The FeiThink Library · Standalone

Starting from the casino in Dragon Quest

3 min

Machine translation from the Chinese original, not reviewed by the author.

A few days ago I was playing Dragon Quest XI S, and the poker game in its casino caught my attention.

The rule is that as long as your 5 cards make any hand, you can use your winnings as the stake for a double-or-nothing challenge: draw 4 cards and compare, beat the face-up card and your stake doubles, lose and it goes to zero, and you can keep challenging.

While playing I had a vague bodily sense, something like:
if you play patiently and do not greedily push your luck, you can steadily run a surplus.
But once you depart from a certain range, hoping to win a few more rounds, the win rate drops sharply, and in the end you lose everything.

So is there a sure-win strategy, a way to win by gambling long?
At first glance this is a typical arithmetic problem in probability. I took it to codex.
I found both places where it matched my feel and places where it differed from a classic probability problem.
This game is not standard fair poker at all: the success rate in the first few rounds is obviously high, then falls sharply.

So the expected-value-maximizing play is: guess 5 times and stop at once.
Then, following this strategy, I did quickly accumulate 100,000 game tokens.

The lesson of this is:

The emotions in the process are utterly unnecessary, purely surplus.

Whether the wild joy of hitting a rare hand,
or the letdown of a huge win wiped to zero in one go.
These emotions are utterly unnecessary.

They are simply one link in the mechanical motion behind this winning formula.

Winning one round is not cause for joy, losing one is not cause for worry.

As long as this mechanical, dull piston stroke keeps going, you accumulate surplus on your own.

Broadly extended,

Many things can be thought through with simple mathematical principles, yet people keep fantasizing that this round is special.

And so they put too much grief and joy, that is, attention, on things that do not matter.

And lose effective attention to what really matters.

And as for what really matters?

Everyone’s answer differs, hard to sum up in one phrase.

But even for the question of finding what matters in life,
mathematical theory gives us a fairly instructive framework.
Should life be about holding what you have, or going out to explore?
The optimal strategy is:

Always stay optimistic about the uncertain, explore the unknown, take risks.

Until the experience you gain from exploring is enough to turn the unknown into the known.

More and more I feel that

philosophy and mathematics really are the two most important disciplines.

Philosophy is responsible for finding you meaning, for defining the objective function for yourself.

Mathematics is responsible for portraying your objective function precisely, and can then help you realize it more effectively.

The question of meaning no longer troubles me.

The task ahead is mainly practice.

And the best way to practice creating is to make money.

As Taleb says, become an entrepreneur.

The lower way to make money is to sell your labor.
The higher way is something like the poker game above: to discover and build a production system.

For me this is a brand new field. A new adventure.

The FeiThink Library · Standalone

Next: In the age of AI, what do I still write?

If this stayed with you, write to me

New volumes arrive here

Email RSS

Starting from the casino in Dragon Quest
The FeiThink Library
First published 2026.08.24

Fei Huang · Writes from Amsterdam on Kant, Dostoevsky, and the moral life.