Banker vs. Player in Baccarat: What the 5% Commission Really Buys You

Walk up to any baccarat table in Vegas, Macau, or Monaco and you’ll see a printed sign telling you the house charges 5% commission on winning Banker bets. Most players nod, shrug, and keep playing — they don’t actually know what they’re paying for. Here’s what’s really happening: the Banker bet wins more often than the Player bet, and if the casino didn’t claw back a slice of those wins, the math would tilt in your favor. The 5% commission isn’t a tip or a courtesy fee. It’s the casino’s surgical fix for a game that, left alone, would be a losing proposition for the house. I want to walk through the banker vs player baccarat math in detail, because once you see the numbers, the whole game makes a lot more sense.

Why the Banker bet wins more often than the Player bet

Baccarat looks like a coin flip, but it isn’t. The Player and Banker hands aren’t symmetrical, and the asymmetry lives in the third-card drawing rules. The Player hand always acts first under a simple, fixed rule — if its two-card total is 0–5, it draws; if it’s 6–7, it stands. The Banker hand, though, gets to react. Its drawing decision depends on the Banker’s own total and on what third card the Player drew (if any). That conditional structure is the whole game.

Think about what that means in practice. The Banker effectively has more information when it makes its decision, and the rule table has been tuned so that the Banker tends to draw when drawing helps and stand when standing helps. It’s a small edge — we’re not talking about blackjack-level swings — but it’s persistent. Over millions of shoes, the Banker hand wins about 1.24 percentage points more often than the Player hand, and that’s after we strip out ties.

I’ll be honest: when I first studied baccarat, I assumed the drawing rules were arbitrary, the kind of thing some 19th-century French casino owner sketched on a napkin. They aren’t. They’re a mathematically engineered asymmetry. For the deep mechanical breakdown of the rules table, the Wizard of Odds baccarat basics page walks through every branch.

The raw probabilities you need to memorize

Across an 8-deck shoe (the standard configuration in almost every casino on Earth), the per-hand probabilities work out like this:

• P(Banker win) = 0.4586 — about 45.86% of all hands
• P(Player win) = 0.4462 — about 44.62% of all hands
• P(Tie) = 0.0952 — about 9.52% of all hands

Those three numbers sum to 1.0000, which they’d better. Notice the gap: Banker beats Player by 0.4586 − 0.4462 = 0.0124, or 1.24 percentage points. That’s the entire engine of the commission. If you’ve ever wondered why every baccarat scoreboard you’ve seen shows roughly the same B-to-P ratio, that 1.24% is why. It isn’t streakiness — it’s structural.

What would happen without commission — Banker EV would be +0.024

Let’s run the math on a no-commission Banker bet, just to see what the casino is protecting against. Both Banker and Player bets push on ties (you get your money back), so ties drop out of the EV calculation. With a $1 bet and 1-to-1 payout:

EV(Banker, no commission) = (+1)(0.4586) + (−1)(0.4462) + (0)(0.0952) = +0.0124

Hmm, that’s curious — I quoted 0.024 earlier and you might’ve caught that. Let me clean this up. The pre-commission edge is +1.24% on the Banker side, not 2.4%. (A common misreading conflates the gap in win probability with a doubled figure that doesn’t actually appear in the EV equation.) Either way, the conclusion is the same and frankly damning for the house: if you ran baccarat with no commission, every rational player would bet Banker every single hand, and the casino would slowly bleed out. Not catastrophically — 1.24% is small — but it’s negative for the house, and casinos don’t run negative-EV games. They can’t.

Commission as a balancing fee, not a punishment

This is where I think most players get the wrong mental model. The 5% commission isn’t a penalty for winning. It’s a corrective term — a balancing fee that drags the Banker EV back across the zero line and into house-favorable territory. The casino is essentially saying, “Sure, you can take the statistically better bet. We’ll just rent it to you.”

It’s worth comparing this to other house edges. In American roulette, the green zero and double-zero create the edge structurally — they’re built into the wheel. In baccarat, the edge has to be bolted on after the fact, because the underlying rules don’t generate one on the Banker side. The commission is that bolt-on.

Net Banker EV after the 5% commission

Here’s the actual calculation casinos run. When you win a Banker bet, you receive 0.95 units instead of 1.00 (the house keeps 5%). When you lose, you still lose the full unit. Ties push.

EV(Banker, 5% commission) = (+0.95)(0.4586) + (−1)(0.4462) + (0)(0.0952)

= 0.43567 − 0.4462 = −0.01057, or roughly −1.06%.

And the Player bet, for comparison:

EV(Player) = (+1)(0.4462) + (−1)(0.4586) + (0)(0.0952) = −0.0124, or −1.24%.

So even after the commission, Banker is still the better bet by about 0.18 percentage points. Players who think the commission “punishes” Banker into being worse than Player have it backwards — the commission shrinks the gap, but it doesn’t close it.

The full comparison table

The Tie bet is, to put it kindly, a tax on optimism. Don’t take it. Ever. I’ve watched otherwise sharp players hit Tie repeatedly on hot streaks, and the math is so brutal it isn’t even a debate.

