Live Arbitrage Betting: How In-Play Arbs Really Work

Live arbitrage betting means backing every outcome of a market while the game is running, at prices from different sportsbooks that add up to less than 100 percent. The gaps are wider and far more frequent in-play than they are pre-match. They also die in seconds rather than minutes, which is the whole story: live arbitrage is not a harder math problem than pre-match arbitrage, it is the same math under a stopwatch.

That changes what you need to be good at. The number on the board matters less than how much adverse movement it can absorb before it stops being an arb at all. Start there.

Try It: How Much Movement a Live Arb Can Absorb

Your first leg is down. The second book is still showing its price. Drag the slider to see what a small move against you does to a live margin before your second bet is accepted.

The default is an illustrative live arb, not a real result: +105 on one side, +102 on the other, which works out to a 1.7 percent margin on a $500 stake if both legs fill at those prices. Nudge the slider to a single point of implied probability and the worst case flips negative. That is the lesson in one drag. A live margin of one or two percent tolerates well under a point of movement, so every second between your two clicks is spending the edge.

Please note: every figure in this guide is illustrative and calculated from the prices shown, not a typical or expected result. Matched bets are never truly risk free, results vary, and only disposable income belongs in a betting bankroll.

What Live Arbitrage Betting Is

An arbitrage opportunity is a complete set of outcomes whose best available prices sum to under 100 percent of implied probability. Convert each price to an implied probability, add them up, and if the total is below 100 the shortfall is your margin. That test is the whole definition, and it is unpacked properly in the guide to what an arbitrage opportunity is.

Live arbitrage betting applies that test to in-play markets. Nothing about the arithmetic changes. The moneyline is still two outcomes, the total is still over and under, and the prices still either sum under 100 or they do not.

What changes is everything around the arithmetic. Pre-match lines are set days ahead and converge slowly as money arrives. In-play lines are regenerated every few seconds by each book's own model, and they never get that convergence time.

Here is why that matters so much.

Why In-Play Markets Produce More Arbitrage Opportunities

Pre-match, sportsbooks have days to watch each other. A price that drifts out of line with the market gets noticed and corrected, which is why pre-match arbs are rare and thin. Live, nobody has days. Every book is recalculating from scratch on a clock.

Four things pull in-play prices apart, and they compound:

• Different models. Each book runs its own in-play pricing model. Two models fed identical facts still disagree on what a 7 point lead with 4 minutes left is worth.
• Different data feeds. Books buy their live data from different vendors with different latencies. A book on a slower feed is pricing a slightly older game state, and for a few seconds it is simply wrong.
• Different suspension policies. Some books suspend a market on every dangerous attack or every possession change. Others leave it up. When one book freezes and the other keeps pricing, the two are describing different moments.
• Different re-price speeds. After a scoring play, one book might re-price in under a second and another in five. That lag is where most live arbs live.

The gap is usually not one book being wrong and another being right. It is the interval between the first book to react and the last. Understanding that reframes the hunt: you are not looking for a mispriced book, you are looking for a market caught mid-update.

The same forces make in-play margins wider than pre-match margins. Books charge more vig in-play precisely because they know they are pricing fast and imperfectly, and if you want to see how that vig is stripped back out to a fair number, the devigging guide and the devigging calculator cover the mechanics. Wider vig at each book plus wider disagreement between books is why live arbs show bigger headline percentages than pre-match ones.

Why the Window Is Seconds, Not Minutes

That reaction lag is also the arb's lifespan. Once the slower book catches up, the two prices agree and there is nothing left to take.

You can see how differently the two markets age inside SmartStake itself. The live board keeps a matched bet only while it stays fresh, dropping anything it has not heard about in 60 seconds. The pre-match board holds one for a full hour. A pre-match arb is a standing offer; a live arb is a moment.

The board reflects the death of an arb explicitly, too. When a live pair's two prices stop covering each other or a line is pulled, the pipeline sends a record retiring that specific pair rather than waiting for it to age out. The board is a stream of arbs being born and killed, not a list you scroll.

That is the mechanism. Now the part that costs money.

