Thesis
Research/Practice

What paper trading can teach—and what it cannot

Use simulation to test a decision process, not to manufacture a track record.

Editorial noteEducational research process only. Not investment advice. Any execution in Thesis is paper-only.

01Use simulation to test repeatability

Paper trading is useful when the object of the test is your process. Can you translate a thesis into explicit entry conditions, size it against a risk budget, review the ticket, and reconcile what happened afterward?

A repeated paper workflow can expose missing fields, ambiguous decisions, and operational mistakes without putting capital at risk. That is a systems test, not evidence of investment skill.

02Treat simulated fills as approximations

A simulator cannot fully reproduce queue position, market impact, partial fills, borrow availability, auction dynamics, or the liquidity available at the moment a real order arrives. Vendor documentation also warns that paper environments differ from live markets.

Record the quote timestamp, order type, limit, and observed spread. If a conclusion depends on an optimistic fill, mark it as fragile rather than rounding the problem away.

03Simulation does not reproduce emotional cost

A paper loss does not affect rent, retirement, or reputation. The absence of consequence changes attention and behavior. It is easier to follow a stop, hold through volatility, or place a large notional trade when nothing real is at stake.

Use conservative risk budgets and pre-written review rules, but do not infer that paper discipline will transfer automatically to live capital.

04Measure process quality before return

Useful paper metrics include missing-source rate, thesis-to-ticket errors, risk-budget breaches, review completion, and whether an outcome was explained by the stated thesis. A simulated return can be recorded, but it should not be presented as a live track record.

Thesis keeps execution paper-only. Its value is the reviewable chain from evidence to decision to outcome—not a promise that simulated results will recur.

05Sources and further reading

These links are provided for source inspection and context. Their publishers do not endorse Thesis.