Baccarat Card Counting: Does It Work? (Honest Answer With the Math)
Card counting made millions for blackjack players. It turned MIT students into legends. It inspired books, movies, and an entire subculture of advantage play. So naturally, people ask: can you count cards in baccarat? The short answer is yes, you can count. The honest answer is that it’s almost certainly not worth your time.
Baccarat card counting produces a theoretical edge so small (roughly 0.01% to 0.05% under ideal conditions) that you’d need to play thousands of hands, sit through dozens of shoes without placing a bet, and wager big at precisely the right moments to see any meaningful profit.
Compare that to blackjack counting, which can yield edges of 0.5% to 1.5%, and the picture becomes clear. This guide explains exactly how baccarat card counting works, the specific system used, and why the math makes it a poor investment of your attention.
- Baccarat card counting is theoretically possible but produces an edge of only 0.01% to 0.05%, far too small to be practically profitable
- The most common system assigns values from +2 to -2 to each card, tracking whether the remaining shoe favors Player or Banker
- A “true count” of 16 or higher signals a potential switch from Banker to Player betting, but this threshold is rarely reached
- In blackjack, removing a single 5 from the deck shifts the player’s edge by about 0.67%; in baccarat, removing any single card shifts the odds by less than 0.01%
- Card counting in baccarat is legal (it’s mental math, not cheating), but casinos can ask you to leave if they suspect you’re counting
- Your time and energy are better spent on bankroll management and bet selection than on trying to count cards in baccarat
Why Card Counting Works in Blackjack But Barely Works in Baccarat
To understand why baccarat card counting is practically useless, you need to understand why it’s powerful in blackjack. The comparison explains everything.
In blackjack, the player makes decisions (hit, stand, double, split) that change depending on which cards remain in the shoe. When the remaining deck is rich in tens and aces, the player benefits disproportionately: more naturals (which pay 3:2), better double-down opportunities, and improved insurance value. A skilled counter can identify these favorable conditions and increase their bets. The edge gain is substantial, often 0.5% to 1.5% over the house.
In baccarat, you make zero decisions during the hand. The third card rules are automatic. The dealing follows a fixed tableau regardless of what’s left in the shoe. Removing a specific card from the deck shifts the probabilities of Player and Banker winning by fractions of a fraction of a percent. The Banker bet’s house edge is 1.06% with a full shoe. Even in the most favorable shoe compositions, it barely dips below 1%.
Peter Griffin, one of the most respected gambling mathematicians, analyzed baccarat card counting in “The Theory of Blackjack.” His conclusion: the theoretical edge available through baccarat counting is so tiny that a player would need to sit at the table for extended periods without betting, waiting for the rare moment when the count becomes favorable, and then wager a very large amount. The practical return on this investment of time and bankroll is essentially zero for the vast majority of players.
How the Baccarat Card Counting System Works
Despite its limited practical value, the system itself is straightforward. If you want to understand it for academic interest or to see the math for yourself, here’s how it operates.
The Card Values
The most commonly referenced baccarat counting system assigns the following values to each card as it’s dealt from the shoe:
| Card | Count Value | Effect on Count |
|---|---|---|
| Ace, 2, 3 | +1 | Pushes count higher (favors Player) |
| 4 | +2 | Strongly favors Player |
| 5, 7, 8 | -1 | Pushes count lower (favors Banker) |
| 6 | -2 | Strongly favors Banker |
| 9, 10, J, Q, K | 0 | Neutral (no effect) |
Running Count vs True Count
As cards are dealt, you maintain a running count by adding or subtracting the assigned values. If the first three cards dealt are a 3 (+1), a 6 (-2), and a King (0), your running count after those three cards is -1.
The running count alone isn’t enough. You need to convert it into a true count by dividing by the estimated number of decks remaining in the shoe. This adjusts for the fact that a running count of +20 means something very different with 7 decks left versus 2 decks left.
True count = Running count / Estimated decks remaining
When to Switch Bets
The default bet in baccarat is Banker, because it carries the lowest house edge at 1.06%. The counting system tells you when conditions theoretically favor the Player bet instead.
The threshold: when the true count reaches 16 or higher, the shoe’s composition slightly favors Player over Banker. At this point, a counter would switch from Banker to Player betting.
When the true count drops to a large negative number, the Banker bet becomes even more favorable than usual. Counters would continue betting Banker (or increase their bet size).
Twenty more hands pass. Running count: +38. Roughly 2.5 decks remain. True count: 38 / 2.5 = 15.2. Still below 16. Continue on Banker.