The Costs Live Arbitrage Adds

Every risk in pre-match arbitrage exists in-play, plus a set that only appears once the ball is in the air. These are the ones that turn a clean margin into a loss:

• Acceptance delay. Most books deliberately hold an in-play bet for a few seconds before confirming it. During that hold you have committed and the price has not. This is the single biggest structural cost of live arbing.
• Re-offer at changed odds. Rather than reject outright, many books come back with the new price and ask you to accept. The widget above models this scenario: your bet fills, but not at the number you clicked.
• Mid-placement suspension. The market freezes between your first and second click. You now hold one side and cannot place the other until it reopens, at whatever price it reopens at.
• Partial fill. Some books accept a smaller stake than you asked for. Your stake split is now wrong, and the two sides no longer pay out evenly.
• Void and settlement mismatches. Two books can grade the same in-play market differently, especially on player props after an early injury or on markets voided for a data error. One leg voided and the other graded leaves you with an open position you never chose.
• Lower limits. In-play maximums tend to start below pre-match maximums, so the same margin returns less cash and the account attention arrives sooner. Managing that is its own discipline, covered in the guide to handling sportsbook limits.

None of this makes live arbitrage betting unworkable. It does mean the calculated margin is a best case that assumes both legs fill at the prices shown, and treating it as anything firmer is how people lose money on a strategy they believed could not lose.

What Happens When Only One Leg Fills

This is the failure mode that matters, so plan for it before it happens rather than in the four seconds you will actually have.

When the second leg is rejected, suspended, or re-offered at a price that kills the margin, you hold a straight bet. You have three responses, and only one of them can be prepared in advance.

Hedge at the current price. Take whatever the other side now offers and accept the smaller margin, or a small loss. Run the numbers in the widget above or the arbitrage calculator before you fire, because the honest answer is often that a small loss is the best outcome available.

Leave it as an outright bet. Sometimes the position you are stuck with is one you would take on its merits, and holding it is a legitimate choice. It stops being legitimate the moment it becomes the thing you do because you froze.

Size so a naked leg is survivable. This is the one you control. If a single unhedged leg would hurt your bankroll, the stake was too big for a live arb regardless of how good the margin looked. The unit sizing in the bankroll management guide applies here with more force than usual, because live arbing generates far more of these moments than pre-match arbing does.

The formula behind the widget's absorbable movement is worth keeping in your head. With $A$ as the two prices added together and $p_2$ as the implied probability of the second leg, the movement the margin can absorb before the worst case turns negative is:

For a two-way market priced near even money, that lands at roughly half your margin in percentage points. A 2 percent arb absorbs about a point. That is your entire error budget.

A Practical Live Arbitrage Workflow

Speed here is preparation, not reflexes. Almost all of the time you can save is saved before the opportunity appears.

1. Fund and log into both books first. A deposit, a password reset, or a two-factor prompt costs more time than any live arb has.
2. Open both bet slips before the alert. Have the game up at both books, not the lobby.
3. Know your stake split in advance. Decide the total stake for a live arb once and let the calculator do the split, rather than working it out mid-market.
4. Place the harder leg first. Fire at the book with the stricter acceptance, lower limits, or more aggressive suspension. If that leg fails you have lost nothing. If you lead with the easy leg instead, a failure on the hard one leaves you exposed on the side you already committed to.
5. Confirm acceptance, then fire the second. Not the click, the confirmation. A pending bet is not a bet.
6. Let alerts find the arb. Nobody wins a seconds-long race by scrolling a board. Set the filters and let the notification arrive.
7. Log both legs immediately. Two independent bets that need to be reconciled later, and reconstructing them from memory after a busy slate does not work.

Point 4 is the one most people get backwards, and it is free to fix.

Live Arbitrage Versus Live Middles

A middle is the close cousin. Instead of two prices that overlap so each side covers the other, a middle is two prices with a gap between them where both bets can win, at the cost of a small loss when the result lands outside the window.

Live markets create middles constantly because totals and spreads move with the score. Bet the over at one number early, and 20 minutes of low scoring can leave the under available several points higher at another book.

The tradeoff is honest: a middle risks a small, defined loss most of the time in exchange for a large payoff when the result lands in the window, while an arb targets a small margin either way. Live conditions produce more of both, and the same execution problems apply to each.

Does Live Arbitrage Beat Pre-Match Arbitrage?

