Analytical error
The hypothesis or the weighting of factors was wrong.
CONSENSUS evaluates each sporting event through a multi-layer analytical process. Different scenarios and data inputs are checked independently before a signal is published. If the evidence does not hold together, the event is passed. Every published assessment is timestamped before the start, recorded at the available market price and settled in public.
RATIONALE: the scenario held across independent checks, and the market price left a usable margin against the internal estimate.
Actual entry from the log. Not a mock-up.
Every signal is fixed before the event starts — the scenario, the status, the odds and the time of record. Once settled, the entry stays unchanged.
* Calculated in units under a single fixed methodology. Past results do not guarantee comparable results in the future.
Not a feed. A handful of signals, and the record that lets you hold us to them.
Every event ends in one of three states — CORE, RISK or PASS. The list is short because the third state is the most common: where the grounds are insufficient, nothing is published.
The scenario cleared the system's main checks: the data is sufficiently complete, the key factors are stable, and the current market price still leaves analytical interest.
The grounds for the scenario are there, but some factors reduce the stability of the assessment. Such a signal is marked separately and is not treated as equal to CORE.
Two separate gates, and an event needs both. The analytical gate asks whether the independent checks support one coherent direction. The price gate asks whether the current market price still covers the model's own margin of error. Agreement between the checks is necessary, never sufficient: a thesis at a bad price is a PASS, not a smaller entry.
A chronology of published assessments: time of record, event, scenario, the original odds and the actual result. Successful and unsuccessful entries are kept the same way.
| Date | Event | Score | Signal | Competition | Status | Odds | Result |
|---|
Once an event is settled, CONSENSUS compares its original assessment against the actual result. The system examines which premises held, which turned out to be wrong, and where the level of uncertainty was misjudged.
The fact is matched against the original assessment — the scenario, the status and the price recorded at the time of publication.
Errors and deviations are classified: a wrong reading, a price without margin, ordinary variance, or a change in conditions after the signal was fixed.
Analytical parameters are adjusted on the new data, and the updated version of the model is recorded before the next block of events.
Every completed event becomes a new observation for the calibration of CONSENSUS. That is what the open log is for: it is the input to the system, not only a report to the reader.
Once events are settled, the forecast assessment is compared against the actual result. Errors are classified, parameters are revised, and the accumulated history is used for regular calibration.
Full statistics for the period, the size of the sample, the dynamics of the result and any change in signal quality. Positive and negative periods are analysed the same way.
Each unsuccessful signal is assigned one of four causes. The verdicts feed the next calibration: errors are not archived — they are worked through.
The hypothesis or the weighting of factors was wrong.
The direction held, but market conditions left no margin.
The scenario sat inside an acceptable range; the alternative outcome landed.
Line-ups, injuries or market moves appeared after the signal was fixed.
For readers who want to understand the process. CONSENSUS discloses the stages of analysis and the information used at each stage, while the internal architecture, weighting logic and calibration parameters remain proprietary.
Every event is analysed across a set of sporting, statistical, contextual and market factors. The composition of that data depends on the sport, the competition and the quality of what is available.
Key hypotheses pass through several independent analytical checks. This reduces the influence of a single error, a random signal or one source of data.
A signal is published only after it clears the internal criteria for quality and stability. Where uncertainty is high, the event is marked PASS.
CONSENSUS does not promise an outcome. Sporting events carry irreducible uncertainty. The task of the system is not to remove risk, but to analyse the data consistently, assess the quality of a scenario, and decline to publish where the grounds are insufficient.
Open a demo account and see today's signals and the full settled record — no email required. CONSENSUS is in public validation: the sample is growing in the open ahead of a commercial launch.
No card · Free while the record is being built
A signal is only useful if its timestamp and market price are known before kickoff. Leave an address and the next ones arrive while the price is still there — along with the periodic result reports.
Unsubscribe at any time
The record is kept at a fixed unit size: the same reference amount on every signal, regardless of confidence or odds. A unit is a unit of measurement for the record, not a sum of money — the block above shows the same history at 0.5 U, 1 U and 2 U. CONSENSUS does not recommend what a unit should represent.
Every signal is fixed and time-stamped before kickoff — that is what makes the record checkable. Delivery before the price moves is what the email list above is for; per-signal delivery timing is part of what the current validation period is working out, and we would rather say that than promise a number we do not yet hold.
The number of signals is not a KPI of the system. If an event does not clear the internal quality criteria, it is not published. On some days there are no signals at all.
The system works with sporting events across disciplines. The published record so far is football: it started with the 2026 World Cup (now complete, final included) and continues with club competitions. The tournament run was strong; the early club stretch has been rougher, and both are in the log. As new competitions are added, their entries appear in the same ledger under the same rules.
Yes. Removing negative results would destroy the point of an open log. Every settled signal stays on the page, and each unsuccessful one is classified in the post-analysis.
CONSENSUS uses a multi-layer analytical architecture with independent checks and versioned calibration. The specific models, internal configuration, weighting and orchestration are proprietary and are not disclosed.
No. CONSENSUS provides informational and analytical assessments of sporting events. Any forecast carries uncertainty, and past results do not guarantee future ones. The decision to use the information rests with the user.
No. CONSENSUS is not a bookmaker, does not accept funds for the purpose of concluding wagers, and is not a party to any agreement on winnings.