Growth simulation: the same player, with and without the rules
Version: 1 · 2026-10-02 · a Monte Carlo of one year of play, 20,000 runs per policy; code in spec/figures/sim/sim.ts
The claim the landing page makes is narrow and testable: for a player who already wins, the discipline rules change the spread of outcomes far more than the average, and they remove the outcomes that end a poker journey. This page shows the simulation behind that claim and every assumption in it. Nothing here promises a return; a losing player loses under either policy.
Assumptions
| Input | Value | Why |
|---|---|---|
| Starting roll | $20,000 | 40 buy-ins at 2/5, the standard |
| Stakes | 1/2, 2/5, 5/10, 10/20 USD, 100 bb buy-ins | The usual live no-limit ladder |
| Win rate by stake | 8, 6, 3, 0 bb/h | Games get tougher up the ladder: a solid winner at 1/2 is a small winner at 5/10 and break-even at 10/20 |
| Standard deviation | 60 bb/h | Common live no-limit hold'em figure |
| Volume | 4 sessions of 5 h a week for 52 weeks | A serious part-time schedule |
| Disciplined policy | Plays the highest stake with 40 buy-ins, moves down under 30, moves up at 50 of the next stake, stop-loss of 3 buy-ins per session | The default rule set, ADR-0003 |
| Undisciplined policy | Plays the biggest game 15 buy-ins allow, drops only under 5, no stop-loss, chases two more buy-ins half the time after a 3 buy-in loss, and plays the next session 6 bb/h worse after any losing session | What the owner's circle actually does on a bad night |
Results
| With the rules | Without | |
|---|---|---|
| Median bankroll after a year | $49k | $16k |
| Tenth to ninetieth percentile | $38k to $67k | $9k to $30k |
| Runs that ended below the starting roll | 0% | 61% |
| Runs that lost half the roll at some point | 0% | 92% |
| Median worst drawdown | 16% | 66% |
| Runs that went broke | 0% | 0% |
Two things the table says. The disciplined player grows at the rate their edge allows, around two and a half times in the year on these assumptions, and never spends a month wondering whether the roll survives. The undisciplined player, with the same cards and the same skill, ends the year behind more often than not, because the games they jump into are ones they do not beat, and the tilt after each loss costs edge they cannot afford.
What the simulation does not say
- It does not say any player will grow their roll. The edge is an input, and most players do not have one.
- It does not model rake, tips, travel or living costs drawn from the roll.
- The undisciplined policy is one plausible bad habit set, not the worst case. Real players also play tired, drunk and at the wrong table.
- Rerun it with your own numbers:
bun spec/figures/sim/sim.ts, thenbun spec/figures/sim/chart.ts.