How the project works, why it exists, and how to read the results.
Monopoly Money Investor is my personal simulation project, built to compare how different portfolio styles behave over time. It is not financial advice, and it is not a public investing platform. Each portfolio starts from the same baseline and is tracked in the same environment, which keeps comparisons fair and easy to follow.
My core idea is simple: if three decision styles manage similar capital, where do results begin to split? This site answers that with clear metrics, holdings data, and historical snapshots. The aim is to focus on process quality, not bold claims or hindsight.
The AI-managed portfolio follows a structured, fully autonomous review cycle. Human input does not influence AI portfolio trade decisions. Any decision to buy, sell, increase, reduce, or replace a position is made and implemented entirely by the AI system using its own reasoning and research workflow.
The advisor-managed portfolio reflects a more traditional FA-style approach to allocation and review. It acts as a steady benchmark and helps show whether the AI setup is adding real value or just producing different choices.
The self-managed portfolio represents hobbyist research and conviction-led decisions. Together, these three approaches create a clear comparison between AI process, advisor structure, and independent judgment.
Every portfolio starts from the same value, so return comparisons stay fair. The main metrics are current value, gain or loss, and return percentage. Daily snapshots feed the chart, so you can see both today’s position and the path taken to reach it.
The head-to-head table is built for quick scanning. It places key metrics side by side so leaders, laggards, and consistency patterns are easy to spot. Holdings count is included too, because concentration can shift outcomes in a big way.
The platform also tries to keep reporting stable. If a live price is missing for a short period, fallback handling keeps views readable and preserves longer-term comparisons.
Weekly rebalancing runs on a set schedule, currently Friday evening, and it applies only to the AI-managed portfolio. During each run, positions are reviewed for thesis quality, risk fit, and relative opportunity. Holdings that no longer fit can be reduced or removed, while stronger candidates can be added, all without human intervention.
This cadence is intentional. Weekly review is quick enough to adapt, but slow enough to avoid noise-driven overtrading. It also creates a clean cycle for review notes, which helps separate repeatable process from short-lived luck.
All simulated buys and sells are logged against available market pricing so changes can be traced and understood.
The app stores holdings, snapshots, and transaction updates in a local database workflow. Market prices come from Yahoo Finance endpoints with checks for ticker resolution and tradability. It supports several asset types and includes fallback logic where direct lookup is difficult.
Development has used AI coding and research tools to speed build work, testing, and analysis. Even so, outputs are interpreted inside this project framework. The tools support the process, but they do not replace accountability.
The engineering style stays simple on purpose: readable PHP components, predictable routing, and shared render helpers for consistent metadata and layout.
This live instance is personal and is not intended as a shared app for public use. I've made the GitHub repository public so anyone can download the code, run a local or self-hosted copy, and test ideas in their own environment.
If you want to try it, clone the repo and configure your own data. Then use your own dashboard to compare outcomes, inspect holdings, and review rebalance decisions over time. This keeps ownership, privacy, and setup control with each user.
You can treat the project as an open-source framework for experimentation. Adjust portfolio structures, improve the interface, or extend the workflow to match your own style.
This is a simulated portfolio environment. It does not include every real-world execution factor, and it is not a recommendation engine. Portfolio values shown here do not include trading charges or platform dealing fees.
Costs, slippage, taxes, liquidity limits, and personal suitability vary by investor and jurisdiction. Results shown here are not promises of future performance.
No content on this site is personal financial advice. This project is educational, analytical, and also a bit of fun. It is my own experiment to see how my AI setup compares with an FA-style portfolio and a hobbyist research portfolio. AI investing tools have been around for years; this one is simply my own version to test how far I can take it.