Arbitrage betting in New Zealand — the rules, the arithmetic, and the brand shortlist

Updated October 2026

Licensed

Available in US

Fast payouts

18+ Only

The arithmetic behind an “arb” is simple to describe and hard to live with in New Zealand. Back every outcome of an event at prices whose implied probabilities add up to less than one, and the bet returns a small, locked-in margin whichever way the result lands. The New Zealand piece is more awkward: since an amendment to the Racing Industry Act 2020 took effect on 28 June 2025, the only operators allowed to take online sports and racing bets from people in New Zealand are TAB NZ and its online sister brand Betcha. A technique that, by its own logic, wants several bookmakers sits inside a market that, by law, has two.

Current as of 31 October 2026; licence and monopoly claims checked against the Department of Internal Affairs (DIA) and TAB NZ’s own statements.

The statute and what it covers

What the Racing Industry Act 2020 now says

The racing side of the New Zealand market has been a monopoly for decades; the sports side collapsed into the same monopoly on 28 June 2025, when section 74AAA of the Racing Industry Act 2020 — inserted by the Racing Industry Amendment Act 2025 — came into force. The wording the Department of Internal Affairs administers is strict: no person other than TAB NZ may offer racing betting, sports betting, or other racing or sports betting to a person in New Zealand, regardless of where that person is resident or incorporated. The Online Casino Gambling Act 2026 covers a different product. It licenses online casino gambling in New Zealand only; it does not license sports betting, and TAB NZ winning the right to apply for a casino licence does not extend the betting monopoly, because casino betting and sports betting sit on separate permits.

TAB NZ and Betcha are two brands of the same operator group. They are not rivals, and the “second legal option” framing is wrong. TAB is the established brand, on the high street and online; Betcha is the TAB NZ app aimed at a younger, online-first audience, with daily odds boosts, a social feed that lets punters follow tipsters and share bets, and a tip section called the Playbook. The two share the same product licence, the same harm-minimisation framework and the same self-exclusion list, so an exclusion on one brand carries across to the other.

What runs the business behind the monopoly

The day-to-day operation of TAB NZ is run by Entain under a 25-year agreement signed in 2023. Entain and TAB NZ split gross profits 50/50, with at least $150m a year flowing to TAB NZ for the first five years; Entain also paid a one-off $100m to the New Zealand racing sector on signing. The result is not just a legal monopoly but a long, capitalised one — the kind of arrangement a competitor needs years and unusual terms to dislodge, and there is no licence to apply for that lets one try.

What the statute does not do

Two things the act does not do are worth naming, because they are easy to assume. The act does not block overseas bookmaker websites from resolving on a New Zealand internet connection, so the page still loads; the liability sits on the offshore operator, not on the player. And section 19(4) of the Gambling Act 2003, inserted in 2025, says an individual cannot be convicted for placing a bet with an offshore betting operator in breach of section 74AAA — so a New Zealand resident who uses one is not published as committing an offence themselves. DIA’s position is the operator is breaking New Zealand law and the player is exposed to the practical risks that come from a regulator that cannot reach the brand.

What may be bet on, and what may not

TAB NZ takes fixed-odds and tote (parimutuel / pool) bets. The published betting rules list Win, Place, Each Way, Quinella, Exacta, Trifecta, First4, Double, Treble, Daily Double, Running Double, Quaddie and Big6, and a Multi and a Same Game Multi (a multi whose legs come from one event). Live betting is in the TAB NZ apps; no statutory restriction on in-play markets has been established, so the limit is the operator’s own terms. Spread betting sits outside TAB NZ’s sports-betting definition; no New Zealand betting exchange has been licensed; and bets on politics and novelty markets are not stated as in or out, so an interested punter checks the rule page for the specific market. The ban covers events held anywhere in the world, not only New Zealand events.

From 1 August 2026, the Racing Industry Act 2020 greyhound racing closure takes effect in stages. From that date, no operator — TAB NZ or otherwise — may take bets on New Zealand greyhound races, and offshore operators are equally barred.

