WHY GLINTARY IS DIFFERENT

A position has to earn its place on the record.

GLINTARY does not publish a market because a model likes a team or because a prediction sounds convincing. A position is published only when the evidence, the odds, the uncertainty, the timing and the final market check all survive the same documented process.

A prediction alone is not enough.

A sports opinion can be correct and still be a bad position. Who wins is one question. What the market was offering when the position was recorded is a different one, and it decides whether the opinion was worth anything.

So every candidate is judged twice: on the research case, and on the odds at which that case is still worth holding. The question GLINTARY answers is not

Who will win?

It is

Does this market still qualify at the odds actually observed now?

A good view at the wrong odds is not a position.

Fair Odds are the de-vigged market benchmark — what the market implies once the operator’s margin is removed. The methodology then adds the required edge buffer and an uncertainty margin, and that produces the Minimum Qualifying Odds for this candidate.

If the fresh Recorded Odds come back below that threshold, the research does not get a second vote. The view may still be interesting. The odds are not good enough. Therefore GLINTARY publishes nothing.

These figures are illustrative. 2.05 is not a global minimum. Every candidate is given its own threshold from its own de-vigged benchmark, its own edge requirement and its own uncertainty.

ILLUSTRATIVE EXAMPLE

Fair Odds: 1.98
Minimum Qualifying Odds: 2.05
Recorded Odds: 1.94
Decision: DOES NOT QUALIFY

Most candidates are supposed to disappear before publication.

Every event that enters the pipeline passes through the same sequence. A candidate that fails at any point stops there; it is not carried forward on the strength of an earlier step.

01Global market scan
02Deterministic screening
03Deep research
04Source verification
05Local and context research
06Red-team challenge
07Final fresh odds refresh
08Policy evaluation
09Pre-event lock
10Public record

Seven research roles contribute to the packet the policy engine reads.

Sports Data: Assembles the factual base for the event: fixtures, form, availability, schedule and result history.
Intelligence: Looks for what the base data does not contain — reporting, context and developments that change the read.
Local Source: Reads the market in its own language and region, where the earliest reliable reporting usually appears.
Market Context: Describes how the market itself has behaved: where it opened, how it has moved and what that implies.
Source Verifier: Checks that each material claim traces to a source that can carry it, and disqualifies the ones that cannot.
Red Team: Argues the opposite case as strongly as it can, so the position has to survive its best available objection.
Synthesiser: Assembles the surviving evidence into one structured packet the policy engine can evaluate.

These roles research, structure and attack a candidate. They do not decide. The gates are deterministic and versioned: they read stored numbers and return the same verdict every time they are run on the same record. Agents cannot vote their way around a failed gate, and a persuasive write-up does not raise a threshold.

The market gets the last word.

Research takes time, and markets move while it happens. So immediately before a position can become official, GLINTARY refreshes that event’s odds again and re-runs the check against the Minimum Qualifying Odds that were set for it.

If the fresh Recorded Odds have fallen below that threshold there is no lock, no position and no publication — no matter how much work went into the candidate.

Zero is a valid number.

Methodology v1.1 sets no daily maximum for official positions and no minimum publication quota. Removed the arbitrary daily OFFICIAL-position cap while retaining singles, flat 1u, one OFFICIAL position per event and the remaining integrity gates.

No arbitrary daily cap.
No minimum quota.
A day with zero positions is a valid outcome.
One official position per event.
Singles only.
Flat 1.00u risk on every position.

Publishing more would be easy. It would also make the record meaningless, because the only way to publish on a day when nothing qualifies is to lower the standard.

Once it is published, it stays published.

Every official position carries:

A Position ID
The time it was locked, before the event
A cryptographic commitment to the locked payload
Its batch and inclusion proof, where applicable
A public verification record

Later information does not rewrite the original. An update is appended next to it, and a correction is recorded as a correction. After the event the result is settled, the unit P&L is computed and the Ledger updates. Wins, losses, pushes and voids all stay where they are.

Verification proves integrity and timing: that this exact record existed before the event and has not changed since. It does not prove that the research was right, and it does not claim the odds were available to everyone. How verification works.

Every position risks the same 1.00u.

GLINTARY uses a flat 1.00u risk for every official position. A unit is a neutral performance measure, not a currency amount.

At decimal odds 2.18, a winning 1.00u position produces +1.18u profit; a loss produces -1.00u. Pushes and voids produce 0u.

A flat 1u risk prevents discretionary stake sizing from distorting the public record: every position counts the same, so the record measures selection, not confidence.

The words on a position, and what they mean.

Recorded Odds: The decimal odds attached to the locked position, observed from the recorded source at the recorded time.
Minimum Qualifying Odds: The lowest odds at which the position still satisfies the published GLINTARY research threshold.
Fair Odds: The de-vigged market-implied benchmark derived from the reference market.
Best Observed Odds: The highest relevant odds GLINTARY observed in the recorded market evidence at that time. It describes what was observed, not what is available to any particular reader. GLINTARY defines this term but does not display a value for it. It cannot be reconstructed reliably from the immutable record, and a figure that cannot be verified does not belong on the public record.
Closing Odds: Comparable recorded market odds near closing, shown where reliable closing data exists for the same line.
Market Reference: A neutral comparison line recorded alongside a position. It is not the Fair Odds benchmark and not the Recorded Odds, and it is never a recommendation.
Line: The handicap or total the selection is measured against. In “Under 49.5 at 1.94”, 49.5 is the line and 1.94 is the odds.

Naming a source is evidence, not promotion.

A public record shows the operator or source at which the odds were observed, the region, the Recorded Odds, the observation time and an evidence identifier. Without those, “the odds were 1.94” is an assertion rather than a record.

That is the whole reason the name appears. GLINTARY carries no operator logos, no affiliate links, no sign-up prompts and no “bet now”, and it never claims the odds were available to you.

Market intelligence ≠ execution ≠ affiliate promotion.

GLINTARY records market intelligence. It does not execute wagers, does not check availability for you and does not promote operators.

Pre-event only.

The current public methodology publishes pre-event positions only. There are no live or in-play positions in this beta.

That is a methodology limit rather than a missing feature. In a live market the score, the clock and the odds all change within seconds, so recording what was true at the moment of a decision — and proving it afterwards — needs its own timing, state and verification rules. Until those exist and are published, GLINTARY does not publish live positions.