Archive / What paper trading has to prove before you use real money

What paper trading has to prove before you use real money

A prompt that writes your paper trading test plan before you start, so you decide in advance what passing looks like and don't move the goalposts later.

Nothing here is financial advice or a recommendation to buy, sell or trade anything. Do your own research.

For: Solo founders, New to AI

Topics: Investment

Why I'm sharing it

Paper trading is practice with fake money, and it's easy to treat it as a formality. Two good weeks, and you're ready. That's how people talk themselves into real money on an idea that was never tested.

I'd rather write the test before the trades. If I decide in advance what passing looks like, I can't quietly change it when the results come in.

What makes it work

It sets the bar before the data. The pass rules are written while you still don't know the answer.

It compares to the simplest alternative. If an index fund over the same dates did better, the idea didn't pass.

It counts trades, not just days. A few weeks can show the plumbing works. It can't show an edge for a strategy that trades rarely.

It includes costs. Paper fills usually ignore fees and slippage, so paper results flatter real ones.

It tests the safety rules too. A stop that never fired during the test hasn't been tested.

The prompt

Prompt

Help me write a paper trading test plan before I start. My idea: [describe the strategy in plain words]. The account I'll paper trade on: [account or tool]. How long I can test: [time]. Write the plan with: 1. The exact rules of the strategy, frozen as of today, so I can't change them mid-test. 2. The benchmark: an S&P 500 index fund over exactly the same dates. 3. Pass rules: beat the benchmark after estimated costs, a minimum number of closed trades, and a maximum drop I'll accept. Suggest numbers and explain each, and mark them as suggestions for me to decide. 4. Fail rules that end the test early. 5. How I'll estimate fees and slippage, since paper fills usually leave them out. 6. A checklist to confirm my safety rules fired when they should. 7. A one-line log format for every trade: date, idea, entry, stop, exit, result, and the reason. End with the questions I should answer before calling the test a pass.

How to use it

Write the plan, then don't touch it until the test ends. If you want to change a rule, that's a new test from day one.

Log every trade, including the ones you'd rather forget.

If the idea fails, that's a good result. You found out for free.

This is a way to test ideas, not a recommendation of any strategy. It isn't financial advice, and nothing here is a recommendation to buy, sell or trade anything.

My own write-up, shaped by the honest results posts other builders shared online about the gap between backtests and forward trading.

Use it for: Setting pass and fail rules for any trading idea before you test it, so the results can actually tell you something.

Back to the archive