They give you a stall, and twelve months.
The most honest thing about business is this: every month, it tells you once that you were wrong.
Feedback learning + risk adjustment (a reinforcement-learning lens). Unlike a one-off choice, it has twelve consecutive feedback loops—whether you can adjust your strategy based on last round's result is the most direct evidence of the "growth" dimension. How fast you come back from a mistake is the real ability.
What we watch is how fast you come back from a mistake, not the final number
Twelve months, one real settlement each month. A loss is immediately visible, with no "fix it later" cushion—this is the biggest difference from Shunxi: high-frequency feedback, set against low-frequency upheaval.
②Game-Theory Core
Feedback learning + risk adjustment (a reinforcement-learning lens). Unlike a one-off choice, it has twelve consecutive feedback loops—whether you can adjust your strategy based on last round's result is the most direct evidence of the "growth" dimension.
Literature anchor: reinforcement learning / behavioral finance—how fast you come back from a mistake is the real ability.
③Players / Duration / Venue
| Item | Spec |
|---|---|
| Players | 2–5 players (multiple tables can run in parallel) |
| Duration | 40–70 minutes |
| Venue | Indoor + host station |
⑦Output
Individual "recovery slope" + "number of decision adjustments" + risk-preference curve.
⑧Consequence Mechanics
Every month settles for real—a mistake is immediately visible, with no "fix it later" cushion. This is the biggest difference from Shunxi (high-frequency feedback vs low-frequency upheaval).
⑨Three Ending Tiers
Mirror
Took a loss in the first 3 months, then clearly changed the approach; recovery slope >1.
"You didn't avoid mistakes—you came back from them faster. That is the real ability."
Iron Armor
No big loss and no big gain throughout (never wrong, never learning).
"Stability has a cost: you lost no money, and you grew no ability."
Unseen
Doubled the bet after a loss.
"You aren't running a business, you're chasing the loss; loss-chasing is the most expensive emotion."
⑩Three Insights
The frequency of feedback determines the speed of learning—this is also why a "weekly review" is more useful than an "annual summary."
Who still dares to move after a loss, and who starts getting cocky after a profit—one game shows it all.
Roles given high-frequency feedback grow fastest; give only an annual target, and the only thing you get back is loss-chasing.
⑪Deployment Checklist
Market · Twelve Stalls
2–5 players · multiple tables can run in parallel · 40–70 min · indoor + host station · twelve months
④Rules & Steps
- Each player runs one stall of the same spec, with the same starting capital and a shared market.
- Each month, choose two of five actions: Stock / Pricing / Credit / Expand / Collect.
- The market gives a demand fluctuation each month (card draw or dice), and good and bad months can both appear.
- After twelve months, settle net worth—but what we watch is "how fast you come back from a mistake," not the final number.
⑤In-Game Currency & Payoff Table
| Item | Rule |
|---|---|
| In-game currency | Cash / inventory / receivables / net worth |
| Key metric | Not the highest, but the recovery slope after a drawdown |
⑥Observation Points & Tags Table
| id | Behavior description | Weight |
|---|---|---|
shrink | Pulls back the moment there's a loss, staying conservative for months after | Courage −2, Wisdom +1 |
double | After a loss, instead doubles down to win it back in one go | Courage +2, Wisdom −2 |
steady | Even through consecutive good months, doesn't increase the position | Wisdom +2, Growth +1 |
credit | Sells on credit often, and accepts it when it can't be collected | Connection +2, Courage +1, Wisdom −1 |
nocredit | Never sells on credit, always cash on the spot | Responsibility +1, Courage −1 |
expand | Expands the stall the moment there's cash flow | Growth +2, Courage +1 |
learn | Took a loss in the first three months, then clearly changed the approach | Growth +3, Wisdom +1 |
solo | Doesn't trade with anyone the whole game | Responsibility +1, Connection −2 |
HostHost Script
- Opening: each player gets one stall of the same spec, the same starting capital, and a shared market; explain the rules only, and never preview the market.
- Monthly action choice: choose two of Stock / Pricing / Credit / Expand / Collect; the host only collects cards and does not comment.
- Monthly settlement: real settlement, numbers produced on the spot, with no "fix it later" cushion.
- Demand fluctuation: decided by card draw or dice, good and bad months can both appear, and the host never changes the number.
- Records: who pulls back after a loss, who doubles down, and who stays steady even through a run of good months—record the behavior tag month by month.
- Closing: first the recovery slope and the number of decision adjustments, then the risk-preference curve.
⑦Three Report Scopes
| Report | Delivered to | Content | Boundary |
|---|---|---|---|
| Personal report | The person | Five dimensions / 16 types / six-dimension decision radar / blind-spot list / ending tier / insight | Visible to the person only |
| Team report | Team / HR | Team personality map + complementary network + cooperation decay curve + three metrics | Aggregated results only, no naming |
| Leader report | Decision-makers | Three charts + five metrics + structural and risk conclusions | No naming, no individual five-dimension raw values |
Game-specific output: individual "recovery slope" + "number of decision adjustments" + risk-preference curve.