_Dorkness]

I just realized that I was counting on my fingers to figure out the odds on that one. Heh. At least counting exponents isn’t as embarrassing as counting units.
Daniel