Ginlix AI
50% OFF

How Do Fund Managers Build a 'Barbell' Investment System That Survives Bull and Bear Markets? — From Sunac's Default to Risk Control Evolution Toward Sharpe Ratio >1.0

#investment_strategy #risk_control #barbell_strategy #fund_management #sharpe_ratio #quantitative_investing
Neutral
A-Share
December 28, 2025

Unlock More Features

Login to access AI-powered analysis, deep research reports and more advanced features

How Do Fund Managers Build a 'Barbell' Investment System That Survives Bull and Bear Markets? — From Sunac's Default to Risk Control Evolution Toward Sharpe Ratio >1.0

About us: Ginlix AI is the AI Investment Copilot powered by real data, bridging advanced AI with professional financial databases to provide verifiable, truth-based answers. Please use the chat box below to ask any financial question.

Related Stocks

01918
--
01918
--
How Do Fund Managers Build a ‘Barbell’ Investment System That Survives Bull and Bear Markets?
— From Sunac’s Default to Risk Control Evolution Toward Sharpe Ratio >1.0

Based on search results and professional financial analysis frameworks [1,2,3], I will systematically analyze the construction logic and practical application of the ‘Barbell’ investment system for you.


1. Profound Lessons from Sunac’s Default Incident

The case of Sunac China (01918.HK) is a typical example of risk exposure in China’s real estate industry in the 2020s. According to the latest data, Sunac experienced:

Risk Exposure Path:

  • 2021
    : Cash flow crisis began to emerge
  • 2022
    : Public debt default, officially falling into liquidity distress
  • November 2023
    : First offshore debt restructuring (billion-dollar scale)
  • December 2025
    : Completed full domestic and offshore debt restructuring, reducing debt by approximately 60 billion yuan [2,3]

Core Reasons for Default Analysis:

  1. Industry Misjudgment
    : Overly bet on the upward real estate cycle, ignoring policy shift signals (“Three Red Lines” policy)
  2. Concentration Risk
    : Overweighting a single industry, lacking diversified allocation
  3. Missing Liquidity Management
    : High-leverage model quickly failed during credit contraction
  4. Insufficient Fundamental Research
    : Inadequate understanding of the enterprise’s true debt structure and cash flow status

Lesson Summary
: Investment cannot rely solely on macro judgments or industry beta; it must establish a
bottom-up risk control system
and
multi-dimensional cognitive framework
.


2. What Is the ‘Barbell’ Investment Strategy?
Strategy Definition and Structure

The ‘Barbell Strategy’ is a classic portfolio management method whose core idea is

allocating assets to two extremes, avoiding the middle ground
:

Configuration Structure Schematic:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
High Odds Offensive End              Middle Gap              High Certainty Defensive End
(15%-25%)                                   (75%-85%)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
• Option Buyer           No allocation to mediocre assets        • High Dividend Stocks
• Deep Contango Index Futures                             • Graded A Merger Redemption
• Event Arbitrage                                  • IC Contango Arbitrage
• Small-Cap High-Growth Stocks                              • Cash Equivalents
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Mathematical Logic

Assume portfolio return is R and risk is σ:

Barbell Portfolio Characteristics:

  • Return Side
    : E[R] = w₁·R₁ + w₂·R₂ (weighted return of both ends)
  • Risk Side
    : σ² = w₁²·σ₁² + w₂²·σ₂² + 2w₁w₂ρ₁₂σ₁σ₂

The key here is controlling overall volatility through

low correlation (ρ₁₂ close to 0)
.

Advantages:

  1. Convex Return
    : One end provides extreme return potential, the other provides a safety cushion
  2. Nonlinear Hedging
    : Avoid systemic risk of linearly correlated assets
  3. Liquidity Management
    : Defensive end provides liquidity to handle extreme situations
Practical Application Cases

According to practical cases shown in search results [1]:

  • Offensive End (15%)
    : Allocate deep out-of-the-money options, loss locked at 3%, profit potential over 300%
  • Defensive End (85%)
    : High-certainty cash cows (Graded A, index futures contango arbitrage)
  • Pulse Risk Control
    : Three-dimensional protection (stop-loss line, position management, stress testing)

3. Cognitive Leap from ‘Stock Picker’ to ‘System Guardian’

The fund manager mentioned by the user experienced an important evolution in investment philosophy, reflected in the redistribution of cognitive weights:

