Betfair Exchange Cricket Historical Odds: Tiers, Fields, and Pitfalls
Historical Betfair Exchange cricket odds come from Betfair's historical data service, which holds Stream API data from 2016. Basic is free with one-minute last traded prices and no volume; Advanced adds one-second data, three ladder levels, and volume; Pro records every tick with the full ladder. Handle suspensions, bet delay, and matched versus offered prices.
The official source: Betfair's historical data service
Betfair sells and gives away recorded Exchange data through its historical data site, historicdata.betfair.com. Per Betfair's own data scientists' documentation, checked on 2026-11-05, the archive holds Stream API data from 2016 onward, covers nearly all Exchange markets, and includes cricket alongside racing, soccer, tennis, golf, and other sports.
It comes in three tiers:
- Basic: free. Prices sampled at one-minute intervals, last traded price only, no volume.
- Advanced: paid. One-second intervals, the top three levels of the price ladder, and volume.
- Pro: paid. Every API tick, at 50 millisecond granularity, with the full ladder and volume.
Prices for the paid tiers are set per package on the site, so check them at purchase time. Betfair's documentation states that Pro files are needed for robust simulations or backtesting across a series of markets, and Australian and New Zealand customers are directed to contact Betfair for Pro access.
How to get the files
The workflow on the site is: log in with a Betfair account, select a data set and period and complete the purchase step (free Basic data goes through the same step, once per period), then filter on the My Data page by sport, date range, event, market type, country, and file type, and download.
Downloads arrive as a TAR archive of bz2 compressed files, one per market or event. Each decompresses to newline-delimited JSON. Betfair recommends downloading in smaller chunks, such as one month at a time, because the site is slow, and points to its historical data API for bulk volumes. Filter to cricket and to the market types you need before downloading; a full cricket season at Pro granularity is large.
What the fields are
The files replay the Stream API. Each line is a market change message with a millisecond epoch timestamp pt and a list mc of market changes. Each change can carry two things:
marketDefinition: the full market state, resent whenever it changes. Key fields includestatus,inPlay,betDelay,marketType,eventName,openDate, andrunners, each with a selectionid,name, andstatus.rc: runner changes, sent as deltas.ltpis the last traded price,tvthe runner's traded volume,trdthe volume traded at each price, andatbandatlthe price and size available to back and to lay.
Because rc messages are deltas, the order book must be rebuilt by applying each change in sequence. A size of zero removes a price level. Which of these fields appear depends on the tier: Basic carries prices without volume, and only Pro carries the full ladder.
Cricket-specific structure
Cricket markets have features that matter for analysis:
- Draws. In multi-day matches the match odds market includes a draw runner. In limited-overs formats it usually does not, and ties are handled by the market rules.
- Long events. A Test can trade for five days, so one market can produce a very large file, with most of the volume in-play.
- Interruptions. Rain delays, reduced overs, and revised targets change the match state without a ball being bowled, and prices move accordingly.
- Side markets. Innings runs, top batter, and similar markets are thinner than match odds, so their last traded prices go stale quickly.
Read Betfair's rules for each market type you study, since settlement under abandonment or a reduced match differs by market.
Pitfalls that distort cricket odds analysis
Matched versus offered. ltp, trd, and tv describe money that actually matched. atb and atl describe what was available but unmatched. A study of where the market priced an event should use matched data or the midpoint of the best back and lay, not one side of the ladder.
Suspensions. Markets are suspended at key moments, and the suspension shows up as a change in status in the market definition. Prices quoted around a suspension are not tradable. Drop those intervals or flag them, otherwise a backtest will appear to trade through a wicket at the pre-wicket price.
Bet delay. In-play markets apply a delay before bets are matched, recorded as betDelay. Any backtest that assumes instant execution in-play overstates what was achievable.
Sampling. Basic data samples last traded prices once a minute. In cricket, a wicket and the repricing that follows can happen inside that minute, so Basic is fine for pre-match and coarse in-play curves but poor for event-level reaction studies.
Thin markets. In low-volume markets, the last traded price can be minutes old. Pair it with traded volume before treating it as a probability.
Aligning odds with ball-by-ball events
The Betfair files have timestamps but no match events. Public ball-by-ball sources such as Cricsheet record deliveries by over and ball, not by wall-clock time, so the two cannot be joined directly. Common approaches are to infer event times from suspensions and sharp price jumps in the Exchange data, or to use a commercial event feed that carries timestamps. Whichever is used, validate a sample of matches by hand. In-play odds history covers the alignment problem across sports.
Other sources of cricket exchange odds
- Betfair's Australian data scientists' files, which publish processed CSVs for some markets, including the Big Bash League and Women's Big Bash League from 2020, per their data sources page.
- Your own collection through the Stream API with a live application key, which gives you exactly the markets and granularity you choose, from the day you start.
- Commercial odds archives, which mostly hold bookmaker prices rather than exchange ladders. Useful for comparison, but not a substitute for matched exchange data.
See what historical odds data actually contains for how archives differ in general.
A practical starting workflow
Start with free Basic data for one competition and one season of match odds. Parse it into a table of timestamp, market, selection, last traded price, and market status. Plot price paths for a few matches against the scorecards to learn the shape of the data. Then decide whether the question needs Advanced or Pro. Questions about pre-match drift rarely need ticks; questions about in-play reaction to wickets usually do.
This page describes data and method and is not betting advice.
Frequently asked questions
- Is Betfair historical cricket data free?
- The Basic tier is free and includes cricket. It samples last traded prices once a minute and has no volume. Advanced and Pro tiers are paid, with prices set per package on historicdata.betfair.com. Even free data must go through the site's purchase step once per period before it can be downloaded. Details checked against Betfair's data scientists' documentation on 2026-11-05.
- What is the difference between Betfair Basic, Advanced, and Pro data?
- Granularity and depth. Basic samples once a minute with last traded price and no volume. Advanced samples once a second with the top three ladder levels and volume. Pro records every API tick at 50 millisecond granularity with the full ladder. Betfair's documentation recommends Pro for robust simulation or backtesting across a series of markets.
- Does Betfair historical data include in-play cricket odds?
- Yes. The files record markets through their whole life, including in-play trading, with an inPlay flag and market status in the market definition. For cricket, in-play usually holds most of the traded volume. Suspensions and the bet delay are recorded too, and both need handling before in-play prices are used in any backtest.
- How do you tell matched from offered prices in Betfair data?
- Last traded price, traded volume, and volume traded at each price describe money that matched. The available to back and available to lay ladders describe offers that were waiting, unmatched. Use matched data, or the midpoint of the best back and lay with enough size behind it, when estimating where the market priced an outcome.