About the author
Xing Xu
I'm Xing Xu. I studied biochemistry and mathematics at UC San Diego, and I'm an independent iOS developer. I build Baccarator, an iPhone app that works out baccarat odds from the cards left in the shoe, and I write the articles on this site.
Why I built Baccarator
Same scoreboard, opposite conclusions.
Whenever my parents and I played baccarat together, we watched the same scoreboard and came to opposite conclusions. They believed in the long streak. I had my own system of waiting for six in a row and then betting the other side.
Neither of us had actually checked. When I started building a baccarat app, I couldn't keep going on gut feeling, so I counted. I simulated 2,000,000 eight-deck shoes and tested both ideas. Both were wrong in the same way: we were both assuming the shoe remembered the last few hands. It doesn't. The only thing that changes the odds is which cards are still left.
That is what Baccarator shows. After each hand you enter, it recalculates the probability and expected value of Banker, Player, Tie, the pair bets and other side bets from the remaining cards. It is a math tool only. There is no real-money gambling in the app, and it is for ages 21 and over.
Method
How the numbers in these articles are calculated
Exact enumeration
Probabilities like Banker 45.86%, Player 44.62% and Tie 9.52% come from working through every possible deal from the shoe under the standard drawing rules, not from estimates or rounded tables.
Simulation where enumeration isn't practical
Questions about a whole shoe, like how often a six-hand streak appears, come from simulating 2,000,000 eight-deck shoes.
Assumptions are stated
Unless an article says otherwise, numbers are for eight decks, standard third-card rules and a 5% Banker commission. Six-deck figures are given where they differ.
Checked before publishing
Every figure is recalculated before it goes into an article, and the English and Chinese versions use the same numbers.
Scope
What these articles will and won't do
What they explain
- Probabilities, house edge and expected value
- How the remaining cards move those numbers
- Why road patterns and betting systems change none of it
What they won't do
At the start of a fresh shoe, every bet has a negative expected value. With some remaining-card mixes, a side bet like a pair bet, Dragon 7 or Panda 8 can briefly turn positive. Banker and Player rarely move far enough to matter.
A positive expected value is a long-run average. The next hand can still lose, and these spots are rare and short. So the articles explain when and why the numbers move, but they don't tell you what to bet, when to bet or how much.
Read the math, or try the calculator.
Start with the articles, or see the same numbers recalculated for your own shoe in Baccarator.