Punto Banco rules. Bet Player (1.24% edge), Banker (1.06% edge, 5% commission on win), or Tie (don’t — 14.4% edge).
| Player win | 1:1 (1.24% house edge) |
| Banker win | 1:1 minus 5% commission (1.06% house edge — best bet) |
| Tie | 8:1 (14.4% house edge — worst bet on the table) |
| Bank wins ~50.7% of decisions; Player ~49.3%; Tie ~9.5%. The 5% commission on Banker exists because the dealing rules favor Banker. | |
Baccarat (specifically Punto Banco) has one of the lowest house edges on the casino floor: 1.06% on the Banker bet, 1.24% on Player, and a brutal 14.4% on the Tie. The simplicity is part of the appeal — you don't make any decisions beyond which side to bet on. Cards are dealt according to fixed rules, and one side wins. Asian high-roller rooms are dominated by baccarat; in some Macau VIP rooms it accounts for over 90% of table revenue.
Two hands are dealt: Player and Banker. (You can bet on either, regardless of which side you sit on — this is not your hand vs the dealer.) Each hand starts with two cards. Card values: 2-9 are face value, 10/J/Q/K count as 0, Ace counts as 1. Only the rightmost digit of the total matters — a 7+8=15 counts as 5. The closer-to-9 hand wins. If either hand totals 8 or 9 on the first two cards (a "natural"), no more cards are drawn. Otherwise, a third card is drawn according to a fixed rules table — the player has no decisions. Wins on Player or Banker pay 1:1, but a winning Banker bet is charged a 5% commission (so a $100 Banker win pays $95). A Tie pays 8:1 (some casinos pay 9:1).
Always bet Banker. The house edge of 1.06% (including the 5% commission) is the lowest non-blackjack bet on the floor. Player is a close second at 1.24%. Tie is a trap — 14.4% house edge — never take it. There are derivative bets (Player Pair, Banker Pair, Big, Small, Dragon, Panda 8, Tiger, etc.) with house edges typically 4-15%; ignore all of them. "Trend" charts in casinos (the colorful grids next to the table showing recent results) feed pattern-recognition instincts, but each hand is essentially independent — past results don't move the next hand's odds. Card counting in baccarat has been studied extensively (Edward Thorp's team analyzed it in the 1960s); the conclusion is that any countable edge is so tiny and so rare that it's not exploitable. Just bet Banker, flat-bet a small percentage of your roll, and accept that the house edge is what it is.
The 1.06% Banker edge factors in the 5% commission. Without commission, Banker wins ~45.9% of hands and pays 1:1; Player wins ~44.6%; Tie hits ~9.5%. If Banker paid 1:1 commission-free, the bet would have a +1.3% edge for the player — that's why the commission exists. The Tie bet at 8:1 expected value: 9.5% × 8 + 90.5% × (-1) = 0.760 - 0.905 = -0.145, a 14.5% house edge. At 9:1 the Tie edge drops to about 4.8%, which is still bad but no longer suicidal. The variance on Banker is low — about 0.93 per bet — making baccarat one of the lowest-variance major casino games, well-suited for large flat bets if you can stomach a long, slow grind.
Baccarat traces to 15th-century Italy and was a French aristocrats' favorite for centuries. James Bond made it famous in Western pop culture (the original "Casino Royale" novel features Chemin de Fer, a baccarat variant where players take turns being the Banker). Modern Punto Banco — dealer plays both hands, fixed third-card rules — was standardized in Cuba in the 1940s and brought to Las Vegas in 1958, where it slowly grew into the high-roller staple it is today.
| Bet | Payout | Notes |
|---|---|---|
| Banker | 0.95:1 | 5% commission — best bet, 1.06% edge |
| Player | 1:1 | 1.24% edge |
| Tie | 8:1 | 14.4% edge — never take it |
| Banker Pair / Player Pair | 11:1 | ~10% edge — skip |
Every game on placebets.ai uses virtual chips that reset whenever you clear browser data. There is no signup, no deposit, no withdrawal mechanism, and no monetary value attached to the chips shown on screen. Use the practice environment to drill baccarat's math and strategy without risk. Decide for yourself whether you ever want to play for real money — we'd statistically rather you didn't.