AI Tips
AI Horse Racing Tips Today: How Our Picks Are Produced
A transparent tour from the raw racecard to today's published shortlist, including the data, model instructions, confidence checks and proof record.
Read articleGuides, methods and results
Understand how the selections are produced, how the proof is measured and where the limits sit before you use today's racing pages.
Start here
AI Tips
A transparent tour from the raw racecard to today's published shortlist, including the data, model instructions, confidence checks and proof record.
Read articleResults
A 3,752-race analysis of win strike rate, confidence-band selection and the difference between an honest comparison and an inflated claim.
Read articleAI Tips
AI Tips
A practical checklist for separating a transparent free prediction from an uncheckable winner claim.
Read articleAI Tips
Tomorrow's cards create a trade-off between publishing early and waiting for better declarations, going and price information.
Read articleAI Tips
A practical guide to what an AI prediction can measure, what it cannot know and how to compare the output with prices and settled results.
Read articleMethodology
Methodology
AI scales structured comparison; humans add context and judgement. The useful contest is a transparent, like-for-like record rather than a slogan.
Read articleMethodology
A practical framework for testing a racing idea while controlling leakage, overfitting, price assumptions and selection bias.
Read articleMethodology
A clear nap-checking routine for comparing a tipster's strongest daily selection with its price, evidence and complete historical record.
Read articleResults
Results
Winning often and making money are not the same. This guide shows how price, stakes and sample size connect the two.
Read articleResults
Three common result terms, the different questions they answer and the mistakes that make performance tables look better than they are.
Read articleResults
The original Fable result, why access stopped and how the returning model is performing against the same published proof window.
Read articleMajor Meetings
Grand National
A race-specific guide that connects the runners with the site's data and product tools.
Read guideMajor Meetings
Pre-race reasoning retained with the result update, rather than rewritten after the finish.
Read guide