Calibration
Do events given a 60% probability happen about 60% of the time? Graded only against forecasts recorded before the event.
Evidence
What BetDNA measures about itself, and what it refuses to claim.
Measurements
What is measured
Coverage
How much history stands behind a sport, a competition or a participant. Thin coverage is reported as thin, not smoothed over.
Out-of-sample validation
A strategy that only works on the data it was found in has not been found. Candidates are tested on data they never saw.
Rejection
The number of candidates rejected is published alongside the number approved, because the first number is what makes the second one mean anything.
What is deliberately not claimed
- No guaranteed winners, and no guaranteed profit.
- No claim that betting can be made risk-free. It cannot.
- No historical return presented as an expectation of a future one.
- No recommendation on an event with no legitimate market price.
- No sport advertised as fully supported while its replay is not yet held to the sport’s own mechanics.
- No placeholder data shown to make coverage look complete. Where genuine data cannot be sourced, the area is marked as blocked and kept out of the product.
Why an empty Today is a good sign
A product that always has something to recommend is telling you about its incentives, not about the sport. Prices are mostly efficient; on most days, on most events, the market is not obviously wrong. A system that finds a strong opinion every single day is not finding edges - it is lowering its standards until something qualifies.
BetDNA reports the empty day.
See what today actually justifies
Create an account and open BetDNA. If it has nothing worth backing today, it will tell you that too.