Wellbeing: the harm-minimisation frame the monopoly buys

The age, the identity and the protections in place

The age for betting in New Zealand is 18 (betting under 18 is an offence under section 89 of the Racing Industry Act 2020); the land-based casino age is 20, and the two are not interchangeable. TAB NZ accounts require a verified New Zealand address and identity documents, and the same controls cover Betcha — the two brands sit on one customer database, which is why a self-exclusion applies to both.

The harm-minimisation tools in the TAB NZ apps are deposit limits, time-outs and self-exclusion, with self-exclusion taking effect across the group. TAB NZ is also required to keep problem-gambling harm minimised and pays a Problem Gambling Levy of 0.74% — in force from 1 July 2025 to 30 June 2028, a fixed-window rate that funds treatment and prevention services. The free Gambling Helpline (0800 654 655, text 8006) and Safer Gambling Aotearoa are the public entry points a punter in a bind is steered to; both are reachable from within the apps.

The limits of the framework

The framework is a New Zealand one. It only covers TAB NZ. The same exclusion does not land at an offshore bookmaker in another jurisdiction with its own self-exclusion lists; the New Zealand register the 2026 online casino law will require is, by its scope, an online-casino register and does not extend to sports betting. So the harm-minimisation rails an arbitrage bettor in New Zealand leans on — a single exclusion that follows them, a regulator that can hear a complaint, a levy that pays for treatment — apply cleanly only inside the TAB NZ pair of brands. Outside it, the rails are the brand’s own.

Crypto, anonymity, and the overseas-only corner

Why crypto comes up at all

Crypto enters the picture because arbitrage bettors who run across brands often need fast settlement, low fees, and a payment method that works across borders without a wire. In the New Zealand context that interest has a sharper edge: an overseas bookmaker taking New Zealand customers is the business model section 74AAA targets, and the question for an arbitrage bettor is less “is Bitcoin anonymous?” than “does using it here put the operator on the wrong side of New Zealand law?”. The law sits with the operator, not the wallet.

What crypto does and does not change

Crypto is pseudonymous, not anonymous: every transaction on a public ledger is traceable, and address clustering, KYC at the on- and off-ramp, and chain analytics tie wallet activity back to a person far more reliably than the word “anonymous” suggests. The offshore brands the comparison below names accept crypto where the local regulator allows; the New Zealand brand does not, because the lawful local option (TAB NZ and Betcha) takes bets in New Zealand dollars through standard New Zealand-facing payment methods, and the player’s interest in privacy in that market is small. The medium that matters is the brand’s licence and the regulator that can hear a complaint, not the rail the funds run on.

The honest version: arbitrage bettors who choose crypto are choosing an offshore operator, and offshore operators are the business section 74AAA tells to withdraw. Crypto is one of several payment rails such brands accept. It does not turn an unlawful offer into a lawful one; the law is set by where the bet is being taken, not how it is being paid for.

Forecasting, bankroll and the analytic frame

What “predictions and strategy” actually means for an arber

An arbitrage bettor is not forecasting outcomes. The point of the bet is that it does not matter which outcome wins: every leg pays, and the cumulative return exceeds the cumulative stake. So the analytic work is upstream of the result and downstream of the price — finding the price differences, sizing each leg so the payout is even (or close to even), and running enough cycles per month that the small per-cycle margin compounds into something meaningful. The forecast is a price; the prediction is the calendar and the volume, not the winner.

That is the part the marketing around arbing tends to skip. A 1% margin on a one-off arb is not a wage; a 1% margin on twenty settled cycles a day across a year is. The workload is the volume, and the obstacle is not the math — an arbitrage calculator does it in seconds — it is the bankroll an operator will let a known arber keep, which the section below covers.

