Our approach
Four steps, in the open.
How a fixture becomes a research entry, why few become qualifying selections, and how every result is kept.
- 01
Estimate the outcome
For each match and market, the model produces an estimated probability. It is an estimate, not a promise about what will happen.
- 02
Assess the price
We compare that estimate with the odds captured during the morning scan. A likely outcome can still be poor value if the price is too short.
- 03
Apply the rules
Most entries stay research only. A qualifying paper selection is one that passes every qualification check — and some days, none do.
- 04
Record the result
The original pre-match selection is kept fixed. We record the outcome and flag any verification issue, such as conflicting scores, as under review.
The daily window
A slate labelled with a date runs from 08:00 on that date to 08:00 the next morning, UK time (including clock changes). Overnight fixtures belong to the same slate.
Production thresholds and source definitions will be published here once confirmed. Example arithmetic for simple markets: expected value = (estimated probability × decimal odds − 1) × 100.