I started arbing in late 2023 with $4,800 spread across fourteen Australian bookmaker accounts and a spreadsheet I did not fully trust. Six months later I had $16,200 and had been restricted at nine of the fourteen books. A year in, I was back down to $12,100 and functionally could not place a meaningful bet at any AU corporate.
That gap between the peak and the year mark is the actual story. The first six months are the part people write Medium articles about. The eighteen months after that are the part that matters, and nobody talks about them because they are unglamorous.
The setup
Fourteen accounts: Sportsbet, TAB, Bet365, Ladbrokes, Neds, PointsBet, BlueBet, BetRight, Unibet, TABtouch, Dabble, Betfair, Palmerbet, and one more I would rather not name because I know someone who works there. I opened them over a three-week period, used the deposit matches where they were reasonable, and funded each with between $200 and $600 depending on the bonus.
I ran a scanner (not mine at the time, a European tool that had patchy AU coverage but was good enough for H2H and line markets). Target threshold was 2% minimum arb margin. Stake sizing was whatever the calculator said, rounded to the nearest dollar. That last detail ended up mattering more than anything else.
The first two months
The first arb I hit was a NRL H2H between the Raiders and Souths. Ladbrokes had Canberra at $2.15, Bet365 had Souths at $2.10. Book percentage 95.1%. I put $487.23 on Canberra and $497.61 on Souths. Canberra won, I pocketed $47.30 above my original stake, and I thought I had cracked something.
By the end of month one I had hit 106 arbs, banked about $1,800, and my bank balance was genuinely confusing me. Money was just appearing. I was checking the scanner every forty minutes at work. My girlfriend thought I was having an affair.
Month two was mostly the same. Bigger arbs on midweek NBA where the slow AU books were lagging US moves, some nice three-way soccer arbs on the A-League, a couple of fat player prop arbs where Sportsbet and PointsBet disagreed wildly on disposal lines. End of month two I was at about $8,100 profit. I started thinking about quitting my job. Which is a warning sign in retrospect.
The first gubbing
BlueBet was first. Day 63. I went to place a $340 leg on a NBL game and the max-bet field showed me $2.75. Same for every market on every sport. No email, no warning, just the capped field.
I was angry for about an hour. Then I did what every new arber does: I contacted customer service to sort out the error. They sent back the we-reserve-the-right-to-restrict-any- customer boilerplate. I argued. They stopped replying. I eventually withdrew the $680 sitting in the account and moved on.
What I did not do was pause to ask why BlueBet had cut me first. The answer, I worked out later, was that I had been hammering them on player prop arbs because their lines were the softest in the market on NBL and AFL. Turnover at BlueBet was probably triple my second- most-used book. I had concentrated risk into the book most likely to notice.
The cascade
Over the next eleven weeks I lost account after account. Unibet at day 81. TABtouch at day 98. PointsBet at day 104, which hurt because it was one of my best books for NRL. Ladbrokes at day 117. Neds at day 122. By the six-month mark, which is when I locked in the "$11,400 in six months" number I still quote at dinner parties, I was down to five functional corporate accounts plus Betfair Exchange.
The gubbings came faster than I expected because I had not been doing any of the things that extend account life. I was staking $487.23 instead of $500. I was placing both legs of every arb inside the same minute. I was ignoring every promotional email because promos were mostly -EV and I was "above that." Every one of those choices saved me a few cents of margin and cost me an account worth thousands in lifetime value. See the gubbing guide for the full list of things I was doing wrong.
The plateau
Months seven through twelve I banked maybe $700 total. The surviving corporates cut my stake limits piecemeal. Sportsbet was unofficially capped around $80 a bet, which makes arbing pointless because you cannot size into the sharp side properly. TAB was fine but had the worst prices of the fourteen, so there was rarely anything for me to do there.
Betfair Exchange became the workhorse because Betfair does not limit in the conventional sense. They charge a premium charge to consistently profitable accounts instead, which is its own kind of pain. But at least the accounts stay open.
Year one total: $12,100 profit on an initial $4,800. Accounting for the capital tied up across fourteen books, the effective ROI was closer to 40% than the headline 250% implies. Still good money. Not job-quitting money.
What I would do differently
Round everything from day one. Every stake, every time, to the nearest $10 or $25 depending on size. The margin hit is maybe 0.2% of arb edge. The account longevity gain is probably 2-3x. Easiest trade in the world, and I missed it because I thought precision was the point.
Claim the promos. Every deposit match, every boost, every money-back-if- draw offer. Some are -EV, but the profiling benefit of looking like a customer who cares about marketing emails is worth more than the EV cost. I ignored all of it because I thought I was optimising. I was over-optimising into a red flag.
Spread wider, thinner. I should have opened twenty accounts instead of fourteen, and I should have actively rotated. No single book above 12% of weekly turnover. I ran three or four books at 25%+ share because they had the best lines for what I wanted to bet. That is exactly the pattern that gets you cut fast.
Stagger legs. Five minutes between legs when the market would cope. Ten minutes where I could get away with it. Most arbs survive the wait. Some close. That cost is noise compared to what aggressive same-minute execution signals to a bookmaker correlating accounts.
Transition to +EV earlier. Pure arbing has a ceiling set by account survival. Positive EV has a higher ceiling and the accounts last longer because you are only betting one side of a market. If I had started mixing in +EV bets from month three, the account death in months four through six would have hit later and less severely.
Would I do it again
Yes, but differently. The returns are real. The money is real. The AU corporate bookmaker landscape is one of the few places in the world where arbitrage still genuinely works for retail-scale capital, because the density of competing books and the softness of some of them creates constant price dispersion.
But you have to go in understanding that every account is depreciating. You are not building a relationship with these companies. You are extracting as much value as you can before they cut you, and then moving on. If you can internalise that at the start instead of learning it by getting gubbed at BlueBet on a random Wednesday, you will get substantially further.
If you are thinking about starting, read the arbitrage guide first. Then size your initial bankroll at half of whatever feels right, because your first drawdown will be from a gubbing you did not see coming, and that is the point at which most people quit. Survive that, and you will be fine.
If you want a lower-friction starting point before you risk real money, track the weekly AFL tips and compare your own numbers to the market consensus.
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.