A betting pattern changes how much you risk. It does not turn a house-edge game into a winning game. Start by separating how you play each hand from how much you put on it.
The examples below describe staking rules, not forecasts or recommended bet sizes.
| Approach | How the stake changes | What to watch | Creates an edge? |
|---|---|---|---|
| Flat betting | Keep the same starting stake each round. | Splits and doubles can still increase the amount at risk within a round. | No. It keeps the starting stake predictable. |
| Martingale | Double after a loss; return to the base stake after a win. | A losing sequence quickly runs into bankroll or table limits. | No. Larger bets do not improve the next hand’s odds. |
| Win progression | Increase after wins, then reset under a chosen rule. A capped doubling sequence is one example. | A later loss can give back earlier gains. Versions use different reset rules. | No. A winning streak alone does not establish an advantage. |
| Advantage-based sizing | Size stakes using an independently estimated edge and the outcome distribution. | A wrong edge estimate or excessive stake can be costly. See the Kelly discussion below. | No. An advantage must already exist; sizing does not create it. |
For the underlying distinction, see Wizard of Odds on betting systems.
Change the virtual-chip amounts to see the stakes a Martingale would require during consecutive losses.
After seven consecutive losses, the required 640-chip stake exceeds the 500-chip table maximum and the 365 chips remaining.
| Loss | Stake | Total lost | Chips left |
|---|---|---|---|
| 1 | 5 | 5 | 995 |
| 2 | 10 | 15 | 985 |
| 3 | 20 | 35 | 965 |
| 4 | 40 | 75 | 925 |
| 5 | 80 | 155 | 845 |
| 6 | 160 | 315 | 685 |
| 7 | 320 | 635 | 365 |
Arithmetic demonstration, not a blackjack simulator. Every row is an assumed loss, not a predicted result. Pushes, blackjack payouts, surrender, splits and doubles are excluded. No probability of a losing sequence is calculated. A win is not “due” after these losses.
JavaScript is off. The worked 5-chip / 1,000-chip / 500-chip example is shown above; changeable inputs require JavaScript.
Are these systems for real? Well, that depends. It might be best to divide betting strategies into two categories, just to separate the effective methods from those that are ineffective (and sometimes sold as scams). Any betting strategy that claims to be able to structure bets that all have a house edge and somehow make you a winner is bound to disappoint you, because it simply isn’t possible to take multiple bets with negative expectations and come out with an overall positive expectation. We’ll call these betting systems. On the other hand, betting strategies that offer you ways to maximize your advantage or limit your risk of going broke when you already have the edge can be an important part of a professional blackjack player’s overall strategy.
Over the years, many people have tried to come up with betting systems that could beat various casino games. They appeal to a lot of players, especially those who don’t understand the math behind gambling. Often, they make some sort of intuitive sense, which leads people to come up with reasons to believe that their system must work. However, as we said earlier, no betting system can turn a series of bets with a house edge into a system that has an edge for the player.
Let’s take a look at one common betting system and see how it works – and why it doesn’t work.
The Martingale System is probably the best known casino gambling system in the world, and also the most popular. It is incredibly simple to use, and seems to offer a “can’t lose” way of playing games that offer bets that are around 50/50 propositions.
In the Martingale System, the player begins by making a small bet – usually the table minimum. If the player wins, they make that same small bet again. If they lose, they instead double their bet. If the player loses again, they double their bet again. The player continues doubling their bet until they win, at which point they revert to their original small bet size.
In an even-money example, a completed cycle ending in a win recovers the previous losses and adds one base unit—provided every required bet can actually be placed. For instance, a player might win their original $5 bet, thus profiting $5. Or they might have the following sequence:
Meaning they would lose $35 but then win $40, thus coming out $5 ahead yet again.
So, why doesn’t this work? Eventually, you will hit a string of losses that will be catastrophic, wiping out your bankroll and any small profits you’ve made so far. Many players believe a long string of losses like that is unlikely, and they’re right; however, it will happen often enough that the casino retains the house edge they had at the beginning of the game. Table limits make this occur even faster, meaning you’ll quickly hit a point where you simply can’t double your bet anymore.
For instance, take a table with a minimum bet of $5 and a maximum bet of $500. The player can double six times:
After that, the player cannot double to $640. For the moment, let’s do an example with single-zero roulette, since the math is easier. If the player makes a bet on one of the even money bets – say, betting on black – they have 18 ways to win, and 19 ways to lose. This means the player will lose 51.3% of the time. This also means that at any given time, there’s about a 0.94% chance that they’ll run into a 7-spin losing streak, putting them in the position of not being able to double. That’s only one time in 106 (we’re rounding a bit, but that’s roughly accurate), but that’s still not good enough to make the Martingale work! The player will make $525 in the 105 times they win, but they’ll lose $635 the one time they hit a 7-spin losing streak.
