Baccarat Strategy

The Honest Answer Upfront

Most baccarat strategy guides try to dress up “bet Banker” with extra noise to feel like a substantial topic. We are not going to do that. Here is what the math actually says, in three lines:

The reason this is the entire baccarat strategy is structural. The Banker bet’s 1.06% house edge is the lowest standard bet in baccarat. The math does not change based on the previous shoe, the dealer, the scoreboard pattern, or the streak you are on. Every other piece of “advice” you can google about baccarat (Martingale, 1-3-2-4, Paroli, Fibonacci, Labouchere, scoreboard reading, pattern following, “trend” trading) either fails to change the math, or actively makes your expected return worse.

The rest of this guide explains why those things do not work, what the four decisions you actually control look like, and how to play a baccarat session correctly when the only real “strategy” is the three lines above.

What You Actually Control: Four Decisions

Baccarat has no in-play decisions, so “strategy” really refers to four choices you make before placing a bet. Each one is small in isolation; together they are the difference between a recreational session that loses at the mathematical floor (1.06%) and one that loses at multiples of that rate.

Decision 1: Which bet to place. Banker every hand. The Banker bet’s 1.06% house edge beats the Player bet’s 1.24% by a small but permanent margin. The Tie bet’s 14.36% house edge is roughly fourteen times worse than Banker. Side bets sit between 2.65% and 30% house edge, all worse than Banker. There is no version of the math that recommends anything other than Banker, every hand, forever. Our bet types and payouts guide walks through every bet with its specific house edge.

Decision 2: Which variant to play. Standard commission-based Punto Banco (1.06% Banker edge) is the default and the second-best you can do. EZ Baccarat (1.02% Banker edge) is marginally better when you can find it. No Commission Baccarat / Super 6 (1.46% Banker edge) is meaningfully worse despite the “no commission” framing that sounds player-friendly. The variant matters because the choice persists across every hand of the session; picking the wrong variant adds 0.4 percentage points to the house edge on every bet. Our variants guide covers the rules for each.

Decision 3: How much to bet per hand. Recreational bankroll guidance is 1% to 2% of your session bankroll per hand. RM500 bankroll means RM5 to RM10 per hand. The temptation to size up after a few wins (“hot table”) or after a few losses (“due for a win”) is the most common variance-amplifying mistake; both feelings are illusions and both cost money. Bet a fixed fraction and let the math work over many hands.

Decision 4: When to stop. Set a stop-loss and a stop-win before you sit down, in absolute ringgit amounts. The stop-loss prevents catastrophic sessions; the stop-win prevents the more common pattern of giving back early gains. Most recreational players are far better at honouring stop-losses than stop-wins; the difference between a winning month and a losing one is often the discipline to walk away when up rather than chasing higher highs.

Notice what is absent from this list. Nothing about reading the scoreboard. Nothing about waiting for streaks to develop. Nothing about progressive bet sizing. Nothing about hand-by-hand “feel.” Those are not decisions that change your outcome; they are decisions that change how you feel about your outcome while it happens.

What Does Not Work: The Honest List

Almost every search for “baccarat strategy” surfaces techniques that promise to beat the house. Almost all of them fail in the same way: by misunderstanding how the math works. Here is the honest summary of the major categories.

Pattern following. “Banker has won 5 in a row, ride the streak” or “Player is due, the shoe owes us.” Both ideas come from the gambler’s fallacy: the belief that independent events develop momentum or memory. They do not. Each baccarat hand has a 45.84% Banker win probability, a 44.62% Player win probability, and a 9.54% Tie probability, regardless of what happened in the previous 1, 10, or 100 hands. A 5-Banker streak does not predict the 6th hand will be Banker. A 5-Banker streak also does not predict the 6th hand will be Player. The next hand has the same 45.84% Banker probability as every other hand.

Scoreboard reading. Big road, bead road, big eye boy, small road, cockroach road. Asian baccarat tables display these prominently and many players study them carefully. The scoreboards are accurate records of past outcomes. They do not predict future outcomes because there is no autocorrelation in baccarat to detect. The pattern-tracking culture is real and longstanding; the math behind it is wrong. Our patterns and scoreboards guide explains exactly what each road represents and why none of them help.

Progressive betting systems. Martingale (double after losses), Paroli (double after wins), Fibonacci (follow the sequence), 1-3-2-4 (a four-bet progression), D’Alembert (increase by one unit after losses), Labouchere (cross out numbers from a sequence). Every progressive system has the same flaw: rearranging the size of negative-expected-value bets does not flip the sign of the sum. The systems produce different equity curves (lots of small wins followed by occasional catastrophes is Martingale’s shape; the opposite for Paroli) but the same long-run loss rate. Our betting systems guide covers each major system with the specific math of why it fails.

Card counting. Theoretically possible, practically useless. Edward Thorp, James Grosjean, and Peter Griffin all studied counting in baccarat and concluded that a perfect counter gains approximately 0.01% per hand in expected value. One basis point. The Banker bet’s structural edge cannot be flipped by tracking deck composition because the game’s symmetry leaves no cards that meaningfully favour one side. Our advantage play guide covers the math.

“Hot” and “cold” tables. Some players believe certain dealers, tables, or studios run hotter or colder than others. They do not. Modern shoes are shuffled to industry-standard random distributions, online RNG baccarat passes regular independent audits, and the math of the game is identical across every table that follows the same rules. Streaks happen because variance is real, not because tables have personalities.

“Following the dealer” or “reading the dealer.” The dealer has no influence on outcomes. The cards are dealt according to fixed third-card rules covered in our rules guide. The dealer applies the rules; they do not make discretionary choices. Believing you can read the dealer is believing the dealer is doing something other than mechanically executing the rules.

