What QCA Is
QCA is a trading system built around a single structural rule: it does not act unless the available evidence clears a defined threshold. The default state is abstention. A position is opened only when the signal passes an explicit, pre-registered gate — and when it does not, the system holds cash and records why it declined rather than treating inaction as a missed opportunity.
This is the opposite of the common failure mode in discretionary trading, where the pressure to "do something" produces trades that exist to relieve the operator's anxiety rather than to express an edge. QCA removes that pressure by making the null action — doing nothing — the structurally rewarded default.
Why It Was Built This Way
The core design claim is narrow and worth stating precisely: the quality of a decision should be judged against the evidence available at decision time, not against the outcome. [INFERRED — this is the design principle the gate encodes, not a measured result.]
Most emotional bias in trading enters through a gap between those two things. An operator who makes a poorly-evidenced trade that happens to win learns the wrong lesson and repeats it; one who abstains correctly but watches the market run without them feels regret and loosens the gate next time. Both are outcome-driven updates to a process that should be evidence-driven.
By forcing every trade through an explicit gate, QCA produces an auditable trail: for each decision there is a record of what evidence was present and whether it met the bar. This makes abstention a logged, defensible action rather than an absence. Whether this discipline improves returns over a discretionary baseline is [UNKNOWN] — we have not run a controlled comparison, and we will not claim an edge we have not measured.
Applying the Same Rule to Ourselves
The honest test of an evidence-gating philosophy is whether its authors apply it when the evidence is inconvenient. Ours currently is. Here is the complete first-party record of QCA's own market presence to date:
- Landing page views: 6 [OBSERVED —
business_learn:landing_view=6] - Call-to-action clicks: 1 [OBSERVED —
business_learn:cta_click=1] - Instagram-attributed traffic: zero [OBSERVED —
gap_id:instagram_attribution=OBSERVED_ZERO] - Hacker News acquisition experiment: did not run; the posting account was blocked as new [OBSERVED —
acquisition_experiment:hacker_news=blocked_new_account]
A marketing page would round these into a growth narrative. The evidence-gating rule forbids that. Six landing views and one CTA click is not a funnel; it is four data points. We report them because the alternative — implying traction we cannot substantiate — would violate the exact principle the product is built on.
A few things do follow directly from this record:
- The single CTA click against six views gives an observed click-through of roughly one in six. This is a raw count, not a rate estimate — the sample is far too small to generalize, and we state the ratio only to be precise about what was seen. [INFERRED]
- The zero Instagram attribution is a genuine negative observation, not a tracking gap we are papering over: it is explicitly recorded as
OBSERVED_ZERO. Whether that reflects no Instagram effort, broken attribution, or a real absence of channel fit is [UNKNOWN]. - The Hacker News experiment produced no acquisition signal at all, because it never executed — the account was blocked before posting. We therefore have no evidence about HN as a channel, which is different from having evidence that it does not work. [INFERRED]
What We Learned
The most transferable lesson is methodological rather than financial. Building a system whose default is abstention changed how we read our own metrics. When "no action" is a first-class, logged outcome inside the trading engine, it becomes natural to also treat "no signal" as a first-class outcome in the business — hence recording Instagram attribution as an explicit OBSERVED_ZERO rather than omitting the line, and recording the blocked HN account rather than quietly retrying until something posted. [INFERRED]
We also learned a concrete operational constraint: new accounts are gated on distribution platforms. The Hacker News block is a reminder that acquisition experiments have their own preconditions, and that a blocked experiment yields no information — it must be re-run under valid conditions before it counts as a result. [OBSERVED]
What Remains Uncertain
Most of what matters is still open:
- Does evidence-gating outperform? [UNKNOWN] We have described a discipline, not demonstrated a return advantage. No backtest or live performance figures are presented here because we will not publish ones we cannot stand behind.
- What is the real conversion behavior? [UNKNOWN] One click on six views cannot support a rate estimate. We need at least one to two orders of magnitude more traffic before the number means anything.
- Which acquisition channels fit? [UNKNOWN] Instagram shows zero attribution; Hacker News was never tested; no other channel has data. The channel question is effectively unanswered.
- Where is the gate threshold correctly set? [UNKNOWN] A gate that abstains too often forgoes real edge; one that is too permissive reintroduces the bias it was meant to remove. We do not yet have enough logged decisions to calibrate it.
The through-line is consistency: a system that abstains from trades on thin evidence should abstain from claims on thin evidence too. This article is written to that standard. When we have measured results — on returns, on conversion, on channels — we will report them with the same tags and the same restraint. Until then, the honest status of QCA is a well-specified discipline with a very small evidence base, and we would rather say exactly that than overstate it.