The math for blackjack is a bit more complicated, since not all bets pay even money, you’ll sometimes have to spend extra money to double down or split, and the player wins far less than half of the hands (especially when taking pushes into account). However, the principle remains the same – the player cannot overcome a series of bets that have a house edge simply by creatively structuring their bets.
English narration and on-screen captions. Illustrative examples use practice chips with no cash value.
Can doubling after every loss always get your money back? Watch where this simplified betting ladder stops working.
Start with a thousand practice chips and a table limit of five hundred. Bet five, then double after each loss: ten, twenty, forty, eighty, one hundred sixty, and three hundred twenty. Seven losing bets cost six hundred thirty-five chips. You have three hundred sixty-five left. The next bet would be six hundred forty. That exceeds both your remaining bankroll and the table limit. A larger starting bankroll only moves the stopping point; it does not remove it. This example assumes each lost bet loses one stake. Real blackjack also includes pushes, blackjacks, splits, and doubles, so it is not a complete game simulation. Changing stake sizes alone does not turn a negative-expectation game into a profitable one.
Use the calculator to explore limits, not to plan a recovery. Set a loss limit you can afford before playing, and do not chase losses.
On the other hand, can betting strategies help you when you already have an advantage? The answer is emphatically yes! In fact, betting strategy is critical for anyone who wants to be a profitable blackjack player. For instance, card counting relies heavily on betting strategy, as the player must bet more when he has the advantage, and less (or not at all) when things swing in the dealer’s favor.
But betting strategy can be much more complex than that. While it’s obvious that players should bet more when they have an edge, it’s harder to figure out how much they should bet on a typical hand. Figuring that out requires the player to take into account the size of their bankroll, as well as the size of their edge over the house.
One method of figuring out the optimal bet size is to use the Kelly Criterion. The Kelly Criterion seeks to maximize expected logarithmic bankroll growth when the probabilities and payouts are known. It does not remove volatility or guarantee a profit. For a simplified bet with only a win or a full-stake loss, the formula is:
f = the portion of the player’s bankroll they should wager on each bet.
b = the odds the player is receiving on each wager (b to 1).
p = the probability of winning the wager.
q = the probability of losing the wager.
For instance, imagine a game where the player receives even money odds (1-1), with an assumed, known 55% chance of winning each bet. This would make the formula look like this:
f = (1*.55 – .45)/1
f = .1
In this case, the Kelly Criterion suggests betting .1, or 10% of the player’s bankroll, on each bet in order to maximize the growth-rate of the player’s bankroll. Of course, players rarely have such a huge edge, and blackjack is a bit more complex since the player sometimes wins differing amounts, and sometimes is forced to wager additional money on a hand. We’ll go into more detail on how to use the Kelly Criterion when playing blackjack in our article on Bankroll Management.
Be wary of anyone selling a betting system: see our warning about sites offering blackjack betting systems.
Test any system with practice chips in our free blackjack game first: you’ll see it changes the swings, not the house edge. If you do play for money, set limits with help from our responsible gaming page, and see our real-money blackjack guide for what to check.
A losing sequence is not evidence that the next hand will win. In a dealt shoe, remaining card composition can affect the odds; the win/loss pattern alone does not tell you the count.
No. Flat betting fixes the starting stake. Basic strategy concerns decisions such as hit, stand, double and split. You can use basic strategy while keeping the same starting stake.
A push returns the original wager. A simple win/loss progression would normally leave its next stake unchanged. Real blackjack also has split hands, doubles and non-even-money results, so define how those are handled before treating any example as a complete system.
A stopping rule limits the amount or time you choose to risk. It does not change the underlying rules or make a negative-expectation bet profitable.
Use the free blackjack table to learn the playing decisions. For counting accuracy, use the card-counting practice hub. Both use virtual chips with no cash value.
Bovada CasinoCasino, Poker & Sportsbook · Fast Bitcoin withdrawals
Source: LCB offer data. Current casino terms apply.
Ignition CasinoSupports Bitcoin transactions · PC, Mac and mobile play
Source: LCB offer data. Current casino terms apply.
Miami Club CasinoWeekly blackjack tournaments · Great for U.S. based players
Source: LCB offer data. Current casino terms apply.
Lucky Tiger CasinoViG Live Blackjack · BTC Bonuses
Source: LCB offer data. Current casino terms apply.