Bankroll, sizing and the edge

A textbook arb on a balanced two-outcome market with prices 1.90 / 1.90 (illustrative — not a quoted price) is a single cycle of “back each side at 1.90, stake each so the payout is even, collect one payout that returns slightly more than the combined stake”. The arithmetic behind that example sits in the section on Selection and odds below, where the margin piece belongs; the bankroll piece is the same calculation repeated across as many cycles as the bankroll covers, which is why the per-cycle return has to clear the friction — payment fees, time spent, lines that get cut — before it becomes income.

A bankroll also has to clear the minimums and the limits of the brands in use. TAB NZ has its own minimum stake and its own cap on winnings per bet, both stated in its terms. Offshore brands named in the comparison below have their own deposit minima, their own per-bet caps, and their own internal rules on what kinds of bets qualify for a promotion or a bonus; an arb that lands a bonus position that the operator then voids because the bet profile is “irregular” has paid the full wager for no return.

The bookmaker’s counter-move

The arithmetic assumes the bets land. The bookmaker’s response assumes they will not. Maximum wager limits on accounts identified as arbers drop sharply, and accounts themselves can be closed. The Wikipedia entry on arbitrage betting summarises the response plainly: bookmakers detect arbing quickly, view it unfavourably, drop max wager limits, and have in the past voided half of an arbitrage bet. That is the part of the “system” the marketing copy does not describe, and it is the part that costs.

Selection, prices and the odds that move

Fundamentals of an arbitrage opportunity

An arbitrage opportunity exists when the odds cover all mutually exclusive outcomes of an event and the implied probabilities add up to less than 1. Convert any decimal odd (say, 2.10) to its implied probability by dividing 1 by the price (1 / 2.10 ≈ 0.476, or 47.6%); add the implied probabilities of all outcomes; if the total is below 1, the bookmakers collectively under-quote the event by that gap. The gap is the bettor’s margin. A 2012 study (Franck, Verbeek and Nuesch) found that 98% of arbitrage opportunities return less than 1.2%, so arb bets involve relatively large sums of money — the size and the volume are not optional. A $100 arb at a 1% margin is a dollar; a $10,000 arb at the same margin is a hundred dollars, and that is what the literature considers a useful cycle.

The “live” variant replaces the pre-match price with an in-play price on the same event. The bookmaker’s in-play feed updates every tick, sometimes faster; the price on the leg the bettor has not yet placed can move before the bet is on, in which case the arb closes while the bettor is mid-leg and the unfilled leg sits as a directional bet at whatever price was caught. Live arb is harder to clear because the time window is short and the price that opened the opportunity may not exist at the moment the bettor acts.

Worked margins from the page’s two illustrative prices

A two-way market at decimal odds of 1.90 / 1.90 illustrates what an even book looks like and is not a quoted price. The implied probability of each side is 1 / 1.90 ≈ 0.5263, or 52.63%; the two together sum to 1.0526, or 105.26%. That 5.26 percentage-point gap above 100% is the bookmaker’s margin: across the whole market, every unit staked returns 1 / 1.0526 ≈ 0.95 units in payout, and the margin is 1 minus that, or about 5%. Two notes on the result. It is a band, not a single figure — the same 1.90 / 1.90 quotes rounded to one more or one less decimal place give 4.7% to 5.3%, depending on which decimal the market settles at; that is the working spread an arber quotes against. And the same 5.26% margin is the floor of the figure — the only way the price gap drops below zero (an arb opening) is if at least one bookmaker offers 1.95 or longer on a side, because 1.95’s implied probability is about 51.28%, and two of those sum to 102.56%, which is a negative 2.56% margin for the bookmaker (a positive 2.56% margin for the arber). The whole arb hunting process is looking for points like that.

The calculation a reader who wants to work it out themselves carries is straightforward. Convert every price on the event to its implied probability; sum across all outcomes; subtract from 1; if positive, an arbitrage opportunity exists, and the arber’s per-cycle margin is that positive number, scaled by the bankroll. Where the sum of implied probabilities falls below zero for a market, the page quotes the result as a band, not a fixed figure, because the underlying prices move and a stated single percentage would not be reliable.

