---
name: fengliu-weak-system
lens: Weak-System Contrarian
attributed_to: Feng Liu (冯柳)
style: contrarian
markets: A-share primarily; adaptable to HK where the same names trade with a discount
description: Assume you have no information advantage, buy where the bad news is already known and priced, and let the position structure rather than the forecast carry the risk control.
---

# Weak-System Contrarian Lens

> **Framework, not investment advice.** This is our own summary of a publicly
> documented methodology, reconstructed as a procedure. It is not written or
> endorsed by the person named, and any verdict it produces holds only under this
> framework.

## What this lens optimizes for

Making money without an information edge. It starts from an explicit admission of
weakness — you will never know the company better than insiders, sell-side
specialists or the supply chain — and builds a procedure that does not require
knowing more. Its edge is supposed to come from *position and expectation*, not from
superior facts.

## Market-fit note

- Designed around A-share market structure: high retail participation, fast and
  violent sentiment swings, wide dispersion between price and expectation, and
  liquid mid-caps that can be accumulated on weakness.
- Translates to HK for dual-listed and China-operating names, where the discount and
  the sentiment cycle are often deeper. Translates poorly to markets where
  institutional ownership dominates and expectations reset gradually.
- Requires a stock that can be *bought* on the way down — enough liquidity and no
  hard limit-down mechanics blocking accumulation.

## Priority signals (walk in this order)

1. **Declare the weakness explicitly.** Write down what you do not know and cannot
   find out. If the thesis requires knowing one of those things, stop — that is a
   strong-side thesis wearing this lens's clothes.
2. **Is the bad news known and old?** The candidate must be down substantially on
   *publicly discussed, well-understood* negatives. A stock falling on something the
   market has not yet digested is not in scope: this framework wants the news
   exhausted, not fresh.
3. **Expectation level, measured separately from valuation.** What is the consensus
   assuming — analyst revision direction, position crowding, retail sentiment,
   media tone? The target is low expectation, which is not the same as a low
   multiple. A cheap stock with high hopes fails this test.
4. **Business quality floor.** High gross margin, a product with real demand, a
   balance sheet that survives the bad scenario. The framework buys pessimism about
   *good* businesses, not cheapness in fragile ones — this is the most frequently
   dropped condition and dropping it converts the lens into a falling-knife machine.
5. **Locate the source of the decline.** Cyclical, sentiment-driven, one-off
   operational, or structural. Structural decline is out of scope; the other three
   are the lens's hunting ground. State which one and why.
6. **Asymmetry of the remaining path.** With the bad news priced, what does normal
   (not good) look like? The framework wants cases where merely *not getting worse*
   produces a return, and where further deterioration is bounded by asset value,
   cash or dividend.
7. **Position structure is the risk control.** Because the framework denies itself
   an information edge, it cannot rely on conviction. Build in tranches with
   pre-committed levels, cap single-name weight against the scenario where the thesis
   is simply wrong, and decide the sizing before the position exists.
8. **Reverse test.** State what the seller believes. If the seller's view is coherent
   and you have no basis to reject it, the correct action is a smaller position, not a
   better argument.

## Disqualifiers (hard vetoes)

- **The decline is structural** — end-market substitution, permanent regulatory
  contraction, technology displacement. This lens has no defence against a shrinking
  denominator.
- **Unresolved integrity or fraud risk.** Weak-side reasoning cannot price fraud;
  "priced in" is meaningless when the accounts may be fictional. Stack a forensic
  lens; if it flags, decline.
- **Balance sheet cannot survive the bad case** — refinancing risk, heavy
  controlling-shareholder pledges, covenant pressure.
- **The bad news is fresh**, still developing, or has an unknown magnitude.
- **Expectations are still high** despite the price fall (analyst estimates not yet
  cut, retail enthusiasm intact).
- **The thesis requires an information advantage** the framework has already
  declared it does not have.

## Scoring rubric

| Gate | Pass condition |
| --- | --- |
| Weakness declared | Written list of unknowables; thesis independent of them |
| News exhausted | Negative is public, old, and its magnitude is bounded |
| Expectation low | Revisions cut, sentiment negative, crowding gone |
| Quality floor | Gross margin, real demand, survivable balance sheet |
| Decline source | Cyclical / sentiment / one-off — named |
| Asymmetry | "Merely stops getting worse" produces a return |
| Structure | Tranches, cap and stop condition pre-committed |

## Typical misuse

- **Dropping the quality floor.** Applying weak-side contrarianism to any beaten-down
  stock, which is the single most expensive error available under this lens.
- **Buying fresh bad news.** Confusing "it fell a lot" with "the news is exhausted".
  Falls with unresolved magnitude are momentum, not contrarianism.
- **Running it in a momentum regime.** In a sustained trending market the framework
  systematically buys early and holds through further decline; that is expected
  behaviour, and it must be sized for it, not rationalized away.
- **Using conviction language.** Any sentence beginning "I'm confident that…"
  violates the framework's own premise. The framework's risk control is structure,
  and conviction language signals that structure has been abandoned.
- **Ignoring the reverse test.** Failing to state the seller's coherent case, which
  is the only cheap check available to a self-declared weak participant.
- **Transplanting to institution-dominated markets** where expectations do not
  overshoot as far or reset as fast.

## Falsifiers

- Analyst estimates are still being cut a quarter later and the operational
  deterioration accelerates → the "news exhausted" gate was wrong.
- Gross margin breaks its historical floor → the quality-floor gate fails and the
  position thesis is void.
- The decline turns out to be structural (a substitute takes durable share) → out of
  scope; exit is required by the framework, not optional.

## Output contract

Report under the skill-level output contract. State the verdict as "under a
weak-system contrarian framework, …", begin with the declared list of unknowables,
name the decline source, and state the tranche structure and single-name cap before
any price target. This is analysis, not investment advice.