Practice and free tables, the desk to learn at.

The desk recommends twenty practice deals per new habit. The number is short enough to fit in an evening; long enough to expose the rough edges of a habit.

Twenty deals

Twenty deals is the working default. A new habit tested across twenty deals reads in the scorecard: did the habit lower the loser score, raise the winner count, or change the read of the table?

The scorecard audit

After twenty deals, audit the scorecard. Look for the changes that the habit was supposed to produce. If the habit did not produce the change, the habit is not yet learned; test for another twenty deals. If the habit produced the change, move to a paid table.

Habits transfer from free to paid. The engine is the same; the joker rule is the same; the drop costs are the same. The only difference is the stake, and the stake sharpens the read rather than changing the habits.

The stake pressure effect

Stake pressure makes the reader pick faster, drop earlier, and read more carefully. A reader who has built the habits on a free table sees the stake pressure amplify the habits rather than disrupt them.

The discipline of free

A reader who plays on free tables only and never moves to paid has built the habits without the stake pressure. The free table is the desk to learn at; the paid table is the desk to play at. Both are useful; the discipline is to use each.

Most platforms offer practice mode against an AI opponent at three difficulty tiers: beginner, intermediate, and advanced. The tiers differ in how the AI evaluates discards, how aggressively it builds its sequences, and how it reacts to a joker appearing on the open pile. A reader who has just learned the rules should start at beginner; a reader who has a few real-money sessions behind them should jump to intermediate; a reader who wants to stress-test a strategy should run advanced.

The beginner AI plays a passive game. It draws, it discards, and it shows when it has a valid show. It does not pressure the table with a joker read and does not push a show on a marginal hand. The beginner tier is useful for the reader who is learning the show mechanic.

The intermediate AI plays a normal game. It reads discards, it watches the open pile, and it takes a first-drop on a weak hand. The intermediate tier is useful for the reader who knows the rules and wants to build a habit.

The advanced AI plays an aggressive game. It reads discards against the player's hand, it punishes a careless joker play, and it pushes a show on a marginal hand when the player is unlikely to challenge. The advanced tier is useful for the reader who wants to stress-test a strategy before taking it to a real-money table.

What practice mode does not teach

Practice mode does not teach the social layer of a real-money table. There is no chat, no banter, no read on a player who tilts after a bad deal. The reader's real-money session will carry the social layer; the practice session will not. The desk's recommendation is to spend enough time in practice to build the habit, then move to a low-stakes real-money table and learn the social layer there.

Practice chips and real chips share the format, the joker mechanic and the drop rule. They diverge on the stake pressure, the variance tolerance and the social layer. The practice session teaches the format; the real-money session tests the reader's tolerance for variance.

The stake pressure is the largest single difference. A reader who loses a practice hand has lost nothing; a reader who loses a real-money hand has lost a rupee amount that the reader's bankroll must absorb. The stake pressure is what makes the social layer matter; it is also what makes the responsible-play controls matter.

When to leave practice mode

The desk's recommendation is to leave practice mode when the reader can play through a fifty-hand session at the intermediate tier with a positive expected value. The fifty-hand window is enough to smooth out the format-level variance and to expose the habit-level mistakes. The reader's first real-money session should be at the lowest stake the lobby offers.

Returning to practice mode

The reader who is iterating on a strategy should return to practice mode between real-money sessions. The practice session lets the reader test a change without risking a stake; the real-money session validates the change under stake pressure. The desk recommends one practice session per strategy change and a real-money session per validation.

Most platforms offer practice mode free of charge, with practice chips that the platform issues at signup and replenishes if the reader loses them all. The free practice model is the standard; the desk's recommendation is to start there.

A few platforms offer a paid practice model. The paid practice uses real money stakes against the AI opponent. The model is useful for a reader who wants to test a strategy under stake pressure without sitting at a real-money table with other readers. The desk's view is that the paid practice model is unnecessary for most readers; the free practice model plus a low-stakes real-money session is the better learning path.

What the paid practice does not teach

The paid practice does not teach the social layer. The reader plays against the AI, not against other readers. The social layer carries its own lessons; the AI does not tilt, the AI does not chat, and the AI does not bring a session-long read to the table. The reader who wants the social layer must move to a real-money table.

Returning to free practice

A reader who has run a few paid practice sessions can return to free practice to iterate on a strategy. The free practice session is the right surface for the iteration; the paid practice session is the right surface for the validation. The desk recommends one free practice session per iteration and a paid practice session only for a stress test.

The reader who wants to test a strategy in practice mode can follow a four-step workflow. The workflow isolates the strategy, tests it under controlled conditions, validates it against a real-money session and writes the result into the reader's notes.

Step one is to isolate the strategy. Pick one change (a different first-drop threshold, a different joker play, or a different discard priority) and run it through every hand in the practice session. Do not mix changes; the workflow tests one change at a time.

Step two is to run the practice session at the intermediate AI tier. The intermediate tier is the most realistic; the beginner tier is too passive, the advanced tier is too aggressive. A fifty-hand session at the intermediate tier is enough to smooth out the format variance.

Step three is to validate the change with a low-stakes real-money session. The practice session isolated the change; the real-money session validates it under stake pressure. A reader who has only the practice session in evidence has not yet tested the change under the social layer.

Step four is to write the result into the reader's notes. The notes carry the change, the practice session result, the real-money session result, and the reader's view of whether to keep the change. The notes are the reader's strategy log; over time the notes become the reader's playbook.

The reader who is opening the practice mode for the first time should run a short session. The desk's recommendation is twenty hands at the beginner AI tier, followed by ten hands at the intermediate AI tier. The session takes about thirty minutes; the session's purpose is to build the habit, not to test a strategy.

Step one is to sit at the beginner table. The reader plays twenty hands without recording anything. The first session is for the rhythm of the table, not for analysis.

Step two is to switch to the intermediate table for ten hands. The reader still does not record anything. The intermediate table shows the reader what a normal session looks like; the reader's job is to read the pace.

Step three is to leave the practice mode and to write a short note. The note captures the reader's first impression. The reader will revisit the note at the third practice session, when the habit is starting to form.

What to do with the first-session note

The first-session note is the foundation of the reader's strategy log. The reader should not over-edit the note; the note's purpose is to record the reader's first impression, not to build a polished analysis. The next session will refine the note; the session after that will refine it further.