What “odds cover all outcomes” requires in practice

The cleanest arbs are markets with two mutually exclusive outcomes — head-to-head, over / under 2.5, a one-on-one match. Markets with three or four runners — a horse race, a correct-score market, a tricast — multiply the difficulty because the arber has to back every runner, and one bookmaker may well shorten a tail runner after the arber has placed on it, breaking the cycle. Same Game Multis are not arbs in the textbook sense — they are single-price accumulators at a known vendor, not a multi-vendor spread.

The brand shortlist and what each name means in New Zealand

How the list below is set

TAB and Betcha are the two online brands of TAB NZ, the only provider lawfully offering racing and sports betting to people in New Zealand since 28 June 2025 (Racing Industry Act 2020, section 74AAA). The other eight names are bookmakers New Zealand punters meet; each names its own offshore licence (or none) and none has a New Zealand licence because none can get one. The list is not a recommendation and not a ranking — a foreign brand presented on a New Zealand-regulated site carries the operator’s unlawful exposure, not the regulator’s blessing. It is a comparison by the same five criteria, so the reader can see what each name brings and what it does not.

The matters the table below compares are the brand’s New Zealand licence status and what the operator’s own terms state on the subject key — what an arbitrage bettor finds there. A “no-data” entry in a cell means neither the brand’s home page nor its terms page, on the materials in hand, state the point for the side being compared. The reader checks the brand’s own terms before staking, the same way they do for any other operator anywhere.

Brand comparison table

The two New Zealand brands return “operator’s terms address account conduct” rather than a clean “no arbing” line because the materials in hand do not surface a published anti-arbitrage rule from either brand — that is not the same as saying neither has one, and a reader who plans to scale up checks the published house rules before they size up. The eight offshore brands return the same shape for a different reason: their sportsbooks publish terms, but a stand-alone “arbitrage” clause is not in the materials in hand. Where a brand names its jurisdiction (Curaçao, the Union of the Comoros) the table records it; the absence of a row marked “no licence named” is itself information.

What the table does not say

A comparison table cannot say how each brand handles a flagged account. The published house rule on an arb — where one exists — typically reads something like “irregular betting patterns may result in stake restriction or account closure”, and the working of that rule sits behind the brand’s customer service. A reader who plans to scale up finds the rule before they scale, because the brand’s tolerance for arbing determines whether the bankroll survives the volume the math requires.

Overview: the working frame a reader keeps

What arbitrage betting is, plainly

Arbitrage betting is placing a bet on each of the mutually exclusive outcomes of an event with different bookmakers so that, whatever the result, the total payout exceeds the total stake by a small margin. The mechanic relies on price differences between brands and is the same whether the brands are local exchanges, monopoly bookmakers with second-screen price feeds, or the multi-vendor mixture the offshore sportsbook market runs on. The mechanic is legal as an activity in most jurisdictions; what is regulated is who is allowed to take the bet, not the bettor’s choice.

What sets arbitrage betting apart from regular betting is that the bettor’s skill edge is in the price comparison, not in the prediction. A punter with a strong opinion who backs one side is exposed to that side losing; a punter who backs both at a margin is exposed to one bookmaker voiding a leg, to the price moving before the second leg lands, and to the account being limited once the brand’s system flags the pattern. The arbitrage bettor’s edge is a small per-cycle margin that compounds by volume; the risks are operational rather than predictive.

The three constraints a New Zealand arb keeps in mind

First, the brand count. Two lawful brands limit the technical scope of a multi-vendor arb inside New Zealand: any opportunity that requires prices from three or more bookmakers cannot be run with TAB NZ and Betcha alone, because the third leg has nowhere lawful to land. A reader who wants a true multi-vendor arb has to use at least one offshore brand, and offshore brands serving New Zealand customers are the section 74AAA target. The honest trade is between the volume an arb needs (high) and the regulatory exposure per offshore leg (real).

