CS2 Arbitrage Betting: How Arbs Work and Why They're Harder Than They Look
Updated · 
Written by Milan Kovac ·
Every so often you'll see two bookmakers price the same CS2 match very differently. One has Team A at 2.10, another has Team B at 2.05. Put money on both at the right ratio and you win either way. That's the whole idea behind arbitrage betting, or "arbing."
It sounds like free money, and in the maths it is. In real life, it's a grind with thin margins and a few ways to lose that nobody mentions in the YouTube thumbnails. Let's be straight with you, bro: arbing works on paper far more often than it works in your wallet.
This guide covers what a CS2 arb is, how to spot one, how to split your stakes, a couple of worked examples, and the real risks, including the one CS2 bettors forget the most: different void rules at different books.
How we wrote this: the maths below is standard implied-probability arithmetic you can check with any calculator. The example prices are made up to show the method, not real lines. Notes on void rules come from published esports rules (Pinnacle's own rules page and a third-party read of Cloudbet's rules in October 2026). Always read your own bookmaker's rules before betting. Nothing here guarantees a profit.
18+ only. Betting is real money and you can lose it. Responsible gambling.
Claim free cases on Clash.gg, no code needed
What arbitrage betting actually is
A bookmaker sets prices so that, across all outcomes, it keeps a small margin. If you add up the implied chances of every outcome at one book, you'll get a bit more than 100%. That extra is the book's cut.
But different books don't always agree. One might be slow to react to a roster rumour. Another might be shading its line because lots of fans backed the popular team. When the *best* price for each outcome comes from different books, the combined total can dip under 100%. That's an arb.
In a CS2 match winner market with no draw, there are only two outcomes. So you need two bets: Team A at one book and Team B at another. In a market with three outcomes, like a best-of-two group game that can end 1-1, you need three bets.
How to spot a CS2 arb
You don't need software to check. You need one formula.
Implied probability = 1 ÷ decimal odds.
Do that for the best price on each outcome and add the results.
- Total over 1.00: no arb. Normal market.
- Total exactly 1.00: break-even before fees. Not worth it.
- Total under 1.00: there's an arb. The gap is your rough margin.
If your book shows American odds, convert first. For positive odds, decimal = (odds ÷ 100) + 1. For negative odds, decimal = (100 ÷ the number without the minus) + 1. So +110 is 2.10 and -150 is about 1.67.
Quick check: a normal market
Say one book prices a match at 1.85 and 1.95.
- 1 ÷ 1.85 = 0.5405
- 1 ÷ 1.95 = 0.5128
- Total = 1.0533
That's 5.3% over 1, which is the book's margin. No arb here.
Quick check: an arb
Now say Book X has Team A at 2.10 and Book Y has Team B at 2.05.
- 1 ÷ 2.10 = 0.4762
- 1 ÷ 2.05 = 0.4878
- Total = 0.9640
That's under 1. The margin is roughly 1 − 0.9640 = 0.036, or about 3.6% of your total stake, before anything goes wrong.
How to split your stakes
Finding the arb is half the job. The other half is splitting your money so you get the same payout whatever happens.
Stake on an outcome = total stake × (that outcome's implied probability ÷ the total).
Using the example above with a total of $100:
| Bet | Odds | Stake | Payout if it wins |
|---|---|---|---|
| Team A at Book X | 2.10 | $49.40 | $103.74 |
| Team B at Book Y | 2.05 | $50.60 | $103.73 |
| Total | $100.00 | about $103.73 either way |
So you put $100 in and get about $103.73 back whichever team wins. That's a $3.73 profit, if both bets go through at those prices and settle normally.
Notice how small it is. You tied up $100 across two accounts to make under four dollars. That's what most arbs look like.
A three-way example
Some CS2 group stages use best-of-two matches that can end in a draw, so books offer a three-way line. Made-up best prices:
- Team A win at Book X: 2.45 → 0.4082
- Draw at Book Y: 4.10 → 0.2439
- Team B win at Book Z: 3.25 → 0.3077
- Total = 0.9598
That's about a 4% margin, but now you need three accounts, three bets placed fast, and three books that all settle the same way. Every extra leg is another chance for something to break.
CSGOEmpire match betting + free case with FREECASES4U
Why CS2 arbs exist at all
Esports markets are smaller and less settled than football or basketball. That's why gaps show up. Common reasons:
- Late roster news. A stand-in is announced, or a star player is rumoured to be sick. Some books move fast, others don't.
- Fan money. Big-name teams attract a lot of casual bets, so some books shorten those teams more than the form justifies.
- Tier 2 and smaller events. Fewer people trade these lines, so prices drift further apart.
- Live betting. In-play prices change round by round. Different books update at different speeds.
- Different market rules. Sometimes what looks like an arb isn't one, because the two books are really offering slightly different bets (more on that below).
The real risks of CS2 arbitrage betting
This is the part that matters. An arb is only risk-free if *every* leg goes through at the price you saw and settles the way you expect. Here's what usually goes wrong.