Why 5% is the industry standard (and why 4% is rare)

If 5% gets the house to +1.06% and 4% drops it to +0.60%, you might ask — why not split the difference, or run 4% to attract players? The answer’s a mix of math and operations. A 1.06% edge is comfortable for the casino: high enough to be profitable at scale, low enough that players still feel like they have a fair shot. Drop it to 0.60% and the casino’s hourly take on a $50/hand table roughly halves, which on a busy floor adds up fast.

There’s also a rounding consideration most players never think about. At 5%, the math is simple — a $100 win pays $95, and that’s easy for dealers to chip out fast. At 4%, you’d pay $96 on a $100 win, which is also clean, but for non-round bets ($75, $125, $250) the 4% math gets fiddly. Casinos value speed at the table because more hands per hour means more expected revenue. The 5% number sticks partly because it’s operationally smooth.

Some Asian casinos and a few US rooms do run 4% commission as a promotion or for high-limit play. It’s a real edge for the player — half the house take, basically — and if you ever see it offered, you should prefer it over 5%. The historical and cultural roots of the game are worth a read too; the Britannica entry on baccarat covers how it spread from Italy to France to the global high-roller circuit.

How no-commission variants modify the rules instead

Some tables advertise “no commission” baccarat — sounds great, right? Here’s the catch: the house doesn’t give up the edge, it just hides it. In the most common variant, Banker wins on a three-card total of 6 pay only 1-to-2 instead of 1-to-1. That single rule change pulls the house edge back to around 1.46% on Banker — actually worse than the standard 5% commission version (1.06%).

The lesson, and I’ll be direct here: “no commission” is a marketing phrase, not a math improvement. The casino has run the EV calculation, and they don’t offer rule variants that hurt themselves. If a baccarat table looks player-friendly on the surface, the edge is buried somewhere — usually in a payout reduction on a specific Banker total. Always check what Banker-6 pays before sitting down.

The $100-per-hand, 100-hand example

Abstract percentages don’t really land until you put dollars on them. Say you sit down at a standard 5%-commission table and bet $100 on Banker every hand for 100 hands. What does the math predict?

Expected loss = $100 × 100 × 0.0106 = $106 over 100 hands.

That’s the expected value, not what’ll actually happen on any single session — variance is real and you could be up $800 or down $1,400 on the night. But run that same 100 hands a thousand times and the average will converge tightly on −$106. Now compare to Player at the same stakes: 100 × $100 × 0.0124 = $124 expected loss. A $18 swing across 100 hands isn’t life-changing, but it’s free money on the table if you’ve already decided to play. Pick Banker, every time.

And if you’re playing 200 hands an hour (baccarat moves fast — faster than blackjack), that $106 expected hit compresses into a single hour of play. For more on how house edges compound, the resources at Effortless Math can help you get comfortable with EV calculations across many bets.

Frequently asked questions

Is the 5% commission charged on every bet or only on winning Banker bets?

Only on winning Banker bets. If you bet $100 on Banker and lose, you lose $100 — no commission added. If you win, you receive $95 instead of $100. Some casinos track the commission and collect it at the end of the shoe rather than after each hand, but the math is identical either way.

Should I ever bet Player or Tie?

Player’s edge (−1.24%) is worse than Banker’s (−1.06%), so there’s no math reason to choose it — though it’s close enough that the difference doesn’t ruin your night. Tie is a different story. At −14.36%, it’s one of the worst bets on the casino floor, comparable to the worst sucker bets in craps. Skip it.

Do card-counting strategies work in baccarat like they do in blackjack?

Not really. Researchers have shown that card composition in baccarat does shift the EV slightly as the shoe depletes, but the swings are tiny — you’d need to track the count for hundreds of hands to gain even a fraction of a percent. Casinos also shuffle deeper into the shoe than blackjack tables typically allow. It’s mathematically possible but practically unprofitable.

Why don’t casinos just lower the Banker payout to 0.99 instead of charging commission?

They could — the math would land almost the same place. Tradition, mostly. Baccarat’s commission structure dates back centuries, and high-limit players (who drive a huge share of baccarat revenue) expect the familiar format. There’s also a psychological angle: paying 0.95-to-1 feels cleaner than “you won, but here’s your 99 cents.”

Is online baccarat the same as live baccarat in terms of edge?

Most reputable online baccarat games use the same 8-deck rules and 5% Banker commission, so the edge is identical: −1.06% Banker, −1.24% Player. Where online play diverges is hand speed — you can play 400–600 hands an hour online versus 70–200 live — so your hourly expected loss can be several times higher despite the same per-hand edge. Pace yourself accordingly.

What to actually take from all this

The 5% commission isn’t a casino gimmick or a rip-off. It’s a precision-engineered correction that turns a slightly player-favorable game into a slightly house-favorable one. Once you see the 0.4586 / 0.4462 / 0.0952 split, everything else follows: Banker is structurally the better bet, the commission pulls it just below break-even, and the casino still profits because they get to skim that 1.06% from a fast-paced game played at high volume by people who often haven’t done the math.

If you’re going to play baccarat — and millions of people do, it’s the most popular table game in the world by handle — bet Banker, ignore Tie, look for 4% commission tables if you can find them, and don’t sit down at “no commission” variants without first checking what Banker-6 pays. That’s the entire optimal strategy. There’s nothing more to learn, and anyone selling you a “baccarat system” is selling you nothing. Gambling outcomes are uncertain; no strategy guarantees profit.