Among emails I receive, a fairly common question is something like this:

Can you tell me the odds of losing six hands in a row at blackjack?

Sometimes it is 5 hands, sometimes 8, sometimes more. No matter, I cringe whenever I get this question.

To me, it’s like an airship designer asking a chemist: “Does hydrogen weigh only half as much as helium?” We all know how THAT turned out.

In either case, there is nothing wrong with the matter-of-fact question being asked, but it is apparent to the recipient that the question is likely a clue to dangerous thinking, whether it be the Hindenburg or the Martingale progression.

I’ll get back to that in a moment. But first, let me respond to the question, very carefully…

## You probably mean “probability”

The first thing I want to do is change the wording of the question from “*What are the odds*” to “*What is the probability*“. Most people asking this question want an answer that is a single number or percentage, representing how likely the event is to occur. “Odds” means something slightly different, although the two are often used interchangeably. If you want to understand the difference, Google it.

## Your question is incomplete

There is a crucial piece of information missing from your question. How do you plan to play?

Let’s say you sit down next to me at a blackjack table, and I look over and tell you: “This is unbelievable! I’m on the most horrible losing streak. My luck is so bad, I bet you $100 that I lose the next six hands in a row.” If you were foolish enough to bet me, I would quickly demonstrate how the probability of losing six in a row can be 100%!

But for the sake of your question, let’s assume the player will use accurate basic strategy on the game. I cannot resist the temptation of pointing out that virtually no player who asks this question will know a completely accurate basic strategy! But I’ll play along anyway.

## Still not clear yet

There is another picky detail to consider. I assume you are not asking for the probability of losing exactly six hands in a row, which implies that the seventh hand cannot be a loss. Instead, you want to know the probability of losing six **or more** hands in a row. We can easily sidestep that complication, if we simply say that you want to know the probability of losing the **next** six hands of blackjack.

## Finally, an answer?

For a typical shoe game, a basic strategy player will win 43.3% of hands, lose 48.0%, and push the other 8.7%.

If I answer the revised and improved question…

*What is the probability of losing the next six hands at blackjack, using basic strategy?*

Since each hand has a 48.0% chance of being a loss, the answer is 0.48 ^ 6 = 0.012 = 1.2%.

But wait a minute…

## That still wasn’t the right question!

That probably isn’t what you meant either. What if you lose the first hand, push the second hand, and lose the next five hands in a row? You would probably want to call that losing six in a row. We should just ignore pushes altogether, and only count hands that end in a win or a loss. Of non-push hands, players win 47.4% and lose 52.6%.

Finally an answer that may satisfy you:

*What is the probability of losing the next six hands at blackjack, using basic strategy, ignoring pushes?*

The answer is 0.526 ^ 6 = 0.021 = **2.1%**.

That is roughly 1 in 47 attempts.

## What about other length losing streaks?

Just because this is a convenient place to do so, I’ll publish the numbers for other lengths of losing streaks.

Probability of losing n hands in a row, ignoring pushes. | ||
---|---|---|

n | Probability | One in |

1 loss | 52.6% | 1.9 |

2 losses | 27.7% | 3.6 |

3 losses | 14.6% | 6.9 |

4 losses | 7.7% | 13 |

5 losses | 4.0% | 25 |

6 losses | 2.1% | 47 |

7 losses | 1.1% | 90 |

8 losses | 0.59% | 170 |

9 losses | 0.31% | 323 |

10 losses | 0.16% | 614 |

11 losses | 0.08% | 1168 |

12 losses | 0.04% | 2219 |

13 losses | 0.02% | 4217 |

## Why I hate answering this question

When players ask questions about how likely it is to lose hands in a streak, it usually means that they are considering a negative progression betting system that would fail if they encounter a certain number of losing hands in a row. Most commonly, they have rediscovered the infamous Martingale system, where you double up after any losing hands to catch up.

The Martingale is the simplest of the negative progression betting systems. The idea is simple. If you lose one hand, just double your bet on the next hand. Eventually, you will win a hand, and make back all of your losses plus a profit of the initial bet. Indeed, this seems to work very reliably, until it doesn’t. The problem is that most people underestimate how likely those “unlikely” streaks of losses really are. Check the table above. The probabilities get small, but not nearly small enough.

## Why you should NOT bet the Martingale

If you think the Martingale is a good idea, you are badly mistaken. Yes, it gives you a high percentage chance of winning a small amount, but it does not take long to run out of luck. For a $10 bettor who wants to be able to handle up to 7 losses in a row, he needs to have a bankroll of $2550, and be willing to bet $1280 on the eighth hand. All this, for a $10 win each time.

And he will run into that fatal eighth loss once every 170 trials on average. You can think of that as making $1700 in profit before you eventually lose $2550. (Yes, I am oversimplifying because doubles and splits make the math very messy. The real numbers still make this a very bad idea.)

