A cricket betting strategy is a method you run before every bet: turn the toss, the pitch, the weather and, in Tests, the venue's draw rate into your own fair price, then bet only where a bookmaker offers more. Cricket suits it: most of those inputs can be measured from past results, and in internationals the teams are named before the toss (ICC men's playing conditions, clause 1.2.1).
Every figure here is illustrative, and every bet goes on before the first ball, in markets the cricket betting guide explains.
A cricket betting model from five inputs
This model has five inputs:
| Input | What it moves | Data to collect | When it is known |
|---|---|---|---|
| The toss | Match prices, at some grounds | Toss and match results by ground and format | 30 to 15 minutes before the start (clause 13.4) |
| Pitch and venue | Innings runs lines | First-innings totals by ground and format | Pitch reports on the day |
| Rain | Limited-overs markets | The forecast and each market's rain clause | Days ahead, firming on the day |
| Draw rate | The Test draw price | Test results by venue and era | Before the series |
| Team news | Player runs lines | The XI and the likely batting order | At the toss |
Write your estimate down before you look at the bookmaker's price, so the market cannot anchor it, then compare with the expected value calculator: edge = offered price / fair price - 1. Judge the method later on closing line value: beating the margin-free close over a few hundred bets says more than a few dozen results can.
Toss and pitch betting: measure both before you price them
The toss: is it worth anything at this ground?
Winning the toss matters only when conditions favour one choice, such as dew in a night match or a wearing pitch, as T20 cricket betting explains. Measure it first:
- From the scorecards the ICC and the home board publish, count one format's completed matches at the ground in a dated window, such as the last 10 seasons, and how many the toss winner won.
- Work out the standard error of a coin flip over that many matches: square root of (0.5 x 0.5 / matches).
- Treat anything within two standard errors of 50% as noise.
Example: the toss winner won 23 of 40 completed ODIs at a ground, 57.5%. Illustrative counts. The standard error is the square root of (0.25 / 40) = 7.9 percentage points, so 57.5% sits 7.5 / 7.9 = 0.95 standard errors above 50%. Confirming a 7.5-point effect at two standard errors takes about (2 x 0.5 / 0.075) squared = 178 matches.
If your ground has far fewer, its batting-first and chasing split has the same standard error over the same matches, so treat the toss as unproven there.
Where the toss does matter, price it in before the coin lands. Rate Team A at 58% if it wins the toss and 50% if it loses. Its pre-toss chance is then 0.5 x 58% + 0.5 x 50% = 54%, a fair $1.85 (1 / 0.54) that becomes $1.72 or $2.00 after the toss. An illustrative pre-toss $1.95 is an edge of 1.95 / (1 / 0.54) - 1 = 5.3%.
The pitch and venue: first-innings scores as the base line
An innings runs line starts from what first innings make at that ground in that format:
- Collect first-innings totals in one format from a dated window, dropping rain-reduced innings.
- Take the median, not the mean, which a few big totals drag up, and count how often the totals passed the line on offer.
- Adjust for this match's pitch, boundaries and XIs.
Example: 15 full first innings at a ground: 214, 238, 247, 255, 261, 268, 272, 279, 286, 291, 297, 305, 318, 342 and 377. Illustrative totals. The mean is 4,250 / 15 = 283.3 and the median, the eighth score, is 279. Against a line of 275.5, 8 of the 15 went over, 53.3%, a fair over price of 1 / 0.533 = $1.88. A line set at the mean, 283.5, was passed by only 7 of the 15 (46.7%).
Fifteen innings is a small sample, so blend it: count a wider median for the format, say 270, as 30 innings, and (15 x 279 + 30 x 270) / 45 = 273 is your base.
Risk: Betting involves risk. An edge worked out from a few dozen matches can be wrong either way, and a bet above your fair price can still lose. See responsible gambling for limits and support.
Rain and DLS: price the forecast into the bet
Match bets follow the official result, and when rain shortens a chase the DLS method resets what the second side needs. A revised target also caps the chase, so an innings over set for the full overs and well above the new target cannot win under a settle-on-runs rule. Runs bets follow their own rain clause, and cricket rain rules sets out which markets stand and which are void.
To turn a forecast into overs: when lost time cannot be made up, the ICC's conditions cut 14.28 overs an hour in an ODI and 14.11 in a T20 international (clause 13.7.2 of each). So two hours lost during an ODI's second innings take about 28 overs off it.
