Value betting software compares every price it can read from the Australian bookmakers it covers with an estimated fair price, and lists the prices that sit above fair, ranked by edge. It builds the fair price from other prices: a reference market with the bookmaker margin taken out, an exchange's prices, or a blend of several.
So a tool is only as good as two things: how fresh its prices are, and how sound its fair price is. Even when both are right, a +EV bet loses often, and the accounts that keep taking those prices are the ones bookmakers restrict. Value betting tools sit beside odds boards and arbitrage scanners among the betting tools in Australia.
How a value bet finder builds a fair price
A value bet finder runs five steps, whatever its screen looks like:
- Read each bookmaker's price, and the reference prices, with the time each was read.
- Match the same event and market across every source.
- Take the margin out of the reference market, a step called de-vigging, to get a fair chance for each outcome.
- Divide each bookmaker price by the fair price: edge = offered price / fair price - 1.
- Keep the rows above a minimum edge and rank them.
With illustrative prices for a two-player match, where the reference market is the low-margin one a tool trusts:
| Step | Working | Result |
|---|---|---|
| Reference prices | Player A $1.62, Player B $2.50 | Implied 61.73% and 40.00% |
| Add them up | 61.73% + 40.00% | 101.73%, an overround of 1.73% |
| Take the margin out in proportion | 40.00% / 1.0173 | Player B's fair chance 39.32% |
| Fair price | 1 / 0.3932 | $2.543 |
| A bookmaker offers $2.70 on Player B | 2.70 / 2.543 - 1 | Edge 6.2% |
On racing, a tool can use the exchange as its reference instead, and which exchange price it takes moves the edge. Against an illustrative back price of $4.80 and lay price of $5.00, a bookmaker's $5.50 is 5.50 / 4.80 - 1 = 14.6% above the back price and 5.50 / 5.00 - 1 = 10.0% above the lay.
The midpoint of the two implied chances, 20.83% and 20.00%, gives a fair price of 2 / (1 / 4.80 + 1 / 5.00) = $4.898, and $5.50 is 5.50 / 4.898 - 1 = 12.3% above that. How to calculate fair odds compares the sources of a fair price in full, and the fair odds calculator runs four de-vig methods on any market you enter.
Risk: Betting involves risk. A price above an estimated fair price can still lose, and often does, and there is no guarantee of profit. See responsible gambling for limits and support.
Why every edge on the screen is an estimate
The edge depends on choices the tool made before you saw the row: which reference it trusted, how it took the margin out, and how old the reference was. Change any of them and the same $2.70 shows a different edge. With the illustrative match above:
| Fair price built from | Player B's fair chance | Fair price | Edge on $2.70 |
|---|---|---|---|
| $1.62 and $2.50, margin removed in proportion | 40.00% / 1.0173 = 39.32% | $2.543 | 6.2% |
| The same prices, margin split equally between the two sides | 40.00% - 0.86% = 39.14% | $2.555 | 5.7% |
| The reference moves to $1.58 and $2.62 (a 101.46% book), in proportion | 38.17% / 1.0146 = 37.62% | $2.658 | 1.6% |
The first two rows are two de-vig methods, and removing the bookmaker margin compares four of them. The third row is the trap a refresh cycle sets: if the tool has not yet read the new reference, it still shows 6.2% on a price worth 1.6%.
A small error in the fair chance does the same damage. If Player B's real chance is two percentage points below the estimate, 37.32% instead of 39.32%, the expected value per $1 is 0.3732 x 2.70 - 1 = +0.8%, not +6.2%.
Why a flagged +EV row still loses, and how often
Take the first row at face value: a 39.32% chance at $2.70. Expected value per $1 = 0.3932 x 2.70 - 1 = +6.2%, yet the bet loses 60.7% of the time. Over 100 bets of $20, the expected profit is 100 x $20 x 0.0616 = about $123. Each winner makes $20 x 1.70 = $34 profit and each loser costs $20, so you need 38 winners to be ahead: 37 winners leaves you 37 x $34 - 63 x $20 = $2 behind.
At a true 39.32% chance, the binomial chance of 37 or fewer winners in 100 bets is about 36%. So about one run in three finishes behind with the edge working exactly as estimated. Positive EV betting and value betting cover the expected value side in more depth.
Risk: Betting involves risk. The $123 is an average over many runs of 100 bets, not what your next 100 will return, and there is no guarantee of profit. See responsible gambling for limits and support.
What to check before paying for value betting software
| Check | Why it matters | How to test it |
|---|---|---|
| Refresh cycle, and a time on every price | A fresh bookmaker price against a stale reference shows an edge that is not there | Compare a flagged row's time and price with your bookmaker account |
| Coverage, by name | Edges at bookmakers you do not hold are no use | Match its bookmaker, sport and market list to your accounts |
| Adjustable de-vig | The method and the reference change the edge, as above | Switch method on one row and see how far the edge moves |
| Price history and closing prices | The only test of a tool is what its prices did afterwards | Check it keeps the closing price for markets it flagged |
| Before the start only | Australian bookmakers cannot take online in-play sports bets | Check rows on games already under way are filtered out |
| Thin references | A reference with little money behind it can put a fair price a long way out | Look for a minimum-liquidity or minimum-market filter |
| Total cost | A subscription and any per-bet charge come off every edge | Price it at the number of bets you expect each month |
The before-the-start check is an Australian one. Under the Interactive Gambling Act, a bet placed after a sporting event has begun is an in-play bet (s 10B), and online in-play sports betting is one of the services the Act bans in Australia. A row on a game in progress is not a bet you can place online.
