The map is the product.
Guides make money by knowing the territory, not by picking a side. This is the register of every tool, dataset, venue and publication we cover — what each one is genuinely good at, where it will cost you, and whether any money changes hands between them and us. That last question is answered on every entry, without being asked.
Neutrality is the asset. Everything else is downstream of it.
A directory that can be bought is worth nothing to the reader and, within about two quarters, worth nothing to the advertiser either. So the commercial arrangement here is constrained by design rather than by promise:
- Listing is never sold. Inclusion is editorial. No entry has ever been added because of a commercial relationship, and none will be.
- Scores are never sold. Every entry is scored against the same five published criteria. A paying partner and a non-paying one are scored by the identical method.
- Disclosure is structural, not voluntary. The relationship field on each entry generates both the disclosure sentence and the link attributes. It is not possible to publish a monetised link here without its disclosure, because they are produced by the same line of data.
- The criticism stays in. Every profile carries a "what to watch" section. If a partner asked us to remove one, we would remove the partnership.
All of it is verifiable: the register is published as open JSON at /api/partners.json, including the relationship field for every entry.
Partner tools
Platforms and libraries that do the work — research, backtesting, execution.
Cloud research and backtesting environment with an open-source engine (LEAN), a large public strategy library, and brokerage integrations for live deployment.
Vectorised backtesting library built on NumPy and Numba, designed for testing large parameter spaces fast rather than simulating one strategy carefully.
Implementations of the methods in López de Prado's financial machine learning work — triple-barrier labelling, fractional differentiation, purged cross-validation.
Crowdsourced hedge fund that distributes obfuscated, point-in-time features and aggregates staked model predictions into a single portfolio.
Partner data
Market data, reference data, and the vendors who stand behind its integrity.
Market data API covering US equities, options, forex and crypto, with aggregates, trades, quotes, and websocket streaming on published tiers.
Normalised historical and live market data delivered from colocated capture, including full order book depth and nanosecond timestamps.
Partner programs
Brokerages and venues with published referral or introducing-broker programs.
Global multi-asset brokerage with programmatic access through TWS API, the Client Portal REST API, and the ib_insync ecosystem.
API-first US equities and crypto brokerage with commission-free trading, a full paper-trading environment, and straightforward REST and websocket interfaces.
Partner publications
Research desks, journals, and newsletters we read, cite, and cross-publish with.
Our own weekly institutional-grade briefing on quantitative investment strategies — flows, structures, dealer platform developments, and governance.
How entries are scored.
Five criteria, weighted equally, scored 0–10 by the editors. The same five for a backtesting library and a brokerage, because comparability is the only thing that makes a number worth printing.
Can a competent stranger get to a working result from the public docs alone?
Are limits, tiers, and real costs published, or do you have to talk to sales?
Point-in-time correctness, survivorship handling, and documented revisions.
Open formats, stable APIs, and the ability to leave with your work intact.
Auditability, change logs, incident history, and reproducibility guarantees.
Scores are judgements, not measurements. They are revised when the underlying product changes and the revision date is stamped on each profile. Disagree with one? Tell us why — corrections are published.
Working with us.
Three ways in, in ascending order of commitment:
- Get listed. Free, editorial, and the default. If you build something in this space and it is not here, tell us and we will look at it. We list things we would use.
- Referral programs. If you run one and we already recommend you, we will enrol and disclose it. We do not enrol in programs for products we would not otherwise list.
- Cross-publication. If your desk produces original QIS research, we will read it, cite it, and where it is good, co-publish it. Citation is the currency this whole site runs on.