Build a Position Sizing & Risk Management Framework: AI Prompt for Claude & ChatGPT

You are a senior portfolio manager and risk analyst with expertise in position sizing, portfolio construction, and risk management for concentrated equity portfolios. Build a position sizing and risk management framework for [PORTFOLIO_CONTEXT].

## Context
The portfolio has [PORTFOLIO_SIZE] in total capital, with a maximum of [MAX_POSITIONS] positions. The investor's risk tolerance is [RISK_TOLERANCE]. This is an educational research exercise in portfolio construction principles; it is not personalized investment advice.

## Required Analysis

### Part 1 - Position Sizing Framework
Develop a tiered position sizing system:

| Tier | Conviction Level | Position Size (% of Portfolio) | Criteria to Qualify |
| 1 - Core | Highest | | |
| 2 - Significant | High | | |
| 3 - Starter/Tracking | Moderate | | |
| 4 - Option/Asymmetric | Speculative | | |

For each tier, define:
- Maximum single-position size
- Minimum thesis requirements (e.g. "completed full bull/bear analysis")
- Time horizon expectation
- Conditions to upgrade or downgrade between tiers

### Part 2 - Kelly Criterion Application
For a hypothetical position with:
- Estimated probability of positive outcome: [P_WIN]%
- Estimated upside if right: [UPSIDE]%
- Estimated downside if wrong: [DOWNSIDE]%

Calculate:
- Full Kelly allocation
- Half-Kelly (recommended practical maximum)
- Quarter-Kelly (conservative)
- Compare Kelly output to the tiered system above - which is more restrictive?

Discuss why full Kelly is rarely appropriate and what fraction is commonly used in practice.

### Part 3 - Risk Budget & Drawdown Management
Define portfolio-level risk rules:
- Maximum portfolio drawdown before mandatory de-risking: [MAX_DRAWDOWN]%
- Single-position maximum loss contribution (position size × max expected loss)
- Correlation-adjusted position limits (reduce sizing when adding correlated holdings)
- Sector/factor concentration limits

Build a drawdown response framework:
| Portfolio Drawdown | Action |
| -5% to -10% | |
| -10% to -15% | |
| -15% to -20% | |
| -20%+ | |

### Part 4 - Scaling In & Scaling Out Rules
Define rules for building and reducing positions:
- **Scaling in:** When does it make sense to add to a position vs. when is it "averaging down into a mistake"?
- **Scaling out:** Rules-based trimming (e.g. "trim 20% when position exceeds 1.5x target weight")
- **Stop-loss vs. thesis-based exit:** When is a hard stop appropriate vs. when should the exit be driven by thesis invalidation?
- **Position sizing after a win:** Do you let winners run to larger-than-target size, or systematically rebalance?

### Part 5 - Correlation & Diversification Check
For a portfolio of [MAX_POSITIONS] positions:
- What is the minimum diversification benefit? (How many truly independent bets are there?)
- How to identify hidden correlations (e.g. two companies both dependent on consumer spending)
- Sector concentration check: maximum % in any single sector
- Factor exposure check: are you inadvertently running a "long duration" or "long momentum" portfolio?

### Part 6 - Pre-Trade Checklist
Before initiating any new position, require:
- [ ] Completed bull vs bear analysis with probability-weighted expected value
- [ ] Defined exit criteria (both upside target and downside thesis-breaker)
- [ ] Position size determined by framework (not emotion or recent performance)
- [ ] Correlation check against existing holdings
- [ ] Portfolio-level risk budget still has capacity
- [ ] Written one-sentence thesis that can be revisited

### Part 7 - Common Sizing Mistakes to Avoid
List the 5-7 most common position sizing errors and how to prevent each:
| Mistake | Why It Happens | Prevention Rule |

### Part 8 - Visual Verification & Chart Follow-Up
- Identify 2-3 multi-year financial trends (e.g. historical drawdown magnitude, volatility cycles, and debt levels) that inform risk and sizing parameters.
- Provide the direct Chartloom visual research link for inspecting individual holding tickers: https://www.chartloom.com/c/ticker/[TICKER]

## Output Format
Tables, headers, bullets. Educational and framework-oriented. This is a research exercise in portfolio construction principles, not personalized investment advice.

**Variables:** If any [BRACKETED] placeholders above are not set, ask the user for each missing value before proceeding.