Royal flush odds — the exact math

Under optimal 9/6 Jacks or Better play, a royal flush hits once every 40390.5 hands — our engine computed this across every possible deal, not from simulation. A dealt pat royal is 1 in 649,740. And droughts are normal: there's a 14% chance of going TWICE the average cycle with nothing. See the drought table ↓
1 in 40390.5hands per royal, optimal play
1 in 649,740dealt a pat royal
~1.98%of total return comes from royals

The drought table

Probability of hitting zero royals in N optimally-played hands:

Hands≈ hours (600/hr)P(no royal)
10,0001778.1%
20,0003360.9%
40,3916736.8%
80,78113513.5%
121,1722025.0%
200,0003330.7%

The 36.8% at one full cycle is the classic e⁻¹ — droughts aren't bad luck, they're the geometry of rare events. Bankroll for the variance, not the average.

Why chasing changes the odds

Royal frequency is a function of strategy. Hold 3-to-a-royal over a high pair everywhere and you'll see more royals — and lose more money, because the EV you burn between them exceeds what the extra royals pay. The trainer keeps you on the line where the 99.5439% lives.

FAQ

What are the odds of hitting a royal flush in video poker?

Playing 9/6 Jacks or Better with optimal strategy, a royal flush arrives once every 40390.5 hands on average — about 67 hours at 600 hands/hour. Being DEALT a pat royal is far rarer: 1 in 649,740.

Can I go 100,000 hands without a royal?

Yes, and it isn't even unlikely: the probability of zero royals in 100,000 optimally-played hands is about 8.4%. Droughts twice the average cycle happen to every regular player.

Do the odds change if I play badly?

Yes — strategy changes royal frequency. Chasing royals too hard raises royal frequency but burns EV elsewhere; ignoring royal draws lowers it. Our numbers assume EV-optimal play, the same line the trainer teaches.

Are these numbers simulated?

No. They come from exact enumeration: every one of the 2,598,960 possible deals, played optimally across all 32 holds, with the resulting category probabilities summed. The probabilities add to 1 to nine decimal places and reproduce the 99.5439% return independently.