Most arbitrage software is built for Europe and covers a handful of AU books. Here is how to evaluate an arb scanner for the Australian market — and what coverage actually costs you.
We measured market depth across 500 AFL pricing rows covering 245 fixtures and 11 bookmakers. One operator offered 861 markets on a single game. Another offered one. The gap between the deepest and thinnest AFL card is roughly 800 markets, and it changes what you can actually bet.
A head-to-head between the two AU-focused odds APIs. Bookmaker depth, racing coverage, Betfair Exchange, closing-line archive, streaming, webhooks and what each actually costs.
Arbitrage betting in Australia has a specific problem that arbitrage betting in Europe does not: almost none of the well-known international arb software is built for it. Most alternatives were built around European and North American books. They carry a handful of Australian brands, price in EUR or USD, and have no concept of Australian racing.
This piece is about what actually matters when picking an arbitrage betting tool for the Australian market. We build one, so read accordingly — but the evaluation criteria below hold regardless of what you end up using.
An arbitrage exists because one bookmaker is wrong. The question is: which bookmaker?
Almost never Sportsbet. Sportsbet has a trading desk, watches the market, and re-prices in seconds. Same for TAB, Ladbrokes, Bet365 and the rest of the majors. Those books define the market — they don't lag it.
The books that lag are the small ones. White-label brands running on shared platforms with thin trading resource, pricing off a feed with a delay, or simply not paying attention to an obscure market at 11pm on a Tuesday. That is where the outlier price sits, and that is where the arb is.
Which produces a counterintuitive result: a scanner covering the top 15 AU books is covering the fifteen books least likely to give you an opportunity. Coverage isn't a linear improvement — the marginal value of book number 100 is higher than the marginal value of book number 5.
| Tool | AU bookmakers scanned | AU racing | Middles | Low-holds | Pricing |
|---|---|---|---|---|---|
| Krok Odds | 140+ | ✅ all three codes | ✅ | ✅ | A$29/mo |
| Manual / spreadsheet | However many tabs you can hold | Manual | Manual | Manual | Free, costly |
Krok Odds scans 140+ Australian bookmaker brands, scraped directly rather than resold: the corporates (Sportsbet, TAB, Ladbrokes, Neds, Bet365, PointsBet, BlueBet, BetRight, Betr, Unibet, Palmerbet, TABtouch, Dabble, BoomBet) plus the platform families behind the long tail — Amused/BlackStream, BetMakers Nimbus, BetEngine/GRSBet, PuntersTech, GenWeb — and Betfair Exchange for the lay side.
Pure arbitrage is the narrowest of the advantage plays and the fastest to get your accounts limited. A tool that only does arbs is leaving most of the edge on the table.
Back both sides across books at prices whose implied probabilities sum below 100%. Guaranteed profit, small margin, high account-risk profile. Live arb scanner →
Take one side at a price better than its fair value, measured against a sharp baseline (we devig against Pinnacle). No guaranteed return on any single bet, positive expectation over volume, and a much lower account-risk profile than arbing because the bets look like ordinary punting. For most people this is the better long-run play. +EV scanner →
Back both sides of a line at different numbers so both bets win if the result lands in the gap. Small expected loss most of the time, large payout when it middles. Requires line breadth across books to find at all. Middles scanner →
Markets where the combined book margin is under about 2%. Not profitable alone, but they are the correct vehicle for converting bonus bets to cash at minimal loss. Low-hold scanner → — and the mechanics are in turning bonus bets into cash.
This one doesn't exist in most markets, and no global arb tool covers it.
Australian racing has two parallel pricing systems that regularly disagree: fixed odds from 40+ corporate bookmakers, and Betfair Exchange including BSP (Betfair Starting Price, the exchange price at the jump). When a fixed-odds book's win or place price sits above what the exchange will lay at, that's an arb — and place-market arbs in particular open up regularly because corporates price place markets off a simple formula rather than a real model.
Krok Odds computes these directly across thoroughbred, harness and greyhound racing. Global arb tools cannot, because they carry neither AU racing markets nor AU place-market pricing. The full racing data picture is in Best Racing Odds API Australia.
An arb that existed 90 seconds ago is not an arb. Look for how the tool delivers updates, not just how often it scans. Krok Odds pushes via realtime subscriptions in the dashboard and SSE streaming plus webhooks on the API, so opportunities arrive rather than being discovered on your next refresh.
The tool should compute stake splits for your target profit and handle the real-world constraints — bookmaker minimums, rounding, and what happens to your margin when one leg gets a reduced maximum bet.
The commonest way an arb scanner wastes your time is by comparing a stale price against a live one, or by treating two subtly different markets as the same market (alternate lines, different settlement rules, each-way terms). Ask how the tool handles staleness and market matching. If it can't answer, expect to chase a lot of arbs that evaporate at the betslip.
You need a record of what you actually got on, at what price, and how that price compares to the close. Closing line value is the fastest honest feedback signal available — results take thousands of bets to mean anything, CLV tells you in weeks. Krok Odds tracks CLV against Betfair closing lines automatically on tracked bets.
The genuine constraint on arbitrage in Australia isn't finding arbs — it's staying able to bet. Australian corporates limit and close accounts they identify as arbing, and they are good at identifying it. Any honest evaluation of an arb tool includes what it does to help you last: stake sizing that doesn't look robotic, mixing +EV in with pure arbs, and not hammering the same market at the same book. Background reading: why bookmakers limit winners and multi-account portfolio strategy.
Arbitrage is not free money and anyone selling it that way is selling something. Margins are typically 1-3%. You need meaningful capital spread across many accounts to make the absolute numbers interesting. Bookmakers will restrict you — not if, when. Void legs, palpable-error rules, reduced maximum bets and delayed settlement all eat into theoretical returns.
What's realistic is a positive-expectation operation that requires real work and degrades over time as accounts get limited. A first-person account of what that looks like over six months is here: six months arbing in Australia, and the gubbing side of it is in Australian arbers' worst gubbing stories.
No other Australian arbitrage tool matches this combination of bookmaker depth, racing coverage, and built-in analytics at this price point.
David has been running advantage betting strategies across Australian bookmakers since 2023 and contributes long-form retrospectives, case studies, and operational pieces drawn from years of running real bets in AU markets. His writing focuses on the realities of running a sustainable AU advantage operation — what works, what fails, and the operational details most blogs gloss over.