How to test a value tool on your own bets
Run it on small stakes, or as paper betting, before you trust it. Log every row you take: the time, the bookmaker, the price and the tool's fair price. After the start, record the closing price, with the margin taken out the same way, and work out closing line value for each bet: your price / closing fair price - 1. With illustrative prices, a row taken at $2.70 that closes at a fair $2.55 beat the close by 2.70 / 2.55 - 1 = 5.9%.
A tool whose rows close shorter than you took them, on average, is finding prices the market agreed with. One whose rows drift out, on average, is flagging prices the market did not agree with, which can come from a stale or thin reference, a de-vig that does not suit the market, or rules that differ. Closing line value explains why the close is the benchmark, and a few dozen bets is still a small sample.
Risk: Betting involves risk. Past results are no guarantee of future results, and a tool that beat the close over a few dozen bets may not keep doing it. See responsible gambling for limits and support.
Restrictions follow winning accounts
A bookmaker can see a value bettor's pattern in its own records: bets placed when its price is out of line with the rest of the market. Under their terms, bookmakers can cut your maximum stake, stop sending promotions or close the account, and why bookmakers restrict accounts lists the usual triggers. A restriction leaves the edge per bet alone and shrinks the money it applies to. With illustrative numbers:
| Over a month of 200 bets at a 4% average edge | Before a restriction ($50 stakes) | After a $5 maximum stake |
|---|---|---|
| Expected profit before costs | 200 x $50 x 4% = $400 | 200 x $5 x 4% = $40 |
| After an $80 a month subscription | $320 | -$40 |
On some racing, a minimum bet limit set by a state racing body (in Western Australia, by the regulator) obliges the wagering operators it covers to accept eligible fixed-odds bets up to a set limit; minimum bet limits covers which races and bets they reach. Racing Queensland has also said (3 July 2025) that its conditions do not oblige an operator to accept a betting transaction completed by automation, which matters if software places your bets.
Risk: Betting involves risk. An expected profit is worked out before costs and restrictions, and either can turn it negative, as the $5 maximum stake does above. See responsible gambling for limits and support.
Rows or counts: what a free tier shows
Free tiers on value tools tend to show less than the full list. Some give a count of prices above fair and the best edge, some show a sample with the bookmaker or price hidden, and some show rows with a delay.
A count tells you the tool is running and how busy the market is. It says nothing about how many of those prices you could take at a useful stake, how old they are, or whether they are in markets you bet.
A best-edge figure is the extreme of the list, so before trusting a large one, check the time on both prices, whether both sides settle under the same rules, and how many bookmakers price the market. A delayed row shows a price that may already be gone: it is good for judging the format, not for betting.
Mistakes that make a value tool look better than it is
With the illustrative numbers above:
| Mistake | What it costs |
|---|---|
| Taking a row without checking when the reference was read | 6.2% on screen for a price worth 1.6% |
| Reading one de-vig method as the answer | An edge of 6.2% one way and 5.7% the other, on the same $2.70 |
| Judging the tool on one 100-bet run | About 36% of runs at a real 6.2% edge finish behind |
| Paying a flat fee once your stakes are capped | 200 bets at a 4% edge and a $5 maximum make $40, against an $80 subscription |
Skip a paid tool when its expected profit before costs (edge x turnover), at the stakes your accounts will take, is below its monthly cost, or when it covers few of the bookmakers you hold.
Where B337's +EV screener fits
B337's +EV screener lists bookmaker prices across sports markets that sit above the fair price, ranked by edge, with filters for sport, market and bookmaker. For sports, the fair price comes from taking the margin out of prices across the market, including exchanges and global market references, and combining the results, so it is an estimate and not a known probability.
With adjustable de-vig you change which prices count and how much margin comes off each, and every edge is worked out again. A free account sees the live count and the best edge, with the rows locked.
The rows (event, market, bookmaker, price and edge) come with Terminal Pro, Full Automation or the Screeners plan, and Terminal View does not include the screeners. On racing, the Terminal's EV overlays compare each bookmaker's price with the exchange market instead. You place each bet from the list yourself, in your own accounts: by hand, or from the Terminal on Terminal Pro or Full Automation. Bets placed through B337 use credits, a per-bet usage charge, so count them in the cost as well as the plan.
Risk: Betting involves risk. A +EV bet can still lose, and often does; the fair price behind every row is an estimate, and a restricted account may get only small stakes on at those prices. See responsible gambling for limits and support.
For free and confidential support call 1800 858 858 or visit gamblinghelponline.org.au.