Cognitive Weight System
Dimension Weight for Traditional Stock Pickers Weight for System Guardians Core Competency Requirements
Stock Selection Ability
70%
10%
Fundamental research, valuation modeling
Capital Management
20%
40%
Position control, rebalancing, liquidity management
Psychological Control
10%
50%
Emotion management, cognitive bias correction, decision discipline
Core Evolution Logic

Phase 1: Stock Picker Mindset (Before Sunac’s Default)

  • Pursue Alpha returns, rely on individual stock selection ability
  • Ignore systemic risk exposure
  • Overconfidence leads to concentrated positions

Phase 2: System Guardian Mindset (After Risk Control Evolution)

  • Treat investment as a probability game, not a prediction game
  • Establish an ‘antifragile’ mechanism (profit from black swans)
  • Limit human weaknesses through quantitative rules

4. Essence of Quantitative Investment: Delegation of Decision-Making Power
What Is ‘Delegation of Decision-Making Power’?

In traditional active investment, investors have 100% decision-making power, but this is precisely the biggest source of risk:

Decision-Making Delegation Model:
┌─────────────────────────────────────────────────┐
│        Full Investment Decision-Making Process   │
├─────────────┬───────────────────────────────────┤
│   Human Decision    │          Systematic Decision               │
├─────────────┼───────────────────────────────────┤
│ ✗ Emotional     │ ✓ Rule-Driven                        │
│ ✗ Inconsistent     │ ✓ Repeatable                          │
│ ✗ Hard to Replay   │ ✓ Traceable                          │
│ ✗ Affected by Cognitive Bias │ ✓ Statistical Advantage                        │
│   Influence       │                                   │
└─────────────┴───────────────────────────────────┘

Decision-Making Delegation:
- 30% Quantitative Strategy (Automatic Execution)
- 40% Risk Control Rules (Mandatory Constraints)
- 30% Human Judgment (Only Used Outside the System)
Practical: Decision Hierarchy Mechanism
  1. Level 1 - Hard Rules
    (Automatic Execution):

    • Auto-reduce position when stop-loss line is triggered
    • Single asset weight cap (e.g., 10%)
    • Reduce leverage when portfolio volatility exceeds limit
  2. Level 2 - Quantitative Signals
    (Semi-Automatic):

    • Technical indicator buy/sell signals
    • Volatility threshold warning
    • Correlation monitoring
  3. Level 3 - Human Judgment
    (Restricted Permissions):

    • Only used in extreme situations not covered by the system
    • Requires double confirmation mechanism
    • Decisions need to be recorded and replayed

5. Path to Achieving Sharpe Ratio >1.0
Core Formula of Sharpe Ratio
Sharpe Ratio = (Rp - Rf) / σp

Where:
- Rp = Portfolio Return
- Rf = Risk-Free Rate
- σp = Portfolio Volatility (Standard Deviation)

Goal: Sharpe >1.0 means for every unit of risk taken,
you get >1 unit of excess return
Case Analysis: How to Reach Sharpe 1.0+

According to search results, the fund manager achieved the following performance indicators:

  • Sharpe Ratio
    : >1.0
  • Maximum Drawdown
    : -13.3%

Achievement Path Analysis:

1. Return Side Optimization (Rp Improvement):

  • Barbell strategy offensive end: Small positions in high-odds targets boost overall returns
  • Defensive end cash flow: High-dividend, arbitrage strategies provide stable returns
  • Dynamic adjustment: Adjust weights of both ends based on market conditions

2. Risk Side Control (σp Reduction):

  • Negative correlation allocation: Low correlation between offensive and defensive ends
  • Volatility management: Control overall portfolio volatility within 15%
  • Tail risk hedging: Option protection to prevent extreme declines

3. Return-Risk Ratio Optimization:

  • After adopting the barbell strategy in 2025:
    • Obtain stable returns from the defensive end (85%) (assuming 8%)
    • Obtain excess returns from the offensive end (15%) (assuming 20%)
    • Portfolio return: 0.85×8% +0.15×20% =9.8%
    • Assume risk-free rate is 3%, portfolio volatility is6.8%
    • Sharpe = (9.8% -3%)/6.8% =1.0 ✓