Five more hands. Running count: +42. About 2 decks left. True count: 42 / 2 = 21. That’s above 16. The system says switch to Player. You’ve played 65 hands of the shoe before the count reached the threshold, and you have maybe 10 to 15 hands left before the cut card ends the shoe.
In this scenario, you’d bet Player for those final hands. The theoretical edge on those Player bets might be 0.03% to 0.05% above normal. On a $100 bet across 12 hands ($1,200 total action), that’s an expected gain of approximately $0.36 to $0.60. For 90 minutes of mental effort.
The Math: Why the Edge Is So Small
The reason blackjack counting works is that specific card removals have large effects on the game’s probabilities. Removing a single 5 from a blackjack deck improves the player’s expected value by approximately 0.67%. That’s huge. It’s enough to flip the house edge into the player’s favor once enough small cards have been dealt.
In baccarat, the effect of removing any single card on the Player/Banker probabilities is microscopic. Removing a 4 (the highest-impact card in baccarat counting) shifts the odds by roughly 0.009% per card. Removing a 6 shifts them by about 0.008% in the opposite direction. These shifts are so small that even a highly favorable shoe composition barely moves the needle.
| Card Removed | Effect on Player Win Rate | Comparison: Blackjack Effect |
|---|---|---|
| Ace | +0.005% | -0.59% (blackjack) |
| 4 | +0.009% | +0.52% |
| 6 | -0.008% | +0.44% |
| 8 | -0.005% | -0.51% |
| King/10 | ~0% | -0.51% |
The blackjack numbers are roughly 50 to 100 times larger than the baccarat numbers. That’s the entire story. Baccarat’s card removal effects are so tiny that even a perfect counting system, executed flawlessly over thousands of hands, produces a negligible return.
Multiple academic analyses, including work by John May and the Wizard of Odds, have confirmed that baccarat card counting offers an expected gain of approximately $0.70 per 100 hours of play under realistic conditions [VERIFY]. That’s not a typo. Seventy cents per hundred hours. You’d earn more picking up coins from the casino floor.
Why Baccarat Resists Counting
Several structural features of baccarat make card counting far less effective than in blackjack.
The 8-Deck Shoe
Most baccarat tables use 8 decks (416 cards). The larger the shoe, the more cards need to be dealt before the remaining composition diverges meaningfully from the starting proportions. In blackjack, single and double-deck games are the most vulnerable to counting because each card dealt has a proportionally larger impact. An 8-deck baccarat shoe dilutes every card removal effect by a factor of 8 compared to a single-deck game.
No Player Decisions
In blackjack, the counter’s advantage comes from making better decisions (when to hit, stand, double, or split) based on the remaining cards. These decisions compound across hundreds of hands to produce a meaningful edge. In baccarat, there are no decisions. The tableau is fixed. Whether the shoe is rich in 4s or 6s, the drawing rules don’t change. The only decision a counter can make is which bet to place (Player or Banker) and how much.
High Shoe Penetration Required
For baccarat counting to reach the threshold where switching bets makes sense (true count of 16+), the shoe needs to be well past its midpoint. Most of the time, the true count stays within a narrow band around zero, never reaching actionable territory. In a typical 8-deck shoe, you might see a true count above 16 in fewer than 5% of shoes played. Practically speaking, you’d sit through 19 or 20 shoes waiting for one opportunity.
The Commission Factor
Even when the count says “bet Player,” the edge is minuscule. And you’ve been betting Banker (with its 5% commission on wins) for the 60+ hands before the count triggered. The commission you’ve paid all shoe long dwarfs whatever tiny edge the Player bet offers in those final hands.
Is Baccarat Card Counting Legal?
Yes. Counting cards is legal in every jurisdiction because it involves only mental math. You’re not altering the game, marking cards, or using a device. You’re thinking. There’s no law against that.
However, casinos are private businesses, and they can refuse service for any reason. If a pit boss suspects you’re counting cards, they can ask you to stop playing baccarat (or leave the property entirely). In practice, casinos rarely bother watching for baccarat counters because the technique is so ineffective. Their surveillance resources focus on blackjack counters, where the stakes are much higher.
The famous cases of advantage play in baccarat, like Phil Ivey’s edge sorting wins, involved techniques far beyond card counting. Courts treated edge sorting differently from counting because it required the casino’s equipment (specifically, imperfect card backs) to be manipulated in a specific way. Card counting involves nothing but your own brain.
What You Should Do Instead
If card counting in baccarat isn’t worth the effort, what is? Here’s where to focus your energy for a genuinely improved baccarat experience.
Bet Banker (Almost Always)
The Banker bet at 1.06% house edge is the best wager on the table. No counting system, no pattern following, no road analysis changes that fact. Bet Banker as your default, and you’re already playing at near-optimal level. Our winning strategies for baccarat guide explains why this is the foundational principle of smart baccarat play.