Second, the per-cycle return. The 2012 literature finding that 98% of opportunities return under 1.2% is the order-of-magnitude ceiling for what an arber clears per cycle, and the per-cycle return has to clear the friction (fees, time, lines cut) before it is income. The page’s worked margin on the 1.90 / 1.90 example shows a per-cycle margin close to zero for the bookmaker — which is the same way around, an arber’s per-cycle return close to a percent when the bookmakers collectively under-quote by that amount. Per-cycle is a percentage; bankroll is the multiplier; volume is the cycle count.

Third, the brand response. Maximum wager limits drop, accounts get closed, and individual bets or halves of bets get voided. The risk to the bankroll is not a losing streak — by construction, an arb has no losing streaks; it is a series of small wins and one larger structural event (a closed account, a voided leg) that can wipe a stretch of small wins at once. A reader who treats arbing as a low-risk income is reading only the per-cycle mathematics and missing the operational tail.

Where the choice leads, by reader

A reader who bets on New Zealand racing and sports for entertainment and wants the protection of a regulated market has one product to choose between — TAB or Betcha — and the choice between them is a question of interface and product depth, not a question of monopoly versus competition. A reader who wants the multi-vendor spread arbitrage arithmetic wants has the offshore market to look at, and that market is what DIA warns about: operators in jurisdictions with minimal consumer safeguards, no guaranteed payout of large wins, and no return of unspent credits. A reader who cares about anonymity or fast cross-border payments has the same offshore market, paid in crypto where the brand accepts it, with the same caveat. None of those choices has the rails TAB NZ and Betcha carry (verified New Zealand ID, the harm-minimisation suite, the regulator’s complaint door, the gambling levy, the Gambling Helpline in the apps). The reader who weighs that trade picks with eyes open.

The brand reviews

TAB

TAB is the established TAB NZ brand — the high-street outlet, the online tab.co.nz product, and the established parimutuel (tote) and fixed-odds book. It takes Win, Place, Each Way, Quinella, Exacta, Trifecta, First4, Double, Treble, Daily Double, Running Double, Quaddie and Big6, and Same Game Multis on the sports side. The brand’s USP, for a New Zealand punter, is that it is the lawful option for racing and the lawful option for sport in one account, which is the position the Racing Industry Act 2020 wrote it into. The first release date on the App Store for the “TAB – Racing & Sports Betting” app is 8 September 2014, and the seller is New Zealand Racing Board, which is the legal identity behind TAB NZ.

Worth it, and for whom: TAB is the right fit for the New Zealand punter who wants racing and sports on one account, with the regulator’s complaint door behind them, and who is not chasing a multi-vendor arb spread — the brand is alone in its lawful lane and has no need to outbid an offshore competitor on price. A punter whose edge is in picking winners will find every product they need; a punter whose edge is in price comparison finds nothing to compare against.

Betcha

Betcha is the TAB NZ online brand aimed at a younger, online-first audience. The App Store listing describes a daily Toolbox of odds boosts and insurance tools for unsettled bets, a Bet Social section where punters follow tipsters and share bets or pools, and a Playbook of tips and match analysis. The brands share one product licence and one self-exclusion list, so a punters’ exclusion on TAB carries to Betcha and back.

Worth it, and for whom: Betcha is the right fit for the New Zealand punter who bets with their phone, follows a tipster or two, and wants the social layer to be part of the product — the daily Toolbox is the differentiator against the parent TAB app, not a price edge. A punter who reads form from a coffee shop will use it; a punter who wants the deepest racing markets or the longest odds comparison will stay on TAB.

Glorion

Glorion is an offshore brand outside the New Zealand licence regime. The “About” page lists Live Chat and email support, an 18+ policy, and product information the materials carry without an explicit arbitrage clause. The brand is in the comparison set because New Zealand punters meet it, not because it has any lawful claim in New Zealand.

Worth it, and for whom: Glorion fits the offshore-curious punter who values support reach (Live Chat on the home page) over the regulator’s complaint door. A New Zealand punter who stakes here is exposed to the practical risks DIA warns about and cannot reach the New Zealand Gambling Helpline or Safer Gambling Aotearoa through any in-product link the operator owns.

