Level 1 · Lesson 4 of 4 · Foundations
Banker vs Player vs Tie: The Only Decision That Matters
What the three bets actually pay
| Bet | Payout if you win | House edge |
|---|---|---|
| Banker | 19 to 20 (1 to 1, less 5% commission) | 1.06% |
| Player | 1 to 1 (even money) | 1.24% |
| Tie | 8 to 1 | 14.36% |
The Banker payout is the most confusing of the three for new players. You bet $100, you win, you get $95 back as profit. The other $5 goes to the house as commission. The reason the casino does it this way, rather than simply paying out a lower amount on the win, is accounting: it preserves the visible "even money" feel of the table while extracting a known house margin.
Some tables run no commission baccarat, also called Super 6, where Banker wins pay full even money except when Banker wins with a total of 6, in which case the payout drops to 1 to 2 (you get $50 profit on your $100 bet, not $100). The house edge on no commission Banker is 1.46%, which is worse than the standard 1.06% on commission Banker. Avoid Super 6 tables unless you're explicitly forced into one.
Why Banker wins more often
Banker wins about 45.86% of hands. Player wins 44.62%. The remaining 9.52% are ties.
The reason is structural. Banker acts second under the third-card tableau, and the tableau is built to give Banker slightly better information. When Player draws a third card, Banker's decision is informed by what that card was. This isn't a coincidence. It's how the rules were designed in the late 1950s, and unchanged since. The 5% commission exists precisely to absorb this advantage. Without the commission, Banker would carry a roughly 1.24% edge in favour of the player. With the commission, the house has 1.06% on Banker and 1.24% on Player. The commission converts most of Banker's structural advantage into casino margin, while still leaving Banker as marginally the best of the two main bets.
It's worth noting that experienced players at Marina Bay Sands in Singapore almost invariably bet Banker. The locals who fill the main baccarat pit there aren't particularly superstitious about which side is running. They've done the arithmetic, and Banker at 1.06% is the rational choice. The tourists who bet Player or chase Tie are the ones keeping the margin healthy for everyone else.
Why the Tie is the worst bet at the table
Tie pays 8 to 1. That sounds like a lot. $100 on a Tie that hits returns $800 profit, plus your stake. People back the Tie because the payout looks generous.
The maths is unforgiving. Ties happen on 9.52% of hands. To break even on Tie at 8 to 1, you'd need ties on 1 in 9 hands, or 11.11% of the time. The actual rate is 9.52%. The shortfall, multiplied across the difference between the true rate and the break-even rate, gives the house a 14.36% edge.
That is the worst-edge bet on any main casino game table outside of slot machines. It's roughly 14 times worse than the Banker bet. A handful of casinos pay Tie at 9 to 1, which cuts the house edge to 4.85%. That's still bad, but it's at least defensible. If you're going to bet Tie at all, do it only on a 9 to 1 table, which is rare. Most modern tables, including all the major London rooms, run 8 to 1.
The pattern problem
A lot of baccarat literature, and almost all the roadmaps you'll see scrolling on the digital displays at Macau tables, are built on the assumption that you can read streaks and reverse them. Banker has won six in a row. Player is "due". Or Banker is "running", and you ride the streak.
Neither is true. The shoe is mechanically reshuffled and the next hand is independent of the last one. The base rate of a Banker win is 45.86%. After six Banker wins in a row, the base rate for the next hand is 45.86%. After six Player wins in a row, the base rate for the next hand is still 45.86% for Banker. The cards don't remember.
This is the gambler's fallacy in its purest form. The reason it persists in baccarat more than other games is that the scoreboard displays at every Macau table are designed to make patterns visually obvious. The Big Road, the Big Eye Boy, the Small Road, the Cockroach Road. They're beautiful. They're also useless for prediction. We'll cover roadmaps in detail in lesson fifteen.
The maths is settled. The next hand's odds don't change because of what the last six were. Bet Banker, or bet Player, or vary between the two for variance reasons if you must. But don't pretend a pattern is information. It isn't.
What to do, in one sentence
Bet Banker. Bet a stake you've decided in advance to bet. Walk when your session budget is spent. The other 99% of what's written about baccarat strategy is either embroidery on this paragraph or actively wrong.
