Data product · raw
The raw odds feed — normalised prices, your logic on top
The raw feed delivers what an aggregator should: 60+ UK bookmakers normalised into one clean schema, one price per book per market, from a single endpoint. No pairing, no ratings — just dependable, deduplicated prices for teams that already have (or want to own) their matching, rating or comparison layer. When you'd rather that relational work arrive done, that's the matched feed.Powers a leading UK matched-betting platform.
$ curl api.oddsrelay.io/v1/odds/standard?region=uk \
-H "Authorization: Bearer or_live_••••••••••••"GET 200 · application/json
{
"event": "Arsenal v Chelsea",
"sport": "football",
"region": "uk",
"market": "match_odds",
"selection": "Arsenal",
"back": { book: "bet365", odds: 2.1 },
"lay": { exchange: "exchange", odds: 2.12 },
"rating": 98.1,
"updated_at": "2026-07-01T14:08:22Z"
}What the raw feed is
One normalised surface across 60+ UK bookmakers: consistent selection naming, consistent market types, consistent structure, so you're integrating one schema instead of sixty. The value is the normalisation and the breadth — the domestic books big aggregators skip are in here alongside the household names — delivered on a monitored, published-freshness cadence. The docs show the response shape and the coverage dashboard lists every book live.
It's the right buy when your product's differentiator is the logic you put on top of prices — a bespoke rating model, a comparison engine tuned to your audience, an in-house matching pipeline you want to control end to end. You get clean inputs and full freedom over what you build from them.
Raw versus matched — an honest fit question
The two products are the same coverage discipline at different altitudes. The raw feed gives you prices and leaves the relational work to you; the matched feed pairs each back price to a current exchange lay price, rates it, and computes the qualifying loss. If you're building an oddsmatcher or arbitrage scanner, that matching layer is a year of engineering you can skip — matched is usually the right buy. If you're building a comparison surface or your own models, raw keeps you in control.
It's a genuine either/or, and we'll say so: the build-vs-buy solution page walks the decision plainly, and the comparison with The Odds API covers where a raw aggregator is the correct choice.
Coverage and delivery
Coverage is deep rather than broad by design: 60+ UK bookmakers normalised into one schema, with bet365 available on the raw feed as an add-on. The feed is pre-match on roughly a three-second polling cycle, monitored, with published freshness. Delivery is one HTTPS endpoint, per-client keys, JSON out — the same integration discipline as the matched feed, so you can move up to matched later without re-plumbing.
Raw coverage is UK today. Regional expansion is shared honestly on the regions view — expanding or roadmap, never claimed as shipped before it's live.
What you get
One normalised schema
60+ UK bookmakers deduplicated into consistent selections, markets and structure — integrate one shape, not sixty.
Depth over breadth
The smaller UK books a shallow feed omits are in the roster, so your comparison isn't missing rows.
Published freshness
Pre-match on a monitored ~3s polling cycle, with freshness published on the live coverage dashboard.
A clean upgrade path
Same endpoint discipline and keys as the [matched feed](/products/matched-feed) — move up when you want the pairing done for you.
An honest note
The raw feed is prices, not opportunities — no pairing, rating or qualifying loss; that relational layer is the matched feed. It's pre-match (~3s polling), not in-play. Raw coverage is UK today, with regions expanding as shown on the coverage dashboard. bet365 is available on the raw feed as an add-on rather than bundled — the matched feed is where bet365 is included as standard.
Questions
What does the raw feed actually deliver?
Normalised bookmaker prices: one price per book per market across 60+ UK bookmakers, in one consistent schema. It does not pair back prices to lay, rate them, or compute qualifying loss — that processed output is the matched feed.
Should I choose raw or matched?
Choose raw if your differentiator is the logic you build on top of prices — a bespoke comparison engine or your own matching model. Choose matched if you're building an oddsmatcher or arbitrage scanner and want the pairing, rating and qualifying loss to arrive done. The build-vs-buy page walks it through.
Is bet365 in the raw feed?
bet365 is available on the raw feed as an add-on. On the matched feed, bet365 is included as standard across every feed type.
Which regions does the raw feed cover?
The United Kingdom today. Regional expansion is shown honestly on the coverage dashboard — expanding or roadmap — and never described as shipped before it is live.
Can I start on raw and move to matched later?
Yes. Both products use the same endpoint discipline and key model, so moving from raw to matched is a scope change, not a re-integration.
Related products
Matched feed
The same coverage with pairing, rating and qualifying loss done for you.
Feed types
The processed outputs available when you move up to matched.
Build vs buy
The honest decision between raw inputs and done-for-you output.
Live coverage
Every book in the feed, with current freshness.
Bookmakers covered
The full directory of covered books.
vs The Odds API
Where a raw aggregator is the right buy.
Clean prices, your logic on top
See the covered books live on the dashboard, then a free trial key puts the raw feed into your own build this week.