Live arbitrage betting produces bigger and more frequent margins than pre-match arbitrage, and it costs you more to collect them. That is the honest trade, and which side of it wins depends entirely on your execution rather than on the strategy.

Pre-match suits you if you want to place calmly, use higher limits, and treat arbing as a scheduled activity. Live suits you if you can sit on a game with two books open and act inside a few seconds without hesitating.

If neither sounds appealing, the same price disagreements feed strategies with far gentler clocks. Positive EV betting takes one side of a mispriced market instead of both, so a single slow click costs you nothing, and line shopping captures a smaller version of the same edge on bets you were already placing. Arbitrage is not the only thing a market disagreement is good for.

Finding Live Arbs Without Watching Ten Books

Manually refreshing in-play pages at several sportsbooks does not work, because the thing you are hunting exists for less time than a page takes to load and compare.

The SmartStake Arbitrage Finder has a Live mode that streams in-play matched bets as they appear, ranked by margin with the best at the top, and retires each one once the two prices stop covering each other. Set a minimum margin, filter to the books you actually hold accounts at, and save the filter so alerts reach you rather than the other way around. The tutorial walks through the controls, and the tool comparison in best live arbitrage betting software covers how the in-play options differ on speed and placement.

For the math itself, the arbitrage calculator gives you the stake split for any two prices, which is the number you want computed before the market moves, not during.

The Bottom Line

Live arbitrage betting is pre-match arbitrage with the preparation moved to the front and the margin for error taken away. The gaps are real, they are wider than pre-match gaps, and they exist because books re-price at different speeds off different feeds rather than because anyone is careless.

Judge a live arb by how much movement it can absorb rather than by its headline percentage, place the harder leg first, and size it so a single unfilled leg is an annoyance rather than a problem. That is the whole discipline, and it is not risk free even when it goes right.

Scan live arbs on the Arbitrage Finder and see what the in-play board looks like during a game.

Frequently Asked Questions

What is live arbitrage betting?

Live arbitrage betting is backing every outcome of a market while the game is running, at prices from two or more sportsbooks that add up to less than 100 percent of implied probability. The idea is identical to pre-match arbitrage. What changes is the clock. In-play prices are recalculated every few seconds off each book's own model and its own data feed, so the disagreement that creates the gap is both more common and much shorter lived. A live arb that lasts a minute is a long one, and many last only a few seconds.

Is live arbitrage betting better than pre-match arbitrage?

Live arbitrage betting produces wider and more frequent margins than pre-match, but it converts a math problem into an execution problem. In-play markets suspend without warning, in-play bets sit in an acceptance delay before a book confirms them, and in-play limits tend to be lower than pre-match limits. Pre-match arbitrage gives you minutes to place both legs calmly at higher stakes. Live arbitrage gives you seconds at smaller stakes for a bigger headline number. Neither is without risk, and the right answer depends on whether you can reliably get two bets down inside that window.

Why do live arbitrage opportunities disappear so fast?

A live arbitrage opportunity exists only while two books disagree, and in-play books re-price constantly. Every scoring play, injury, red card, or timeout triggers a recalculation at each book, on its own model, off its own data feed, at its own speed. The gap is usually the lag between the first book to re-price and the last. Once the slower book catches up, the prices agree again and the arbitrage is gone. The SmartStake live board reflects this directly: a live matched bet drops off if nothing has been heard about it for 60 seconds, against a full hour for a pre-match one.

What happens if only one leg of a live arb gets accepted?

You are left with a straight bet on one side, which is the single biggest risk in live arbitrage betting. You have three options. Hedge immediately at whatever the other side is now offering, which usually means accepting a smaller margin or a small loss. Leave it as an outright position, which is a real gamble and should be a deliberate choice, not a default. Or size every live arb small enough that a naked leg is survivable. Sizing is the only one of the three you can decide before it happens, which is why it matters most.

Do sportsbooks limit live arbitrage bettors?

Yes. Sportsbooks monitor for the betting patterns arbitrage produces, and they respond by cutting maximum stakes, adding manual review to your bets, or restricting the account. In-play limits tend to start lower than pre-match limits, so live arbitrage bettors often feel the squeeze sooner. Nothing about this is a rule violation, but it is a business decision the book is free to make, and planning around it is part of the strategy rather than an afterthought.