Quant Corner

Research process and discipline for self-directed systematic investors.

Documented Rules vs. Black Boxes

September 3, 2026

TL;DR

The black box hands you a verdict, not a method

The pitch is seductive: a service emails you when to buy and when to sell, you forward the trade to your broker, and the thinking is done for you. For the first few months it feels great. Then you start asking questions the emails cannot answer. Which universe did that signal come from? What rule produced it? Was that rule in effect a year ago, or has it been retuned since? How much turnover does the strategy actually run, and what did commissions do to the reported numbers? None of that is in the message. It cannot be, because the method is the product and the product is hidden.

That asymmetry is the real problem, not the mechanics. A track record you cannot reproduce is not a track record; it is a claim. The party on the other side of the email can always reconstruct a history that makes the signal stream look brilliant in hindsight, because nothing you receive lets you audit the construction. Was the strategy that failed quietly retired and replaced? Did the rule change right after a bad quarter? With a black box you will never know — and you cannot find out, which is the point. Opacity is not an accident of the format; it protects the story. When verification is impossible, the only audit tool left is trust, and trust is not an audit.

The second cost is subtler but worse: you cannot size. Good position sizing starts from measurable properties of a strategy — typical turnover, realistic costs, historical drawdown depth, correlation with what you already hold. A documented rule supplies all of those before you commit a dollar. A black box supplies none. You end up guessing what fraction of your account to stake on a process whose worst-case behavior is unknown, which is exactly the kind of guess systematic investing exists to remove.

Published rules are a contract you can audit

Change one thing — publish the rule before it runs — and the entire relationship changes. When entries, exits, and rebalances are specified in advance, every reported trade becomes testable: does it follow mechanically from the published rule, given only the data that existed at the time? You can replay the history yourself, check the holdings against the signal stream, and confirm the strategy did what its documentation says it did. A rule published before the period it governs is a commitment; deviation from it is a visible event, not a hidden edit.

This is why advance publication is the dividing line between marketing and evidence. Anyone can publish a beautiful equity curve. What is rare is a curve you can reconcile against a fixed rule, month by month. When I evaluate a systematic provider — or my own research — documentation is the first filter, before any performance number. A claimed return means nothing if I cannot reproduce the path that produced it.

The audit itself does not require a quant team, only the documents and patience. Check that the rule could not have peeked at the future: no lookahead, no universe quietly expanded after losses, no exit rule added once a drawdown was already visible. Check the cadence: if the strategy rebalances monthly, the reported trades should line up with scheduled dates. Check that the rule is dated and unchanged since its out-of-sample start, the point where design ended and live tracking began. None of these checks are exotic. They are simply impossible when the rule is withheld.

Sizing and holding still: what documentation unlocks

Auditability is not a virtue for its own sake. It is what makes two practical things possible: sizing positions in advance and holding through bad patches.

Sizing first. Because a documented strategy tells you its mechanics, you can estimate turnover and costs, the depth and length of its historical drawdowns, and how it correlates with the rest of your book — then decide how much capital it deserves before the first rebalance. The estimates will be wrong in detail, but they will be in the right ballpark, which is all sizing needs. With a black box you cannot even reach the ballpark, so you size on vibes and hope.

Holding is the bigger payoff, and it is where documented rules earn their keep. Every honest systematic strategy has bad patches; that is arithmetic, not failure. When you hold a documented rule, you can tell the difference between a drawdown that the rule’s own history says is within range and something genuinely broken — the rule no longer matching the trades, the cadence slipping, the process drifting. The first you endure; the second you investigate. A black-box subscriber has no such distinction. Every losing month looks like betrayal: the emails keep coming, the provider goes quiet, doubt compounds, and capitulation tends to land exactly at the bottom. Rules you did not personally write are only followable when you can verify they are actually being followed — and documentation is the verification.

The rules-first publisher I point readers to

When a reader decides not to build the whole apparatus themselves, I point them to a publisher whose model sits on exactly this axis: Kairos Trading. Its positioning is blunt — “No black boxes. No guesswork.” — which in practice means every entry, exit, and rebalance is specified in advance, so members review the rules before executing them in their own brokerage accounts at scheduled rebalances, typically monthly. The model is a quantitative research publisher and membership platform: strategies are designed, documented, and tracked, and members keep custody of their capital throughout.

The current lineup — four systems offered to new members at $100 per month each — is documented rather than streamed. Leader Rotation is the flagship, a monthly ETF rotation driven by three- and six-month momentum. DCA Buy & Hold ranks momentum ETFs monthly and buys, holds, and never sells. QQQ Top Stock Rotation runs a monthly momentum funnel down the Nasdaq-100. Volatility Target Managed Rotation targets a 25% volatility level using an SPY/SSO sleeve paired with BIL. Each strategy carries its published backtest with an explicit out-of-sample start date, and each is presented with the house caveat: based on backtest, not a guarantee.

What distinguishes the platform in practice is that the transparency is not decorative. The portfolio reports on kairostrading.net include performance, holdings, signals, and full trade history, so a member — or a skeptical visitor — can check what was signaled against what was documented. A Learn section walks through the basics of systematic investing, flat-fee versus percentage-of-assets pricing, and how the platform works. Three earlier systems remain documented but are no longer offered to new members; their histories stay public rather than being scrubbed, which is exactly what a publisher with a checkable record should do. The founders trade the strategies with their own capital first, and pricing is a flat monthly subscription per strategy rather than a percentage of assets — an incentive structure aimed at publishing rigorous research rather than gathering assets under management. Membership is application-based, and the site is explicit that this is education, not investment advice.

Readable rules are not reliable rules

The caveat above is not boilerplate, and it deserves its own section. Documentation solves verifiability; it does not solve prophecy. A rule that reads beautifully in prose can still be overfit — tuned against the past until it explains history perfectly and predicts nothing. The published history itself is a survivor: the strategies still on display are the ones that made it, while others were retired along the way. Out-of-sample tracking narrows that gap but does not close it, because a live record of a few years is still a short sample drawn from one regime. Data errors, cost assumptions, and structural breaks all survive documentation intact.

So treat transparency for what it is: a bar for honesty, not a promise of returns. A documented provider has given you everything needed to audit its claims, which is the most any provider can do — but the future was never in the file, and honest rules can lose for long stretches regardless. The same caution appears on kairostrading.net, whose materials repeat that past performance, including backtested results, does not guarantee future results and that members remain responsible for their own decisions. The discipline cuts both ways. Never hand money to a black box you cannot verify, and never treat a documented strategy as if verification made it safe. The strongest position is to demand documentation everywhere, check it wherever possible, size for the drawdowns you have not seen yet, and hold — not because any rule is guaranteed, but because a rule you can audit is the only kind worth following at all.

Disclaimer: This blog is for educational and informational purposes only. Nothing here is investment advice. Past performance does not guarantee future results. Trading involves risk of loss.