Manage Your Bankroll
Bankroll management has a far greater impact on your session outcomes than any counting system ever could. Bringing 40 to 50 times your base bet, setting win targets and loss limits, and walking away at the right moment are the closest things to genuine “strategy” in baccarat.
Avoid the Expensive Bets
The Tie bet (14.36% house edge) and most side bets (5% to 25% house edges) cost you orders of magnitude more per dollar wagered than any counting-derived edge could ever recover. Skipping the Tie bet saves you more money per session than counting cards would earn in a year.
Understand Volatility
Short sessions produce wild swings. Long sessions grind down your bankroll. Understanding baccarat volatility helps you set realistic expectations and avoid the emotional traps that cost more than any house edge.
Practice the Game, Not the Count
If you want to sharpen your understanding of baccarat, spend your practice time on our free baccarat simulator. Watch how the roads develop over a shoe. Get comfortable with the pace, the betting, and the psychology of the game. That awareness is far more valuable than a counting system that earns seventy cents per hundred hours.
- Theoretically produces a tiny positive edge under specific shoe compositions
- The counting system itself is simple to learn (simpler than blackjack counting)
- Legal in all jurisdictions since it involves only mental calculations
- Can marginally improve Player/Banker bet selection in rare circumstances
- Produces an edge of only 0.01% to 0.05%, effectively zero in practical terms
- Requires sitting through dozens of shoes before actionable counts appear
- Expected profit is approximately $0.70 per 100 hours under realistic conditions
- The 8-deck shoe dilutes card removal effects to insignificance
- Time and mental energy are far better spent on bankroll management and bet selection
- Casinos can still ask you to leave if they detect counting behavior
The Honest Verdict on Baccarat Card Counting
Baccarat card counting is a real technique with real mathematics behind it. The counting system works as described: assign values, track the running count, convert to a true count, and switch from Banker to Player when the count says so. Everything about the process is valid.
What isn’t valid is the expectation that it will make you money. The edge is too small, the opportunities too rare, and the time investment too large. The Banker bet’s 1.06% house edge is already one of the best wagers in any casino. Counting cards to shave a few hundredths of a percent off that number is like polishing a coin before spending it. Technically better. Practically pointless.
If you’re drawn to baccarat because of the math, focus on the math that matters: the odds and house edge of each bet, the impact of side bets on your hourly cost, the relationship between bet sizing and volatility, and the discipline of solid bankroll management. That’s where the real edge is. Not in counting cards, but in counting the things that actually matter.
Baccarat Card Counting FAQs
Yes, it’s technically possible. You assign values to each card (+1, +2, -1, -2, or 0), maintain a running count, convert it to a true count by dividing by the remaining decks, and switch from Banker to Player betting when the true count exceeds 16. The system is legal and mathematically sound. However, the edge it produces (0.01% to 0.05%) is so small that it’s not practically profitable.
No. Card counting is legal in all jurisdictions because it involves only mental math. You’re not using a device or altering the game. However, casinos are private businesses and can ask you to stop playing or leave the property if they suspect counting. In practice, casinos rarely monitor for baccarat counters because the technique is too ineffective to threaten their bottom line.
Under realistic conditions, academic analysis suggests an expected gain of approximately $0.70 per 100 hours of play. The edge per bet is roughly 0.01% to 0.05% during the rare moments when the count becomes actionable. Compare this to blackjack, where skilled counters can earn $25 to $100+ per hour. Baccarat card counting is not a viable income strategy.
In blackjack, removing specific cards (especially 5s and aces) shifts the player’s expected value by 0.5% to 0.67% per card. This is large enough to flip the house edge. In baccarat, removing any single card shifts the probabilities by less than 0.01%. The 8-deck shoe further dilutes the effect. Additionally, blackjack players make decisions (hit, stand, double) that exploit favorable compositions; baccarat players make no decisions during the hand. For more on baccarat odds, see our house edge guide.
Bet Banker as your default (1.06% house edge), avoid the Tie bet (14.36%), skip most side bets (5% to 25% edges), and practice solid bankroll management. These fundamentals save you more money per session than any counting system could earn in hundreds of hours. Our winning strategies guide covers the full approach.
Historically, yes. Edge sorting has produced far larger advantages (Phil Ivey won roughly $20 million using the technique). However, edge sorting requires specific conditions (particular card brands with asymmetric back patterns, a cooperative dealer) and has been successfully challenged in court by casinos. In [current_year], it’s extremely difficult to execute because casinos now use cards designed to prevent it.
