Historical Odds Data
Up to 10 years of odds,
synced to game state.
25+ years of stats and results across 5 major sports and 125+ leagues, with up to 10 years of pre-match and in-play odds (basketball since 2016) and multi-book coverage across every sport. A clean, machine-readable time-series built for model training, backtesting, and market research.
Implied win probability from the moneyline
One game, pre-match line through the full in-play path.
Illustrative.
A clean odds time-series, joined to game state
Every game arrives as a sequence of timestamped odds, each tagged with the live score and clock at that moment - so the price and the situation are already joined for you.
- Pre-match lines and the full in-play path
The closing line before tip-off and the complete live sequence, not just open and close.
- Joined to game state
Every snapshot carries the live score, period, and clock.
- Captured from multiple sportsbooks
The same game across books, so timing and disagreement are visible in one series.
- One consistent format
No scraping, no stitching feeds together, no per-sport reformatting on your side.
years of stats and results
leagues across 5 major sports
lines from multiple sportsbooks
refresh available
5 major sports, 125+ leagues
Stats and results since 2000 in every sport, pre-match and in-play odds since 2016 in basketball, and multi-book odds across all five. New leagues land in the same format, so adding coverage never means re-integrating.
Basketball
NBA, WNBA, NCAA (M/W), EuroLeague, Japan B.League, China CBA, Spain ACB, Germany BBL
Football
NFL, NCAAF, CFL
Baseball
MLB, NPB, KBO, Mexico LMB, NCAA
Ice Hockey
NHL, KHL, SHL, Liiga, DEL
Soccer
Premier League, La Liga, Serie A, Ligue 1, MLS, Argentina Primera
+ ~100 more leagues
Four things in every record
Every record is labeled and ready to model against.
The lines
Moneyline, spread, and totals - for the full game and for the periods within it (halves, quarters, sets).
The game state
Live score, period, and clock attached to every snapshot.
Pre-match and in-play
The closing line before tip-off and the complete live path - not just open and close.
Built-in signal
Which sportsbook each line came from, with the sharp book and the most-coverage book flagged for each league.
What teams do with it
One dataset, mapped to the work your team already does.
Model training & backtesting
A clean, labeled history of price, state, and outcome to learn from.
Closing-line value studies
Measure CLV and how lines migrate from open to close.
In-play model calibration
Tie live price to score and clock to calibrate live-pricing models.
Market efficiency research
Compare sharp versus softer books and quantify where the edge sits.
Settlement & grading
Period and final scores to grade outcomes and reconcile bets.
New-market expansion
Stand up a new sport or league on data you already trust.
A simple pull, live in days
No bespoke per-sport handling, and no pipeline to build and maintain - your engineers are loading real data in days.
Pull by API
JSON or CSV from a single authenticated endpoint. CSV loads straight into your warehouse.
Backfill once
Take the history for the sports and leagues you want.
Refresh daily
Scheduled updates daily keep the series current as new games settle.
Why teams trust this data
Built by people who have done this at scale.
One source of truth
Multiple books and a synchronized game-state feed, normalized end to end.
Proven with an industry infrastructure provider
Amelco, which builds trading and sportsbook technology for major operators, purchased Odditt historical odds data to support model testing.
Data pedigree
Our CTO was DraftKings' first data hire and scaled their data org from 1 to 100+.
Scoped to the coverage you need
Pricing is scoped to each buyer's coverage.
How it is packaged
One-time backfill - priced to the sports, leagues, and depth you take.
Recurring refresh - as often as daily, keeps the series current for ongoing model work.
Scales with you - start with one sport and grow into the rest.
Typical engagement
- Backfill:
- one-time, priced to scope
- Refresh:
- recurring monthly subscription
- Sample:
- one-off CSV sample, on request
- Format:
- JSON or CSV, into your warehouse
Frequently asked questions
Straight answers on coverage, format, and pricing.
How far back does Odditt's historical odds data go?
Stats and results go back 25+ years, to 2000, in every sport. Pre-match and in-play odds go back up to 10 years (basketball since 2016), with multi-book odds across all five sports.
Does the data include in-play odds with the score and game clock?
Yes. Every odds snapshot carries the live score, period, and clock at that moment, from the closing line before tip-off through the complete in-play path, not just open and close.
Which sports and leagues are covered?
5 major sports and 125+ leagues: basketball (NBA, WNBA, NCAA, EuroLeague and more), football (NFL, NCAAF, CFL), baseball (MLB, NPB, KBO and more), ice hockey (NHL, KHL, SHL and more), and soccer (Premier League, La Liga, Serie A, Ligue 1, MLS and more).
Is the data from multiple sportsbooks?
Yes. Lines are captured from multiple sportsbooks and every line is tagged with the book it came from. The sharp book and the most-coverage book are flagged for each league.
Which markets are included?
Moneyline, spread, and totals, for the full game and for the periods within it (halves, quarters, sets).
Can I use it to backtest a betting model or measure closing line value?
Yes. Teams use it for model training and backtesting, closing-line value studies, in-play model calibration, market efficiency research, and settlement and grading with period and final scores.
How is the historical odds data delivered?
By API, as JSON or CSV from a single authenticated endpoint. You take a one-time backfill for the sports and leagues you want, then refresh as often as daily.
How much does historical odds data cost?
Pricing is scoped to each buyer's coverage: a one-time backfill priced to the sports, leagues, and depth you take, plus an optional recurring monthly refresh.
Can I get a sample first?
Yes. Tell us a sport and we'll send a CSV sample so your team can load the real data and judge it directly.
Tell us a sport, and we'll send a sample.
We'll send CSV samples so your team can load the real data and judge it directly. Backfills are running now - start before football season kicks off.