A blackjack betting system changes the size or order of wagers. It does not change the probability of the next hand. Martingale, Fibonacci, D’Alembert, Paroli, Oscar’s Grind, 1-3-2-6 and similar progressions can rearrange the pattern of wins and losses, but they cannot erase a negative house edge.
On this page
- The betting-system verdict in one table
- The math every system must survive
- Why a progression cannot predict the next hand
- Martingale: many small targets, one steep sequence
- Other negative progressions
- Positive progressions
- Flat betting: the honest recreational baseline
- Card counting is not a progression
- Kelly betting only belongs with a positive edge
- How to evaluate any blackjack system
- Common sales claims and the missing fact
- Blackjack betting-system directory
- Frequently asked questions
- More progression systems in the archive
- Worked ledger: same outcomes, different exposure
No progression guarantees a profit or prevents a losing session. If a system’s explanation begins with “you cannot lose if…” and omits bankroll limits, table limits, doubles, splits, pushes and house edge, it is not a complete mathematical argument.
The betting-system verdict in one table
| Approach | What changes | What does not change | Main risk |
|---|---|---|---|
| Flat betting | Nothing; the opening stake remains constant. | Rules, hand probabilities and house edge. | Ordinary blackjack variance and cumulative exposure. |
| Negative progression | Bets rise after losses. | The next hand does not become “due” to win. | Rapid stake growth, table limits and catastrophic sequence loss. |
| Positive progression | Bets rise after wins. | A streak does not create a future edge. | Returning recent winnings to a negative-expectation game. |
| Card-counting spread | Bet size responds to an estimated change in remaining-card composition. | The estimate can be wrong; variance and casino response remain. | Estimation error, inadequate bankroll, poor conditions and detection. |
| Kelly sizing | Stake is tied to a measured positive edge and bankroll. | Kelly cannot manufacture an edge where none exists. | Overbetting when the edge or variance estimate is wrong. |
The math every system must survive
For a fixed game and strategy, a useful long-run estimate is:
expected loss = total initial action × house edge
If a basic-strategy game has an assumed 0.5% house edge, $10,000 of initial action has a theoretical expected loss of $50: $10,000 × 0.005. Actual short-term results can be far above or below that figure. Changing the order of the $10,000 in wagers does not change the percentage edge; it changes dollar volatility, the chance of reaching a limit and the shape of the results.
Blackjack makes progression claims especially messy because not every round is a simple win or loss. Hands can push, blackjacks can pay 3:2 or 6:5, doubles add a wager, and splits can create multiple hands. A progression that treats every outcome as even-money coin tosses is not modeling blackjack faithfully.
Why a progression cannot predict the next hand
Past wins and losses do not, by themselves, tell you that the next hand is due to reverse or continue. In a hand-dealt game, removed cards can change the composition of the remaining shoe, but a win/loss streak is not a card-counting system. A valid count tracks specific cards or tags, estimates decks remaining and connects the resulting true count to a measured edge. “I lost five hands” contains nowhere near that information.
This distinction separates a betting progression from advantage play:
- A progression reacts to bankroll history.
- A count-based spread reacts to information about the remaining cards.
- Basic strategy reacts to the player’s hand, dealer upcard and stated rules.
Martingale: many small targets, one steep sequence
Martingale normally doubles the next wager after a loss and returns to the base wager after a win. With a $10 base, one common sequence is:
| Consecutive loss | Next wager | Total already exposed if that wager loses |
|---|---|---|
| 1 | $20 | $30 |
| 2 | $40 | $70 |
| 3 | $80 | $150 |
| 4 | $160 | $310 |
| 5 | $320 | $630 |
| 6 | $640 | $1,270 |
The target is still one $10-unit profit, but the required wager grows exponentially. A finite bankroll, a table maximum or the need to double and split can stop the sequence before recovery. Pushes complicate the progression, and a blackjack payout does not repair the fundamental constraint. A larger bankroll only moves the failure point; it does not remove it.
Read the full Martingale blackjack analysis.
Other negative progressions
Fibonacci
Fibonacci increases stakes through a sequence such as 1, 1, 2, 3, 5, 8 and 13 units after losses, then moves back after wins. Its growth can feel gentler than Martingale, but the wager still rises because of past losses rather than a new player edge. See the Fibonacci system guide.
D’Alembert
D’Alembert usually adds one unit after a loss and subtracts one after a win. The slower progression reduces exponential shock, but it still assumes that balancing wins and losses will produce recovery. Blackjack payouts and outcome frequencies do not promise that balance on the player’s schedule. See D’Alembert in blackjack.
Oscar’s Grind
Oscar’s Grind generally raises the wager after a win and tries to end a cycle one unit ahead. The cycle target does not change expected value. A long loss-heavy sequence can require large stakes or end at a bankroll or table limit. See the Oscar’s Grind analysis.
Positive progressions
Positive progressions increase wagers after wins. They avoid chasing losses with the largest bet, but they still return money to the table when a past streak gives no reliable prediction about the next hand.
Paroli
Paroli commonly doubles after wins for a limited number of steps. The cap can make the risk easier to understand than an unlimited loss progression, but it cannot turn a negative-expectation game positive. See the Paroli guide.
1-3-2-6
The 1-3-2-6 system follows a four-step unit sequence after consecutive wins. Completing the sequence produces a favorable cycle result; failing early loses part of the preceding wins. The probability of completing a cycle and the game edge still control the long-run outcome. See 1-3-2-6 explained.
