Dragon Tiger Simulator
Play with pretend chips, or run thousands of rounds and watch the math take over.
Dragon Tiger Simulator
This simulator deals Dragon Tiger from a real shoe, shuffled with your browser's cryptographic random source. Use Play to bet pretend chips round by round, or Classroom to bet one unit for up to 100,000 rounds and watch the average result settle onto the house edge. No money, no sign-up, and nothing is stored.
A simulator shows how the game behaves. It cannot predict it.
Each round is two cards from a shuffled shoe. Past results in the shoe do not tell you what comes next, and no betting pattern changes the house edge. The exact figures behind every bet are on our Probability Calculator.
Pretend chips: 1,000
Dragon
Tiger
Place a bet and press Deal. Bets stay in place between rounds.
This shoe so far (newest last)
8 decks, 416 cards left; the shoe is reshuffled when 52 remain.
What the Classroom Mode Should Show
With the standard rules (eight decks, Tie at 11 to 1, half the stake lost on a tie), a one-unit bet every round has these exact expected results. The last column is how far a 10,000-round run typically strays: about 95% of runs land inside it.
| Bet | Expected result per unit | 95% of 10,000-round runs within |
|---|---|---|
| Dragon or Tiger (half lost on a tie) | -3.73% | ±1.94% |
| Tie at 11 to 1 | -10.36% | ±6.31% |
| Suited Tie at 50 to 1 | -13.98% | ±13.14% |
The Tie and Suited Tie bets pay rarely and pay big, so their runs wander far more widely before settling. That spread, not the center, is what short sessions feel like. Why no staking system moves the center is on Winning Strategies.
How the Simulator Works
- The shoe: four, six or eight standard 52-card decks, shuffled with an unbiased Fisher-Yates shuffle driven by
crypto.getRandomValues. - The deal: one card to Dragon, then one to Tiger. Aces are low and kings high; suits do not rank. The shoe is reshuffled when 52 cards remain.
- Settlement: Dragon and Tiger pay even money and lose half (or all) of the stake on a tie, depending on the rule you choose. Tie pays 8, 10 or 11 to 1; Suited Tie pays 50 to 1 when both cards are identical in rank and suit.
- Checked against the exact math: 2,000,000 and then 20,000,000 rounds under three rule sets, with every frequency and average within normal random error of the figures on our Odds & Payouts page.
Playing baccarat too? Our sister site has a Baccarat Simulator built the same way.
Frequently Asked Questions
Is this Dragon Tiger simulator random?
Yes. It shuffles a full shoe using your browser's cryptographic random number source, the same source used for encryption keys, and deals two cards per round until 52 cards remain, then reshuffles. Nothing is weighted or scripted.
Can the simulator help me win at Dragon Tiger?
No. It shows how the game behaves, not what comes next. Over many rounds every bet drifts toward its house edge: 3.73% on Dragon or Tiger with eight decks and half lost on a tie, 10.36% on a Tie at 11 to 1, and 13.98% on a Suited Tie at 50 to 1.
Why does a short run sometimes show a profit?
Short runs are noisy. After 100 one-unit Dragon bets the average can easily sit several points above or below the expected -3.73%. The classroom chart shows how that spread narrows as rounds pile up, while the center stays at the house edge.
How was the simulator checked?
We ran the same code for 2,000,000 and then 20,000,000 rounds under three rule sets (eight decks at 11 to 1 with half lost on a tie, six decks at 10 to 1, and four decks at 8 to 1 with all lost on a tie). Every frequency and average result landed within normal random error of the exact figures.
Gambling involves risk of loss. If it stops being fun, see our Responsible Gaming page.