BillyBets

BillyBets is an offshore sportsbook and casino with three odds formats (Decimal, American, Fractional) and a sportsbook with cashback and reload bonuses, a coin shop that converts wagering activity into free bets and free spins, and a five-level VIP programme. The brand’s combined sportsbook-and-casino shape is what it markets to; the materials do not name a specific arbitrage clause.

Worth it, and for whom: BillyBets is a fit for a punters’ profile that wants the conversion loops (coin shop, VIP levels) on top of the betting — the gamified side is the selling point, not the price. A reader who values a New Zealand regulator’s harm-minimisation toolkit does not get one through an offshore casino-led brand, and the coin-shop conversion is a casino-economics structure the New Zealand pair of brands does not run.

Roobet

Roobet is the brand name of Raw Entertainment B.V., registered at Korporaalweg 10, Curaçao, under Curaçao licence OGL/2024/687/0427. The brand carries a strong crypto identity and a sportsbook that is one product among several. The licence is Curaçao’s and is not a New Zealand licence; the brand names its jurisdiction plainly.

Worth it, and for whom: Roobet is the right fit for the punters’ profile that wants crypto rails (where the brand accepts them) and a Curaçao-licensed operator with a published company number, name and address — that much of the corporate identity is on the page. A New Zealand punter who stakes here accepts DIA’s stated risks (no consumer safeguards equivalent to New Zealand, no guaranteed payout of large wins, no return of unspent credits), and the harm-minimisation rails of the New Zealand pair do not apply.

22Bet

22Bet is an offshore operator under Curaçao licence 8048/JAZ (holder TechSolutions Group N.V.), with a sportsbook that takes single bets, accumulators, systems and chains, plus live betting with real-time odds and score updates. The product range is wide and the licence is publishable.

Worth it, and for whom: 22Bet is a fit for a punters’ profile that wants the four bet types a sportsbook offers (single, accumulator, system, chain) on one platform and prefers a Curaçao licence with a named holder. A reader who wants a multi-vendor arb spread will look here knowing the per-leg edge has to clear account-closure risk and the New Zealand regulatory protection does not extend.

Rabona

Rabona launched on 23 December 2019 as the sixth online sportsbook in the Soft2Bet group. It runs under a Curaçao licence and is a sportsbook-led brand in that group. The brand sits in the comparison because New Zealand punters meet it; the licence position is the same as the others — Curaçao, not New Zealand.

Worth it, and for whom: Rabona is a fit for a punters’ profile that recognises the Soft2Bet group, a multi-brand operator with a published launch date and a sportsbook-led product. The New Zealand reader who stakes here is on the same offshore terms as every other brand in that group, with the same DIA-flagged risks and no New Zealand regulator behind them.

Golden Panda

Golden Panda runs as an offshore sportsbook and publishes a dedicated Sports Terms and Conditions page that covers fixed-odds betting, Combo Bets, the Bet Builder feature, live scoreboards and broadcasts, and stake and odds limits. The dedicated terms page is the differentiator; the licence position is the standard offshore one (no New Zealand licence).

Worth it, and for whom: Golden Panda is the right fit for a punters’ profile that wants a sportsbook with a published dedicated sports terms page rather than a casino-led operator’s general T&Cs — the terms depth is the brand’s point of difference. A reader who wants the New Zealand harm-minimisation suite still does not get it, and a punters’ profile who cares about Bet Builder on a New Zealand basis will find TAB NZ offers the same product.

BetNinja

BetNinja is an offshore brand that states it is licensed by the Union of the Comoros. The brand runs a casino-led product with sports on the side; the materials do not surface a stand-alone arbitrage clause.

Worth it, and for whom: BetNinja is a fit for the punters’ profile who lands on a casino-first brand that also takes sports bets — the audience is one that came in through the casino side and the sportsbook is the secondary product. The licence jurisdiction is named (the Union of the Comoros) but does not extend to New Zealand; the New Zealand pair is what gambling-harm-minimisation rails attach to in this market.