Example: an innings over at $1.90 that you rate at 55% in a full match, a 15% chance that rain cuts the innings after you bet, and a 5% chance of the over if it does. Illustrative estimates.
| Rule for a cut innings | Chance of winning | Expected value per $1 |
|---|---|---|
| No rain risk | 55% | 0.55 x 1.90 - 1 = +4.5% |
| Settled on the runs scored | 0.85 x 55% + 0.15 x 5% = 47.5% | 0.475 x 1.90 - 1 = -9.75% |
| Void if overs are reduced | 55%, on the 85% of bets that stand | 0.85 x (0.55 x 1.90 - 1) = +3.8% |
The same bet at the same price goes from worth taking to worth leaving on the rule alone. Under a settle-on-runs rule the lost value moves to the under. Rated 45% in a full match, the under is worth 0.45 x 1.90 - 1 = -14.5%. With the rain risk its chance is 0.85 x 45% + 0.15 x 95% = 52.5%, worth 0.525 x 1.90 - 1 = -0.25%.
Risk: Betting involves risk. A forecast is a chance, not a certainty, and your rain estimate can be wrong. See responsible gambling for limits and support.
Test draw rates: anchor the draw on a dated set
A Test draw price starts from how often Tests at the venue end drawn:
- List every Test at the venue in a dated window from the same scorecards, split by era, such as the last 10 years against the 10 before: pitches, scoring rates and day-night Tests change the game.
- Anchor on the recent era if the two differ, then move it for the forecast, the pitch and both batting line-ups.
Example: illustrative counts for one ground: 4 draws in 11 Tests in the earlier decade (36.4%) and 2 in 12 in the latest (16.7%). Anchor on 16.7%, a fair draw price of 12 / 2 = $6.00. With 12 Tests the standard error is the square root of (0.167 x 0.833 / 12) = 10.8 percentage points, so treat the anchor as a range.
A forecast that wipes out a day takes a fifth of a five-day Test's scheduled days, so lift the anchor. Moved to an illustrative 26%, the fair price is 1 / 0.26 = $3.85, and a bookmaker's $4.50 is an edge of 4.50 / (1 / 0.26) - 1 = 17.0%. At the dry-weather 16.7%, it was a bet to leave. Test cricket betting covers the three-way market.
Risk: Betting involves risk. A draw rate from a dozen Tests is a rough estimate, and even a well-priced draw fails most of the time. See responsible gambling for limits and support.
Team news, batting orders and player runs lines
A player runs line rests on how many balls the batter is likely to face, which depends on batting position. The nomination before the toss names 11 players (clause 1.2.1), not a batting order, so lines move once the XI and the order are clear.
Example: a batter expected to face 30 balls at No. 3 in an ODI, at a strike rate of 90, projects to 30 x 0.90 = 27 runs. Moved to No. 5 with 20 balls expected, the projection falls to 18. Illustrative figures. A line of 24.5 set before the change now sits well above it.
Batting scores tend to be skewed, with many low scores and a few big ones, so a line set at a batter's average is often passed less than half the time, as player runs betting works through. Under an all-in rule, the NSW default (clause 5.1.1), a bet backing a player who is left out of the XI stands and loses, which is the risk of betting before the team is named.
Risk: Betting involves risk. A line that looks out of date can still be right, and a bet on it can lose. See responsible gambling for limits and support.
Mistakes that undo a cricket strategy
| Mistake | The cost, on this page's numbers | Do this instead |
|---|---|---|
| Trusting a toss record from 40 matches | 57.5% was within one standard error of a coin flip | Work out the standard error first |
| Building a runs line from the mean | At 283.5, only 7 of 15 innings went over (46.7%) | Use the median of full innings |
| Ignoring the rain clause on runs bets | +4.5% became -9.75% under a settle-on-runs rule | Read the clause, or skip runs bets in rain |
| Betting player runs before the XI is named | Under an all-in rule, a left-out player's bet loses in full | Wait for the XI, or check the rule |
Where B337's screeners fit
Your estimates cover the cricket; when a cricket match is listed on B337's +EV screener, the screener covers the comparison across sports markets. It lists bookmaker prices above a fair price built from the market, with the margin taken out, and ranks them by edge. It does not model the toss, the pitch or the weather, and its fair price is an estimate, not a known probability.
The screener rows come with Terminal Pro, Full Automation or a Screeners plan arranged with the team, and a free account shows the live counts, with the rows locked. Any bet is yours to place in your own account, and bets placed through B337 use credits. For free and confidential support call 1800 858 858 or visit gamblinghelponline.org.au.
Risk: Betting involves risk. A +EV bet can still lose, and often does, and a thin sample can make a price look better than it is; there is no guarantee of profit. See responsible gambling for limits and support.