Python Financial
Python Financial is a software company focused on quantitative finance and systematic trading research. We create tools and practical learning resources for people who research markets, test trading ideas, compare platforms, and build financial applications.
python.financial is our reference site for understanding trading research platforms, programmable tools, and the ideas behind them. It brings together reviews, guides, comparisons, concepts, tools, implementations, and benchmarks in one place.
Recommendations are based on the stated use case, source material, executed examples, and published benchmark evidence. Readers should use the same evidence to reach their own conclusion.
Editorial process
Each tool or platform page starts with the best available primary evidence: current official documentation, pricing and plan limits, product specifications, supported integrations, and hands-on output where practical. Source repositories, package metadata, release histories, and software licenses are checked when they exist. Closed-source services are not treated as if their internal behavior were independently inspectable.
- Facts and recommendations are kept separate.
- Strengths and limits are described for the same kind of work.
- Pricing, plan limits, licenses, maintenance, data coverage, and supported integrations are checked at the source and dated when they can change.
- Code examples and feature reproductions are treated as demonstrations, not proof that a strategy works or that an independent implementation replaces the full service.
- Proprietary data, broker connectivity, operational infrastructure, and other non-replicable services are separated from calculations that can be reproduced programmatically.
- Publication and update dates change only when the page changes meaningfully.
Automated checks help find broken links, inconsistent metadata, inaccessible controls, and mismatched benchmark files. They do not replace source review or responsibility for the published result.
How reviews are produced
Reviews examine a question using official product material, hands-on testing, published studies, exported results, public APIs, and available source code. They distinguish company claims, results reported by other authors, our measurements, and our interpretation. When a review discusses someone else's experiment, it does not mean we have repeated that experiment ourselves.
Each review states the date through which evidence was considered and links to its sources. If we reproduce a platform feature or make our own measurements, we identify the reference implementation separately and describe what the comparison can and cannot establish.
How benchmarks are published
A benchmark is published only after every included implementation produces the required result under the same stated rules. The page records the machine, software versions, inputs, warmup and measurement procedure, and the parts of the workflow that were excluded.
Every release includes exact values and downloadable evidence. A narrow runtime result is not turned into an overall ranking of the tools.
Corrections and contact
Frameworks change, documentation can be unclear, and mistakes are possible. To report an error, provide a current source, or question a benchmark assumption, email contact@python.financial. Corrections that materially change a page are reflected in its updated date.