FunBet

FunBet is an offshore sportsbook with a wide menu (football, ice hockey, American football, basketball, volleyball, tennis, rugby, baseball), live betting and a 24/7 bookmaker. The product range and the round-the-clock framing are the brand’s pitch; the licence position does not extend to New Zealand.

Worth it, and for whom: FunBet fits a punters’ profile who wants the breadth of an offshore sportsbook with a 24/7 claim — the breadth and the hours are the selling point. The punters’ profile who stakes here from New Zealand takes the same DIA-flagged risks as the rest of the offshore group, with no New Zealand regulator and no in-product harm-minimisation rails equivalent to TAB NZ.

A note on what the choice does

Picking the New Zealand pair means a quieter book

The punters’ profile who runs their book through TAB NZ and Betcha stays inside the New Zealand regulator’s view, sees the harm-minimisation toolkit in the app, and has the Gambling Helpline and Safer Gambling Aotearoa at hand if they want them. What the punters’ profile gives up is the multi-vendor spread an arbitrage strategy wants — by statute, not by accident.

Picking the offshore names means a busier book and a different risk

The punters’ profile who adds one or more offshore brands gains the multi-vendor spread and the fast crypto rails where the brand accepts them. What they take on is exposure to the offshore operator’s terms (including whatever voiding rules sit behind the brand’s “irregular betting” clause), the DIA-stated risks (no guaranteed payout, no return of unspent credits, minimal consumer safeguards), and the absence of the New Zealand regulator’s complaint door.

The arithmetic does not change

A 1% margin on a $10,000 cycle is the same number under both regimes. The deductions are different. Under TAB NZ the deductions are the operator’s house margin (close to 5% on a 1.90 / 1.90 market, less on longer prices) plus the friction of a New Zealand payment rail. Under an offshore book the deductions are the same per-side margin, the same friction, and additionally the per-leg account risk of stake restrictions or closure. The reader who weighs these picks with that arithmetic in view is the reader who treats arbitrage as a math problem with operational risk, not as a strategy with no downside.

Frequently asked questions

Does sports arbitrage betting work?

Sports arbitrage betting works as an arithmetic exercise: if the implied probabilities of every outcome of an event sum to less than 1, a stake on each outcome at those prices returns a small locked-in margin whichever result lands. What “works” does not include is escaping the bookmaker’s response — accounts flagged as arbers see max wager limits drop and accounts closed, with individual bets or halves of bets voided in some cases.

How is the return on a bet worked out from the stake and the odds?

Return on a winning bet = stake × odds; net profit = stake × (odds − 1). For decimal odds of 2.10 and a stake of $50, the return is $105 and the profit is $55; for fractional odds a reader converts first. The odds of an accumulator’s legs multiply into one price, and the same formula applies to the multiplied price.

What is the bookmaker’s margin and how can it be seen in the odds?

The bookmaker’s margin is 1 minus the payout share, where the payout share is 1 / (sum of 1/odds across every outcome). For a balanced two-way market at 1.90 / 1.90 the implied probabilities sum to about 1.0526, so the payout share is about 0.95 and the margin is about 5%. A sum below 1 across all outcomes is the line an arbitrage bettor is hunting for; the magnitude of the gap is the per-cycle margin.

How does an arbitrage bet get placed in practice?

In practice the bettor scans several bookmakers for the same event, identifies a price spread where the implied probabilities of every outcome sum to under 1, sizes each stake so the payouts match (or are within margin), places each leg in turn, and lets the cycle settle. The leg-by-leg race matters: the second leg is at the price at the moment the bettor clicks, and that price can move before the bet lands — the arb closes mid-cycle on a bad tick.

Which bookmaker may lawfully take sports and racing bets in New Zealand?

TAB NZ and its online sister brand Betcha, under section 74AAA of the Racing Industry Act 2020. The amendment came into force on 28 June 2025; there is no licence to apply for and no licence register for bookmakers in New Zealand.

Content created by the «dogracingnz.com» team