Key numbers
| Bet | Win rate | Payout | House edge | Expected loss per $100 over 100 hands |
|---|---|---|---|---|
| Banker (5% commission) | 45.86% | 19 to 20 | 1.06% | $106 |
| Player | 44.62% | 1 to 1 | 1.24% | $124 |
| Tie (8 to 1) | 9.52% | 8 to 1 | 14.36% | $1,436 |
| Tie (9 to 1, rare) | 9.52% | 9 to 1 | 4.85% | $485 |
| No commission Banker | 45.86% | 1 to 1 (1 to 2 on 6) | 1.46% | $146 |
Sources: Marina Bay Sands casino, Hippodrome Casino baccarat, DICJ Macau monthly statistics.
Welcome to the lesson on Banker versus Player versus Tie.
I'm Annabel, and this is the lesson that determines whether you'll play baccarat well or badly for the rest of your life. The decision you make before the cards are dealt is the only decision in the game, and it's worth understanding exactly what each of the three options costs you.
Let's start with the numbers, because the numbers settle the argument.
Banker carries a house edge of one point zero six percent. Player carries a house edge of one point two four percent. Tie pays eight to one and carries a house edge of fourteen point three six percent. That last figure is not a typo. Tie is roughly fourteen times worse than Banker. It is the worst main bet on any major table game.
The three payouts. Banker pays even money minus a five percent commission. Win one hundred on Banker, get ninety-five back as profit. Player pays even money straight. Tie pays eight to one. There's also a variant called no commission baccarat or Super Six where Banker pays full even money except when Banker wins with a total of six, which pays one to two. The edge on no commission Banker is one point four six percent, which is worse than the standard Banker bet. Avoid Super Six tables unless you have no choice.
Why does Banker win more often? Because the tableau is built in Banker's favour. Banker acts second under the third-card rule, and the rule gives Banker information about what Player just drew. Banker wins about forty-five point eight six percent of hands. Player wins forty-four point six two percent. Ties happen nine point five two percent of the time. The five percent commission on Banker wins is the casino's way of clawing back most of Banker's structural advantage and turning it into house edge. Without the commission, Banker would actually favour the player. With it, the house has one point zero six percent on Banker, one point two four percent on Player, and Banker stays the marginally better bet.
I've sat at tables in Singapore and watched the experienced local players at Marina Bay Sands. They almost universally back Banker. Not out of habit. Out of arithmetic. They know the structure of the game and they take the better side. The tourists who drift to Player or chase the Tie are the ones funding the room's margin.
Now the Tie, because this is where most casual players burn money.
Tie pays eight to one. That looks generous. The maths is unforgiving. Ties happen on nine point five two percent of hands. To break even on a Tie bet at eight to one, you'd need ties on roughly one hand in nine, or eleven point one one percent. The actual rate is two percentage points lower than break-even. The gap gives the house a fourteen point three six percent edge.
A handful of casinos, mostly in Asia, pay Tie at nine to one. That cuts the edge to four point eight five percent. Still bad, but defensible. Most modern tables, including every major London room, pay eight to one. So unless you're at a specific nine-to-one table, the Tie is a tax dressed up as a long-shot.
Now the part where most baccarat players go wrong, which is patterns.
Almost every Macau table has a digital scoreboard above it showing what's called a roadmap. The Big Road. The Big Eye Boy. The Small Road. The Cockroach Road. They're beautiful, geometric displays of recent hand outcomes, designed to show you streaks and patterns. Players use them to decide whether to follow a streak or bet against it.
Here's the thing I most want you to know. The roadmaps are decoration. They are not data. The next hand is mechanically independent of the last one. After six Banker wins in a row, the odds of Banker winning the next hand are forty-five point eight six percent. After six Player wins in a row, the odds of Banker winning the next hand are still forty-five point eight six percent. The cards don't remember.
This is the gambler's fallacy in its purest form. The reason it persists in baccarat more than other games is that the displays are designed to make patterns visually obvious. They make you feel like the shoe is telling you something. It isn't. We cover the roadmaps properly in lesson fifteen. For now, the one thing to know is that they don't change the odds of the next hand.
So what do you do? You bet Banker. You bet a stake you decided to bet before you sat down. You walk when the session budget is spent.
Here's the closing arithmetic. Stake one hundred per hand over one hundred hands. On Banker, your expected loss is one hundred and six dollars. On Player, one hundred and twenty-four. On Tie, fourteen hundred and thirty-six.
Bet the side with the lowest edge. Bet the stake you decided to bet. Walk when the budget's gone.