On top of that, blackjack is a terrible choice for the Martingale system. With the necessary doubles and splits, sometimes both on the same hand, you will actually need a far larger bankroll to play correct basic strategy. If you choose to ignore basic strategy doubles and splits, you can get by with “just” the $2550, but at a substantially increased loss rate on all your bets.

## It’s not only the Martingale

**Any** progressive betting system is doomed to fail. Any series of negative expectation bets is guaranteed to create a negative expectation overall. If you are just playing for fun, you will lose less money by flat-betting. If you are playing for profit, you need to find a true way to get an edge, such as counting cards.

Hopefully you are reading this before you learn the hard way. After all, we don’t fly airships full of hydrogen any more either.

You have a 1/20 prob. of being dealt a blackjack w only a 1/25 chance (assuming 8-deck shoe) of having your blackjack pushed. Suppose you are playing a 13 loss Martingale, then that means you will have 200 strings that end with you getting a blackjack and making $10 plus your bet on the string. Simply counting $10 profit on every string ending in a loss and multiplying by the number of times you can expect to win and then subtracting by the total lost win you do run through the string is incorrect. Perhaps, you can figure the math using martingale relative to expected losses while figuring in the mathematics of a 3/2 payoff on the probability of each length string ended by a blackjack?

You’re right. I ignored blackjacks in this simplified comparison, just like I ignored doubles and splits. The math gets quite a bit messier, although I suppose it would fairly easy to account for only the blackjacks. The conclusion is however unavoidable. In blackjack, your expected loss is the total of your bets times the house edge, unless you vary your bets based on the deck composition instead of some scheme based on a string of wins or losses. No progression can change that fact. Feel free to work it out in laborious detail if that interests you. It doesn’t interest me.

You also didn’t factor in the 3/2 payoff on a blackjack that can be the streak ender. If you bet, say $640 after your 6th loss in a row and hit a blackjack, you make $330, not $10. Also, since double-downs and most splits are in your favor, while you certainly need a much bigger bankroll for the possibility that you’ll get deal an 11 when the dealer has a 6, those hands also put the odds in your favor when you have the most money on the table.

Question and Observation regarding raising your bet when the count is a negative 7 to 13 true negative of 4′ Has anyone run simulations on this.

So far , 100 hands only, it works almost as well as a true positive count of 4′. If Simulation agree, Why?

And when bets are only the true count 1 or greater than 1?

Positive true counts don’t affect the win percentage as much as you probably think and hope. See this: https://www.blackjackincolor.com/truecount5.htm

Why does your basic strategy recommend always standing on a pair of sevens, where as you hit a hard 14 when the dealer shows 6 or more ?

You don’t mention what dealer upcard you are looking at, nor what the rules of the game are. I’m going to assume that maybe you are looking at a single deck game, and the unusual advice of standing with (7,7) vs a dealer Ten. (Or Surrender if available!). Well, single deck is a lot different. The effect of removing each card is substantial in single deck. In this case, what card would you prefer to draw here? Another 7. But there are only two 7s left in the deck. That swings this decision to make standing better than hitting.

My usual caveats: This ONLY happens in single deck. Refer to the Strategy Engine for your exact game’s optimal strategy.

I’m embarrassed to tell you that I was looking at T-T, not 7-7. These 74 year-old eyes play tricks on me.

Sorry

I know i m off topic but i don t know where to post.

I had a new idea to count down cards.

Is there anyone who can give me an help please?

my mail is federico.betti90 @ gmail.com

How do you know when to increase betting when I play for five or ten bucks I win when I go up I lose? any help on staggering my betting to improve winning percentages.

There is no magic betting system of staggering your bets that can or will work. If you want to beat the game, learn to count cards. Then you can raise your bet when you know you have an advantage. If that doesn’t appeal to you, the cheapest way to play the game is to flat-bet the minimum.

Why dont you advise splitting 8s on ten and ace and ace on ace like other srategys thanks

It sounds like you have generated a strategy for a “no-peek” game using the strategy engine or trainer. In that case the plays you mention are correct.

But if you are playing a standard US-style game, make sure you choose “peek” instead. The strategy for those hands will then change.

Edited to add: It looks like you are probably in the UK, where no-peek is the norm. As my charts indicate, don’t split against a dealer Ace, because you could lose both bets to a dealer blackjack. (Against a ten, only split aces.) That can’t happen in a peek game.

Great helpful advice re martingale. The pushes and table maximums make any negative progression a terrible strategy.

It’s clear you never heard of “Jim’s rhythm system”.

I can’t reveal the details, but it’s a modified anti martingale system that limits your losses and maximizes your gains, best played on a 6 deck shoe, and playing two hands or one hand, depending on the circumstance. If other players at the table don’t like you ‘changing up the cards’ by being in and out with one or two hands, you may want to find an empty table. You will never draw attention from the pit, and you never count cards. This system will lose more than it wins,but the wins can be huge.