Systematic discovery and validation

Tomatoes From Mars®

A quantitative research framework for finding promising trading ideas in historical tick data, then testing whether they can survive deeper scrutiny.

Historical tick dataSystematic filtersSimulated executionValidation

From noise to research candidates

Evidence before conviction

Markets contain many possible patterns

Most apparent patterns are random, temporary, or too fragile to trade. The challenge is to search systematically without mistaking a good-looking backtest for a durable result.

Ideas need a repeatable structure

Tomatoes From Mars® records what was tested, the conditions applied, the execution assumptions used, and the resulting behaviour so that candidates can be compared on a consistent basis.

Discovery is not conclusion. A candidate is useful only if it continues to make sense after out-of-sample testing, cost assumptions, parameter changes, and risk review.

The discovery workflow

Repeated filter-and-test loop

Tomatoes From Mars® workflow

The framework scans historical data through a defined sequence of filters, signal tests, simulated execution, and result logging.

Tomatoes From Mars workflow showing testing range, regime filter, timing filter, signal filter, simulated execution, and results logging
A repeated filter-and-test process over historical tick data.

1. Define conditions

Select the historical range, instruments, sessions, data resolution, and test boundaries.

2. Filter and test

Apply regime, timing, and signal filters to examine predefined market conditions.

3. Compare and validate

Log results, assess robustness, and promote only credible candidates for deeper work.

Paid quantitative research services

Research engagements

Independent research using Tomatoes From Mars®

TFM Quantitative Trading offers paid quantitative market-research engagements using the Tomatoes From Mars® framework. Engagements may include historical data analysis, systematic signal research, backtest design and review, and research documentation tailored to an agreed analytical brief.

Each engagement is scoped before work begins, with the agreed research question, data assumptions, deliverables, timetable, and fee confirmed in writing.

Request a research engagement

To discuss paid research, email contactweb@tfmquantitativetrading.com with a short description of the market, instruments, historical period, and question you would like researched.

TFM will reply with availability, proposed scope, expected deliverables, and pricing before accepting an engagement.

Request paid research
Service boundaries: research is provided for informational and analytical purposes only. It is not personalised investment advice, a recommendation to buy or sell any instrument, portfolio management, brokerage, or custody of client assets. TFM does not accept client money or execute trades on a client’s behalf.

What the framework is not

Important boundaries

Not a prediction engine

It studies historical conditions; it does not claim to predict markets or future returns.

Not investment advice

The framework describes research methodology and does not provide investment recommendations.

Not a profitability guarantee

Research results can fail in live conditions and require careful, ongoing validation.