6. In-Depth Analysis of the ‘10-40-50’ Cognitive System
1. Stock Selection Ability (10%): From Art to Science

Traditional Misunderstanding
: Spend 80% of energy on individual stock research
Evolution Direction
: Establish a reusable stock selection framework

Systematic Method
:

  • Fundamental filter: ROE, free cash flow, moat assessment
  • Valuation model: DCF, relative valuation, PEG
  • Risk exclusion mechanism: Debt ratio, cash flow断裂 risk, corporate governance
2. Capital Management (40%): Mathematical Foundation of Risk Control

Core Formula: Kelly Criterion

f* = (bp - q)/b

Where:
f* = Optimal position ratio
b = Odds (profit/loss ratio)
p = Win rate
q = Loss rate (1-p)

Practical Application
:

  • If a strategy has a 60% win rate and 2:1 odds
  • f* = (2×0.6 -0.4)/2 =40%
  • Actual Application
    : Use 0.5×f* =20% position considering estimation error

Rebalancing Mechanism
:

  • Time trigger: Adjust weights quarterly
  • Amplitude trigger: Adjust when weight deviates from target by ±5%
  • Volatility trigger: Reduce position when portfolio volatility exceeds limit

###3. Psychological Control (50%): The Hardest Part of Investment

Cognitive Bias List
:

Bias Type Performance Countermeasure Mechanism
Confirmation Bias
Only look at information supporting one’s own views Pre-set falsification tests
Loss Aversion
Unwilling to stop loss when losing Mandatory stop-loss rules
Anchoring Effect
Stick to purchase price Focus on relative value
Overconfidence
Overestimate own prediction ability Decision recording and replay
Herd Effect
Blindly follow the crowd Independent thinking framework

Psychological Training Methods
:

  1. Decision Log
    : Record logic, emotion, and result of each transaction
  2. Regular Replay
    : Review decision quality monthly, not just results
  3. Stress Testing
    : Assume response plans for extreme scenarios
  4. Third-Party Perspective
    : How would you operate if it were someone else’s account?

##7. Investment Competitiveness in the AI Era: Understanding Business Essence

Core Abilities AI Cannot Replace

In the AI era, the unique value of fund managers lies in:

1. Deep Business Understanding

  • AI can process data, but it is difficult to understand the essence of business models
  • Case: Sunac’s risk lies in the fragility of the ‘high leverage + pre-sale system’ model when policies shift
  • This requires comprehensive judgment of business logic, policy environment, and industry cycle

2. Unstructured Information Interpretation

  • Management interviews, grassroots research, supply chain information
  • These require human intuition and experience accumulation

3. Extreme Situation Response

  • AI is trained on historical data and may fail in new crises
  • Human crisis handling experience is crucial in extreme situations
Collaboration Model Between AI and Humans
┌─────────────────────────────────────────┐
│        AI-Enhanced Decision Framework   │
├─────────────────┬───────────────────────┤
│   AI Excels At        │    Humans Excel At            │
├─────────────────┼───────────────────────┤
│ ✓ Data Processing      │ ✗ Business Essence Understanding         │
│ ✓ Pattern Recognition      │ ✗ Logical Reasoning             │
│ ✓ Execution Discipline      │ ✗ Creative Thinking           │
│ ✓ Emotional Neutrality      │ ✗ Value Judgment             │
└─────────────────┴───────────────────────┘

Best Practice:
- AI is responsible for: Data screening, signal generation, risk monitoring
- Humans are responsible for: Strategy design, exception handling, value assessment

##8. Discounted Free Cash Flow (DCF) and Investment Essence

Core Philosophy of DCF

DCF Model
:

Enterprise Value = Σ [FCFt/(1+WACC)^t]

Where:
FCF = Free Cash Flow = Operating Cash Flow - Capital Expenditure
WACC = Weighted Average Cost of Capital
t = Time Period

Why Is DCF the Essence of Investment?

