Why everyone scrapes, and why they should stop
Ask "where do I get racing data in a spreadsheet" on any forum and the standard answer is still "write a scraper". Scraped data breaks silently, has holes exactly where you need prices, and sits on the wrong side of the sites you take it from. The scrapers exist mostly because people do not know licensed options exist at hobbyist prices. They do.
The Racing API, tier by tier
The free tier gives basic daily racecards and results, enough to build against. Basic (£27.99 a month) adds advanced racecards, per-horse history and a first analysis endpoint. Standard (£59.99) adds bookmaker odds, twelve months of queryable results, search across horses, jockeys, trainers and owners, and the full analysis endpoints: strike rates, A/E and profit-loss computed for you. Pro (£99.99) extends queries to any entity's full history, adds future racecards, breeding data and odds-movement history. This site runs on it, which is the plainest endorsement we can offer.
Betfair's APIs
The Exchange API is free with a funded account and one-time key setup: delayed prices, markets and traded volumes, which is plenty for studying market moves. Live-price keys cost serious money and exist for trading tools, not hobby projects. Historical Betfair SP files are downloadable free and are the standard for proofing bets at exchange prices, which is exactly how our The Weigh-In keeps its records honest.
What the licensed route cannot give you
Ratings owned by private firms (Timeform figures, Racing Post ratings) are their intellectual property and no API sells them cheaply; sectional times remain fragmented and partly paywalled. Everything else a serious hobbyist needs, cards, results, form, odds, analysis, is available above board for less than two streaming subscriptions.