Baccarat strategy, tested

We tested every popular idea against three quarters of a million live-dealer hands. Most of what is sold as strategy does not survive the data. Here is what does.

Baccarat strategy advice is mostly folklore: follow the streak, bet against the chop, read the derived roads, wait for the shoe to "settle." We built a collector that logs every hand from dozens of live-dealer tables around the clock, then tested those ideas the way a statistician would: find a pattern on one half of the data, judge it on the other half, and check whether random shoes produce the same "winners."

What the live-dealer data says

• Past hands do not predict the next one. After a Banker, the next hand is Banker 50.7% of the time. After a Player, 50.8%. After three Bankers in a row, 50.1%. After ten in a row, 48.9%. Every conditional we could construct lands within noise of the base rate.
• Streaky shoes do not stay streaky online. A shoe that has already thrown two dragons continues its next run at 50.0%. A shoe that looks consistent by any measure we tried starts its next run at the base rate.
• Derived roads are decoration. Big Eye Boy, Small Road and Cockroach Pig, followed or faded, land between 49.8% and 50.8% on 13,000 to 15,000 held-out bets each.
• Position in the shoe, hour of the day, table, and the previous shoe on the same table carry no information.
• Card counting on the main bets is real but worthless: with every card known, Banker is a positive bet on about 1 hand in 40,000, by a fraction of a percent.

What that means for how you play

If the next hand cannot be read from the last ones, the things that decide whether you leave a session ahead are the size of your bets, when you stop, and whether you are still following your rules at hand sixty. That is what Rack Chasers is built to hold constant: a call on every hand under a fixed rule set, units sized to your buy-in, and a lock at your target. You bring the read of the table; the system keeps you honest.

Why the venue matters

Not every shoe is dealt the same way. Casino-floor shoes in our dataset show streak structure that studio-shuffled live-dealer shoes do not, and that difference is measurable. We wrote that comparison up separately.

Questions

Does any betting system beat baccarat?

No system that sizes bets from past results changes the expected value. Martingale, Fibonacci, Labouchere, 1-3-2-6 and Paroli only change variance and, for negative progressions, raise the risk of ruin sharply.

Is live-dealer baccarat rigged?

Not in our data. The outcome streams from Pragmatic, Evolution and Ezugi pass every randomness test we run, including one that catches flawed shuffle machines. That is bad news for pattern players and good news for fairness.

What is the break-even hit rate in baccarat?

A Banker bet pays 0.95 to 1, so it needs to win 51.28% of resolved bets to break even. A Player bet pays 1 to 1 and needs 50.00%.

Figures on this page are replays and measurements of logged hands under the current rule set as of 2026-09-23. They are not a forecast and do not guarantee future results. Decision support only; every bet is your own decision. 21+ or your local gambling age.