Baccarat Strategies: What Works, What Doesn't and the Maths Behind It

No baccarat strategy can beat the house edge over time. Every hand is settled by fixed drawing rules, so you make only two real decisions: which bet to place and how much to stake. The best way to play is to bet on the Banker with flat stakes. That bet has the lowest house edge in standard eight-deck baccarat, at about 1.06%.

Betting systems such as Martingale, Paroli and 1-3-2-6 change how your results swing from session to session. They don’t change how much you can expect to lose on average. Reading patterns on the scoreboard doesn’t help either, because each hand is independent.

This guide looks at the most common baccarat strategies, with worked examples and verified odds. It also covers the Mikki Mase claims and the mistakes that cost players the most money. If you need the rules first, read our guide on how to play baccarat.

In punto banco, the version used in almost every online and live casino, the player and dealer make no drawing decisions. The third-card rules are fixed. Your only choices are Banker, Player or Tie, plus any side bets.

The house edge on each bet in a standard eight-deck game is shown below.

If you’re asking how to win at baccarat, these numbers are the honest answer. You can win in the short term, because results vary from session to session. Over many hands, though, your average loss moves towards the house edge on the bets you choose. A strategy can only help you avoid the worse bets and control how much you stake.

The Banker wins slightly more often than the Player because of how the drawing rules work. The casino charges a 5% commission on Banker wins to make up for that. Even with the commission, Banker remains the cheapest bet on the table.

Worked example: what commission costs

• Stake £20 on Banker and win: you are paid £20 minus 5% (£1), so your profit is £19.
• Stake £20 on Player and win: your profit is £20.

Player pays more when it wins, but it wins less often. Over time, Banker still costs you less.

Expected cost per 100 bets of £10

• Banker: £1,000 staked × 1.06% = about £10.60
• Player: £1,000 staked × 1.24% = about £12.40
• Tie at 8:1: £1,000 staked × 14.36% = about £143.60

These are long-run averages, not the result of any single session. They show why the Tie bet is so costly. Flat betting on Banker is not exciting, but it is the strategy that costs the least. Stake sizing is the only other part you control.

Most baccarat betting strategies are progressions borrowed from roulette and other even-money games. Each one changes your stake after a win or a loss. None of them change the house edge on the bet itself.

The same maths applies at the roulette table. You can compare the results in our roulette strategies guide.

Martingale on the Player bet

With ties ignored (they push on Player and Banker bets), Player wins about 49.3% of decided hands. Suppose you start at £10 and double after each loss. Six losses in a row cost £10 + £20 + £40 + £80 + £160 + £320 = £630. Your next stake would need to be £640, and winning it brings back only your original £10 profit.

The chance of six Player losses in a row is about 1.7%, or roughly once in every 59 sequences. That sounds rare, but you will run many sequences in a session. When the streak comes, it wipes out dozens of £10 wins. Table maximums can also stop the progression before it recovers.

1-3-2-6 on the Player bet

You stake 1 unit, then 3, then 2, then 6 as long as you keep winning. You go back to 1 after any loss. Four wins in a row earn 12 units. The chance of completing all four steps on Player is about 6%. In most sequences you lose either 1 unit (a loss on the first bet) or 2 units (a loss on the second bet). The system limits your losses per sequence, but the average loss on every unit staked is still 1.24%.

The Paroli works in a similar way. Three straight wins (1, 2, then 4 units) earn 7 units, and that happens only about 12% of the time on Player.

Live and online baccarat tables show roadmaps such as the Big Road and Bead Plate, and land-based tables hand out scorecards. Many players use them to follow streaks (“Banker is hot”) or bet on a change (“Player is due”).

The cards don’t remember earlier results. Over hundreds of hands in an eight-deck shoe, the cards already dealt have almost no effect on the next result. A run of five Banker wins doesn’t make the sixth more or less likely. The roadmaps are a record of what has happened, not a forecast.

Roadmaps can still be useful in one way. If you play at live dealer baccarat tables, they let you see the shoe’s history before you sit down. Just don’t treat that history as a signal for your next bet.

Mikki Mase is a gambler and social media personality. He has said in interviews that he has a long-term edge at baccarat. Reports say he has shown casino win/loss statements to back this up. He has never published his method. When asked, he has spoken only in general terms about “behaviours and characteristics in cards” and about reverse-engineering how casinos design their games.

That means there is no “Mikki Mase strategy” anyone can test. Many sites selling a version of it are guessing. With fixed drawing rules and independent hands, there is no known pattern-based method that can turn Banker or Player into a positive-expectation bet. Many viewers and commentators remain sceptical of his claims.

A sensible approach is to treat any system promising guaranteed baccarat profits with suspicion, especially if it costs money.

• Betting on the Tie. An 8:1 payout looks tempting, but the 14.36% house edge is more than thirteen times the Banker edge.
• Avoiding Banker because of the commission. The 5% charge is already included in the 1.06% figure, and Banker is still the best bet.
• Not reading no-commission rules. In the common version where a Banker win on 6 pays half, the Banker edge rises to about 1.46%. That is worse than the Player bet. EZ Baccarat, where a three-card Banker 7 is a push, has a Banker edge of about 1.02%. Check the paytable before you sit down.
• Playing side bets without checking the edge. Pairs, Dragon Bonus and similar bets usually cost far more than the main bets.
• Chasing losses with progressions. Martingale feels safe until one losing streak costs more than all your earlier wins.
• Ignoring speed. Fast online tables deal many more hands per hour. A low edge on a high number of hands still adds up to real money.

1. Set a session budget and a unit size. A unit of 1–2% of your session budget lets you sit through normal swings.
2. Choose standard commission tables unless the rules of a no-commission variant are clearly shown and no worse.
3. Use a time limit. Expected losses grow with every hand you play, so shorter sessions cost less.
4. Check bonus terms before playing with bonus money. Baccarat is often excluded from wagering or contributes only a small percentage. Our guide to casino bonus terms explains what to look for. A cashback bonus with no wagering is often a better fit for table game players.
5. Play at licensed sites. Our casino reviews check licences and game providers so you can be confident the game runs as described.

• No strategy can make baccarat a profitable game in the long run. The house edge applies to every hand.
• Bet on Banker with flat stakes. At about 1.06%, it is the lowest-edge bet in standard baccarat.
• Progressions such as Martingale, Paroli and 1-3-2-6 only change the pattern of wins and losses, not the average cost.
• Scorecards and roadmaps record past results. They don’t predict the next hand.
• Avoid the Tie bet, and check no-commission rules before you play.

Baccarat is entertainment, not a way to make money. Only stake money you can afford to lose. Set deposit and time limits, and stop if gambling stops being fun. Gambling is for adults aged 18 and over. If you’re worried about your gambling, use the casino’s self-exclusion tools and contact a support service in your country. For more on the game, browse our baccarat guides.