Why People Keep Searching for Baccarat Strategy

If the honest answer is “bet Banker,” why are there thousands of baccarat strategy guides, hundreds of betting systems, dozens of pattern-tracking methodologies, and an active culture of scorecard study? The answer is psychological, commercial, and cultural in roughly equal parts.

Pattern-seeking is wired into human cognition. The same mental machinery that recognises a tiger in the long grass also produces dragon streaks and ping-pong patterns on a baccarat scoreboard. Recognising patterns where none exist is not a failure of intelligence; it is a default cognitive setting that evolution selected for. Casinos do not have to teach players to look for patterns; humans bring that instinct with them.

Casinos benefit from strategy-thinking. A player who believes their strategy gives them an edge plays longer, bets larger, and absorbs more total wagering than a flat-bet recreational player. Casino floor management welcomes scoreboard study, system play, and pattern reading because all three increase the total volume of wagered money. The house edge applies to volume, so anything that increases volume increases the casino’s expected revenue. Resort casinos in Macau and Genting do not actively discourage scoreboard culture; they print the scorecards and provide the pens.

Affiliate content sells the dream. A baccarat strategy guide that tells you to bet Banker and skip everything else generates fewer clicks and fewer ad impressions than one that promises a system to beat the house. The result is a content ecosystem where the honest answer (a paragraph) is buried under thousands of pages of false strategy (an industry). The honest answer is also the boring answer; pattern-following at least feels like agency, even when it changes nothing.

The win you remember overrides the losses you forget. A scoreboard reader who hit a four-Banker run after spotting a “dragon” pattern remembers that win vividly. The dozen sessions where the same approach lost are filed under “bad night.” Selection bias in memory produces a strong subjective sense that the strategy “works sometimes,” which is enough to keep playing it.

None of these reasons make the strategies effective. They explain why the strategies persist despite being ineffective. The math has not changed since the 19th century.

The Mathematical Floor: 1.06% Is As Good As It Gets

The Banker bet at standard Punto Banco rules has a 1.06% house edge. EZ Baccarat tightens this to 1.02% when available. These are the two best bets in the game; there is no third option that does better at standard tables.

Compared to other casino games, this is genuinely competitive:

Baccarat is harder to play badly than blackjack (no decisions to get wrong) but cannot be played as well as blackjack (no basic-strategy advantage to capture). For a casual player who is not committed to perfect blackjack execution, baccarat at 1.06% may genuinely be the lowest-edge game accessible to them.

What this means at the table: for every RM1,000 you wager on Banker over a long session, expected loss is RM10.60. Over many sessions and many thousands of hands, the actual loss will converge toward this number. Variance can push individual sessions sharply above or below the average, but the average itself does not move.

No “strategy” you apply changes the 1.06%. The number is fixed by the rules and the payouts; the only way to change it is to change the table you play at (a different variant) or stop playing (the house edge does not apply to bets you do not make).

Bankroll Discipline: The Real Strategy

Bankroll management is the only baccarat strategy that actually delivers measurable results, not because it changes the math, but because it changes how much you wager. The house edge applies to total wagering; reducing your total wagering reduces your total expected loss in direct proportion.

Set a session bankroll before you sit down. This is the amount you are prepared to lose during this session, full stop. If you brought RM500 to the casino, your session bankroll is RM500. If RM500 is more than you can afford to lose, the bankroll is too high, not the game.

Bet 1% to 2% of bankroll per hand. RM500 bankroll means RM5 to RM10 per hand. This sounds small if you came expecting to bet RM50 per hand on a RM500 bankroll. It is intentionally small. Bet sizing of 10% or more produces extreme variance: a few unlucky hands can wipe out a session in minutes. Bet sizing of 1% to 2% lets the math work over many hands and produces sessions that match the expected math much more closely.

Set a stop-loss and a stop-win before you start. Stop-loss is the easier of the two. Pick a number (commonly 50% to 80% of session bankroll) and walk away when you hit it. Stop-win is harder and more valuable. Pick a number (commonly 30% to 50% above session bankroll) and walk away when you hit it. The wins you take are the wins you keep; the wins you leave on the table are usually the wins you give back.

Walk away when you hit either limit. Not “one more shoe to see if I can get it back.” Not “let me see if this streak continues.” Walk away. The next session is a new session.

What Playing Correctly Looks Like

A session of correct baccarat play has a specific shape. It looks nothing like the cinematic baccarat scene and very little like what most players actually do at the table.

You arrive with a fixed bankroll and a fixed bet size. You sit at a Punto Banco table (preferably 5% commission, accepting EZ Baccarat if available, avoiding No Commission). You note the stop-loss and stop-win you set before you arrived.

You bet Banker every hand for the duration of the session. You do not look at the scoreboard. If the scoreboard is showing a 6-Banker streak, you bet Banker. If the scoreboard is showing a 6-Player streak, you bet Banker. If the scoreboard is alternating ping-pong style, you bet Banker. You bet Banker because the math says Banker; the scoreboard does not change the math.

You ignore the side bets. Dragon Bonus offers, Pair bets, Lucky 6, Super 6, Tie payouts: you do not place them. The fact that they have higher payouts is the trade for their higher house edges; you have chosen not to participate in that trade.

You bet the same amount every hand. Not larger after wins, not larger after losses, not larger when the scoreboard “looks favourable.” Flat betting at 1% to 2% of session bankroll, no progressions.

You walk away when you hit your stop-win or stop-loss. Not at a “natural” stopping point, not “after one more shoe,” not when the dealer changes. At the number you set before you arrived.

This style of play is not exciting. It also produces the lowest expected loss baccarat allows, with the lowest variance per session. The boredom is the strategy working.

Frequently Asked Questions