Pre-registered trial protocol
Published , before any trial under this protocol.
This page states, in advance, how a future trial will be run and counted. It records no trial, reports no outcome, and proves nothing on its own. A trial under this protocol has not been run: it still needs approval of its inputs and timing before it runs.
1. The break-even formula
Break-even tickets = ceil(guarantee ÷ net per-ticket contribution), where net per-ticket contribution = ticket price − variable cost per ticket. The result is rounded up to the next whole ticket.
An optional fixed-cost stack may be entered: venue cost, production cost, marketing cost, other fixed costs, and a sponsor offset that is subtracted. Tickets are then ceil((guarantee + entered fixed costs − entered sponsor offset) ÷ net per-ticket contribution). A blank field is not entered and is not zero. The recorded trials entered none of these fields, so their counts stay ceil(guarantee ÷ net per-ticket contribution).
2. The miss definition
A miss is recorded when a comparison-arm pick differs from the grounded pick. The grounded pick is the Qloo-grounded pick; each comparison arm (Gut, Web) is scored against it, separately.
3. The counting rule
A trial counts only when the grounded pick AND at least one comparison pick both came from real runs. If the grounded pick is missing, or if no comparison pick came from a real run, the run is not a trial and is not counted.
4. The fail-closed rule
An outage means no trial, never a substitute. If a required evidence leg fails, or a required arm is unavailable, no verdict is produced, no pick is substituted, and nothing is counted in place of the trial.
5. Appended regardless of outcome
The trial will be appended regardless of outcome — whether the comparison arms agree with the grounded pick or miss it, and whether the break-even number is high or low. No trial run under this protocol is dropped for its result.
6. Insufficient data → no claim
If the data needed for a claim is absent, the trial/report states no claim rather than substituting or estimating. A missing leg, a missing pick, or a missing input produces a stated absence — never a substitute value, never an estimate presented as a result.
Change log
v1 — 29 October 2026: initial publication of this protocol, before any trial under it.
v2 — 30 October 2026: an optional fixed-cost stack may be entered (venue cost, production cost, marketing cost, other fixed costs, and a sponsor offset that is subtracted). A blank field is not entered and is not zero. The recorded trials entered none of these fields, so their ticket counts are unchanged.
Any protocol change is logged here with its date and reason, and never applies retroactively to already-recorded trials. A trial recorded under an earlier version stays judged by the version in force when it ran.