  1. Cash Flow Is King
    : Profits can be manipulated, but cash flow cannot
  2. Long-Termism
    : DCF focuses on long-term value creation ability
  3. Margin of Safety
    : Conservative DCF assumptions provide downside protection
Sunac Case From DCF Perspective

Lessons From DCF Failure for Sunac
:

  • Traditional DCF model assumptions:

    • ✓ Going concern assumption
    • ✓ Capital availability
    • ✓ Market efficiency
  • Actual Situation of Sunac:

    • ✗ Cash flow断裂 (pre-sale fund supervision)
    • ✗ Financing channels closed
    • ✗ Equity dilution due to debt restructuring

Lesson
: DCF needs to combine
scenario analysis
and
stress testing
to consider value realization paths under extreme situations


##9. Building a Complete Risk Control System That Survives Bull and Bear Markets

Three-Layer Risk Control Architecture

First Layer: Pre-Risk Control (Prevention)

┌──────────────────────────────────────┐
│  Pre-Risk Control Checklist          │
├──────────────────────────────────────┤
│1. Single asset weight cap (e.g.,10%) │
│2. Industry concentration limit (e.g.,<30%) │
│3. Liquidity requirement (cash + high-liquidity assets >20%) │
│4. Debt ratio screening (exclude high-debt targets) │
│5. Valuation margin of safety (require 30% discount) │
└──────────────────────────────────────┘

Second Layer: In-Process Risk Control (Monitoring)

Real-Time Monitoring Indicators:
- Portfolio volatility > target value → Reduce position
- Maximum drawdown >-10% → Trigger warning
- Correlation突变 >0.8 → Check concentration
- VaR(95%) exceeds limit → Reduce risk exposure

Third Layer: Post-Risk Control (Response)

Extreme Situation Response Plan:
- Market crash >20%: Activate hedging tools
- Individual stock default: Stop loss immediately and re-evaluate
- Liquidity crisis: Activate cash reserves
- Black swan event: Restart stress testing
Antifragile Mechanism Design

The concept of ‘antifragile’ was proposed by Nassim Taleb, referring to a system that not only resists pressure but becomes stronger under pressure.

Antifragile Design in Investment
:

  1. Option Protection
    : Purchase put options as insurance
  2. Tail Risk Arbitrage
    : Reverse layout during market panic
  3. Nonlinear Return Structure
    : Limited loss, unlimited profit
  4. Evolution Under Stress Testing
    : Optimize the system after each crisis

##10. Practical Advice: How to Build Your Own Barbell System

Step 1: Determine Risk Preference and Goals

Self-Assessment Questionnaire
:

  • What is the maximum drawdown you can accept? (e.g.,-15%)
  • Expected annual return? (e.g.,10-15%)
  • Investment period? (e.g., over 3 years)
  • Tolerance for volatility?
Step 2: Design Configuration for Both Ends of the Barbell

Defensive End (70-85%) Configuration Example
:

Asset Class Allocation Ratio Return Expectation Risk Characteristics
High Dividend Stocks 30% 6-8% Low volatility
Bonds/Fixed Income+ 25% 4-5% Very low risk
Money Market Fund 15% 2-3% Risk-free
REITs 10% 5-7% Medium-low volatility
Gold ETF 5% 0-5% Inflation hedge

Offensive End (15-30%) Configuration Example
:

Asset Class Allocation Ratio Return Expectation Risk Characteristics
Tech Growth Stocks 10% 15-30% High volatility
Option Strategy 5% -100%~300% Extreme distribution
Emerging Theme 5% 20-50% High uncertainty
Merger Arbitrage 5% 8-15% Event-driven
Special Opportunities 5% Uncertain Scenario-dependent
Step3: Establish Execution Rules

Buy/Sell Rule Template
:

Buy Conditions (All Must Be Met):
1. Valuation below 30th percentile of historical distribution
2. Fundamental score >80 points
3. Technicals not in a downward trend
4. Portfolio weight not exceeding cap

Sell Conditions (Any Triggered):
1. Valuation above70th percentile of historical distribution
2. Fundamental deterioration (e.g., earnings miss)
3. Stop-loss line triggered (single stock -15%)
4. Find better alternative opportunities
Step4: Continuous Optimization Mechanism

Monthly Replay Checklist
:-
[ ] Is the portfolio Sharpe ratio up to standard?
[ ] Is the maximum drawdown within tolerance?
[ ] Do the weights of both ends need adjustment?
[ ] Are there new risk exposures?
[ ] How is the decision quality? Any violations?


Ask based on this news for deep analysis...
Alpha Deep Research
Auto Accept Plan

Insights are generated using AI models and historical data for informational purposes only. They do not constitute investment advice or recommendations. Past performance is not indicative of future results.