Parlaying “the house’s money”
Once a chip is won, it is the player’s money. Calling it house money can make an otherwise unacceptable wager feel painless. A positive progression can be used as a predefined entertainment pattern, but not as proof that recent winnings are less real or less exposed.
Flat betting: the honest recreational baseline
Flat betting does not beat blackjack. Its advantage is operational clarity: the initial wager is predictable, tracking is easier and a loss does not automatically trigger a larger risk. At the same average stake and house edge, expected loss still follows total action.
For recreational play, the most defensible approach is usually:
- Choose an affordable total-loss and time limit before play.
- Reject 6:5 blackjack and compare the complete table rules.
- Use accurate basic strategy.
- Flat-bet at the lowest practical stake and leave room for correct doubles and splits.
- Skip high-edge side bets.
- Track all deposits, cash buy-ins, withdrawals and cashouts.
Those steps reduce avoidable cost and control exposure. They do not promise a win.
Card counting is not a progression
A trained counter may vary bets when the estimated remaining-card composition creates a player advantage. That is an information-based betting decision, not a claim that losses must reverse. It also requires accurate counting, true-count conversion where applicable, favorable rules, sufficient penetration, an appropriate bet spread and a bankroll sized for variance.
Even a real edge does not guarantee a profitable session. Mistakes, poor conditions, heat, backoffs and estimation uncertainty can erase it. Begin with the card-counting guide, then study bet spreads and risk of ruin.
Kelly betting only belongs with a positive edge
The Kelly criterion sizes a wager from an estimated positive edge and payoff. It is a bankroll-growth model, not a method for making a negative game profitable. If the estimated edge is zero or negative, the Kelly wager is zero.
Full Kelly can be highly volatile, and an overestimated edge leads to overbetting. Advantage players often consider fractional Kelly because real casino inputs are uncertain. The calculation must reflect the actual blackjack outcome distribution rather than an oversimplified even-money coin. See the Kelly criterion guide.
How to evaluate any blackjack system
- Identify the claimed source of edge. If it is only a win/loss sequence, there is no new information about the next hand.
- Write the complete wager path. Include the largest wager and cumulative amount at risk.
- Add real blackjack outcomes. Model pushes, blackjacks, doubles, splits and surrender.
- Apply table limits. Ask what happens when the required next wager exceeds the maximum.
- Apply the bankroll limit. “Enough money” is not an infinite bankroll.
- Calculate total action and expected value. Do not judge a system only by the percentage of cycles that end ahead.
- Measure drawdown and risk of ruin. Average result alone does not describe a tolerable strategy.
- Reject cherry-picked sessions. A winning screenshot is not a reproducible test.
- Look for an unfalsifiable reset rule. A seller who discards failed sessions but counts successful cycles is manipulating the record.
Common sales claims and the missing fact
| Claim | Missing fact |
|---|---|
| “You only need one win to recover.” | The wager required to reach that win may exceed the bankroll or table maximum. |
| “It wins most sessions.” | Many small wins can be outweighed by a less frequent large loss. |
| “You are playing with casino money.” | Winnings in your stack are your money and can be lost. |
| “The next hand must turn.” | A streak alone does not create a reversal probability. |
| “Card counters spread their bets, so progressions work.” | Counters respond to estimated card composition, not to a loss sequence. |
| “The system worked for me.” | A finite result cannot establish positive long-run expectation. |
Blackjack betting-system directory
Frequently asked questions
What is the best blackjack betting system?
No progression beats a negative house edge. For recreational play, an affordable loss limit, good rules, correct basic strategy and a small flat bet are more defensible than a loss-recovery sequence.
Can a betting system make consistent profit?
Not without a real, measurable source of positive expected value. A progression can produce a high percentage of small winning cycles while retaining a lower-frequency large loss. “Consistent” does not mean profitable after all outcomes are counted.
Does Martingale work if I stop after a few steps?
A step limit caps the worst progression loss, which is useful risk control. It also means the sequence can end before recovery. The cap improves transparency, not expected value.
Is flat betting profitable?
Flat betting in a negative-expectation game is still negative in the long run. It simply avoids progression-driven stake spikes and makes exposure easier to track.
Can card counting make a betting spread profitable?
Under suitable hand-dealt conditions, accurate counting and disciplined bet variation can create a small player edge. It is difficult, volatile and not guaranteed, and the casino can restrict or end play.
Do stop-loss and win goals change the odds?
No. They can limit time, money exposed and chasing behavior, but they do not change the probability or expected value of the next hand.
More progression systems in the archive
These older systems reorganize bet sizes; none changes the probability of the next hand or removes the house edge. The pages are retained so readers can inspect the rules and risks directly.
- 1324 Doubled
- Baby Progressive
- Dubois
- Power Press
- Pump and Dump
- Stern
Worked ledger: same outcomes, different exposure
Take the five-result sequence L-L-W-L-W. Flat betting one unit produces −1−1+1−1+1 = −1 unit after five units of action. A double-after-loss progression stakes 1, 2, 4, 1 and 2 units, producing −1−2+4−1+2 = +2 units after ten units of action. That selected sequence favors the progression because its wins land on larger wagers.
Reverse the order to W-L-L-W-L. Flat betting still finishes at −1 unit. The progression stakes 1, 1, 2, 4 and 1 units, producing +1−1−2+4−1 = +1 unit after nine units of action. Other orders create large losses. The comparison does not prove that either sequence has a positive expectation; it shows why session examples must report total action and maximum stake. Judge a system over the full distribution under the same game rules, bankroll, table limits and stopping policy—not by one attractive path.
