Baccarat is often described as a “low-edge” casino game, but that reputation depends heavily on the rule set sitting behind the felt. In 2026, most casinos offer classic commission baccarat, at least one no-commission option, and either Dragon Bonus side bets or a separate Dragon/Tiger table. These choices are not cosmetic: they change expected return (RTP), house edge, and how sharply results can swing from session to session. This article compares the most common rules you will meet today, using widely published maths for standard multi-deck games and explaining what those percentages mean in real play.
Classic “punto banco” baccarat pays even money on Player wins and pays Banker wins at 1:1 minus a 5% commission. The commission exists because Banker has a slightly higher probability of winning across the full shoe. Without the fee, a Banker bet would be too strong for the house, so the commission is the balancing mechanism that keeps the casino advantage small and predictable.
For the standard multi-deck version commonly dealt in live casinos and live studios, the usual 2026 reference figures are consistent across reputable odds calculators: the Banker bet has a house edge of about 1.06% (RTP ~98.94%) and the Player bet sits around 1.24% (RTP ~98.76%). The Tie bet is the outlier, with a much steeper house edge around 14.36% (RTP ~85.64%), which makes it expensive over time even if it occasionally lands in your favour.
From a risk perspective, commission baccarat is relatively steady compared to slots or high-variance table games because the main bets pay close to even money. That does not mean it is “safe”. Long streaks happen, and the small edge becomes noticeable as the number of hands increases. The practical point is that the baseline game rewards patience and bankroll discipline rather than chasing big payouts.
Players often focus on the 5% fee emotionally (“I hate paying commission”), but what matters is the effective long-term cost. If you wager £10 on Banker and win, your profit is £9.50 rather than £10. Over a large sample, that reduction is exactly what produces the ~1.06% house edge on Banker. In other words, the commission is not an “extra” disadvantage on top of the maths — it is the maths.
Another practical detail is how casinos handle rounding. Many live tables round commission to the nearest chip unit. With small bets, rounding can nudge your actual cost slightly up or down. On £5 or £10 minimums the difference is usually minor, but if you are making very small stakes in certain online rooms, it is worth checking whether commission is rounded or calculated precisely.
If your goal is to minimise house edge on a standard baccarat table, Banker remains the best mainstream option despite the fee. Player is close, and some players prefer it simply because it is simpler (no commission tracking). The Tie bet, however, changes the entire risk equation because it pays a high multiple while landing relatively rarely, creating bigger swings and a much higher expected loss rate.
No-commission baccarat became mainstream because it removes an operational headache for casinos and a psychological hurdle for players. You see the bet paid instantly at even money, and you do not need to manage commission chips. However, the house edge does not disappear — it is shifted into a rule that trims payout on a specific Banker result or changes the Banker win condition.
The most common version in 2026 is the so-called “Super 6” style rule (often branded as No Commission Baccarat): Banker wins generally pay 1:1, but if the Banker wins with a total of 6, the payout is reduced (commonly to 0.5:1 profit, meaning you receive 1.5 units back for every 1 unit staked). This single adjustment restores the casino advantage because a Banker win on 6 happens often enough to matter over time.
Depending on the exact no-commission rule set, the Banker bet house edge commonly rises to around 1.46% under the “win on 6 pays half” model, while Player remains roughly the same as in classic baccarat. Some casinos advertise alternative “EZ” style variations where a Banker 6 pays 1:2, and published calculators may show slightly different edges. The key point is that “no commission” does not automatically mean “better RTP” — you must read the payout table.
Even if the Banker edge is worse in many no-commission games, there are reasons some players prefer them. First, the experience is simpler: no commission bookkeeping, no extra chips, and usually faster dealing. If you are playing casually and want clarity, the small edge difference may be an acceptable trade-off for smoother play.
Second, in some venues the no-commission table has lower minimum stakes, better speed options, or more comfortable limits than the classic commission table. Your actual “cost” per session is not only house edge — it also depends on how many hands you play per hour and whether you are forced into larger bets than you planned. A slightly higher edge on smaller stakes can sometimes be easier on your bankroll than a lower edge at a higher minimum.
Finally, the best way to judge no-commission baccarat is to treat it like any other game: check the Banker payout rules, confirm the Tie payout (often unchanged and still expensive), and then decide whether the convenience and game speed match your style. If the casino offers a no-commission version with unusually favourable terms, that can shift the decision — but those are the exception, not the standard in 2026.

Dragon-themed options appear in baccarat in two main ways. The first is Dragon Bonus, a side bet that pays when the Player or Banker wins by a large margin (typically 4 points or more, with higher payouts for bigger margins). The second is Dragon/Tiger, a separate casino game where you bet which of two hands (Dragon or Tiger) will be higher, with an optional Tie bet paying a large multiple.
Dragon Bonus side bets usually look attractive because they offer headline payouts that standard baccarat does not. The trade-off is a bigger house edge and much higher variance. Published analyses for common Dragon Bonus pay tables (eight decks) show that the Player Dragon Bonus can sit around 2.65% house edge, while the Banker Dragon Bonus may be dramatically worse (often quoted near 9%+ depending on the table). That means the “fun” part is paid for in expected value, especially on the Banker side.
Dragon/Tiger, while easy to learn, also carries a different profile from baccarat. In many casino versions, the main Dragon and Tiger bets carry a house edge around the 3–4% range, while the Tie bet can be extremely costly because of its high payout and low probability. Some guides quote Tie house edges above 30% for common 11:1 payouts, which makes it one of those bets that can produce dramatic results but is mathematically harsh over repeated play.
Risk profile is not only about house edge — it is about volatility and the likelihood of large short-term swings. Dragon Bonus is designed to produce long stretches of small losses punctuated by occasional bigger hits. If you place it regularly, you should expect streaky results. For many players, this is enjoyable in small doses, but it changes the feel of baccarat from a steady grind into something closer to a side-bet lottery.
Dragon/Tiger behaves similarly: the main bets resolve quickly and feel straightforward, but the Tie bet turns the game into a high-variance proposition. If you treat Dragon/Tiger as a “fast coin-flip with a house fee”, it becomes easier to manage. If you treat the Tie as the core of the game, you are effectively choosing a high-cost gamble where the long-term maths is stacked against you.
If your priority is maximising RTP and keeping risk controlled, the classic Banker bet in commission baccarat is typically still the most efficient mainstream option. If your priority is entertainment and you accept bigger swings, Dragon Bonus or Dragon/Tiger can deliver that — but the sensible approach is to separate them mentally from your core bankroll. In practical terms: keep stakes small, set a clear limit, and remember that higher payouts usually mean you are paying for volatility through a larger house edge.