1. The odds move before your second bet
You place Team A at 2.10. By the time you open the other site, Team B has dropped from 2.05 to 1.90. Now:
- 1 ÷ 2.10 + 1 ÷ 1.90 = 0.4762 + 0.5263 = 1.0025
The arb is gone and you're holding a normal bet with a slight loss built in. You either take a small guaranteed loss on the second leg or leave the first bet running. This is the most common way arbers lose money.
2. Your bet gets cut or rejected
Books can cap your stake, ask you to accept a new price, or reject the bet. If one leg only takes $20 instead of $50, your stakes are no longer balanced.
3. The two books settle the match differently
This one is big in CS2. Esports rules aren't the same at every book. A couple of real examples of how they can differ:
- One map voided. Pinnacle's esports rules say that if a map is voided because of retirement, disconnection, disqualification, walkover or another admin decision, all bets on the match period are void. A third-party review of Cloudbet's rules in October 2026 reported the same approach there. But that review also said some other books settle the match winner on the official result instead. If your two legs sit at books that do it differently, one bet can be refunded while the other loses.
- Postponements. Pinnacle voids an esports match that hasn't started 30 hours after its scheduled time. The same review reported Cloudbet uses 30 hours too, while other books use 24 hours or less. A match that starts 26 hours late could be live at one book and void at the other.
- Format changes. Many books void match bets if the number of maps changes, say a Bo3 cut to a Bo1. If only one of your books treats it that way, your arb breaks.
- Stand-ins and overtime. Most books say stand-ins don't void bets, and most count overtime in match and map winner markets. But "most" isn't "all." Read both.
Our full breakdown is in CS2 betting void rules and CS2 overtime betting rules.
4. Account limits and closures
Bookmakers make money from margin, and arbers take that margin away. So most books watch for it. Patterns like always betting the best price in the market, odd stake amounts like $49.40, and only betting when prices are out of line can get an account limited to tiny stakes or closed. Many books' terms give them the right to do this. Before you start, read the terms on any book you use, and assume a profitable arbing account won't stay unrestricted forever.
5. Fees, currency and rounding
A 2% arb can vanish because of:
- Deposit or withdrawal fees
- Currency conversion between accounts
- Crypto network fees and price swings if you hold a coin like BTC instead of a stablecoin
- Rounding your stakes to whole numbers
Run the numbers on the money that actually lands back in your wallet, not the payout on the bet slip.
6. Palpable error rules
If a book posts a price that's clearly wrong (like the favourite at 5.00 by mistake), its terms usually let it void or resettle the bet at the correct price. Those "too good to be true" arbs are often exactly that. Your other leg stands, and now you have an unbalanced bet.
A safer way to think about it
If you're curious about arbing, treat it as a maths exercise first, not a money-making plan.
- Paper trade for a couple of weeks. Write down arbs you find, the time, and whether both prices were still there a few minutes later. You'll learn fast how often they vanish.
- Use books whose rules you've actually read. Compare void, postponement, stand-in and format rules side by side before betting both sides.
- Place the harder leg first. Usually that's the price most likely to move or the book most likely to cut your stake.
- Keep stakes small. Only use money you can afford to lose, because sometimes you will lose part of it.
- Track every bet. Use a sheet like our CS2 bet tracker so you see real profit after fees, not imagined profit.
- Don't chase. If a leg fails, take the small loss. Doubling up to fix it is how a $4 arb becomes a $100 problem.
Arbing vs hedging vs value betting
People mix these up, so here's the quick difference:
| Approach | What you do | Profit if done well | Main risk |
|---|---|---|---|
| Arbitrage | Back every outcome at once, at different books | Small, locked in | A leg failing or settling differently |
| Hedging | Bet the other side later to cut risk on an earlier bet | Varies, often smaller than the original upside | Giving away value to feel safe |
| Value betting | Back one side when you think the price is too high | Uncertain, long-term only | Losing streaks, being wrong |
If you want the hedging side, read CS2 hedge betting. For how prices are built in the first place, see how CS2 odds are calculated.
Where CS2 bettors usually go wrong with arbs
- Thinking it's risk-free. It's low-risk only when everything works. It often doesn't.
- Ignoring rule differences. Two "match winner" markets aren't always the same bet.
- Betting on locked-in edges smaller than fees. A 0.8% arb doesn't survive a 1% withdrawal fee.
- Using a VPN or someone else's details to open more accounts. Don't. Books can void bets, keep winnings and close accounts, and it can break the law where you live.
- Forgetting the time cost. Watching odds for hours to make a few dollars isn't a great hourly rate.
Is CS2 arbitrage betting worth it?
For most people, honestly, no. Arbs exist and the maths is real, but margins are thin, gaps close in minutes, rules differ between books, and accounts that win this way tend to get limited. Learning the method still makes you a sharper bettor, because it teaches you to read implied probability and compare prices, which helps with everything else.
If you just want to enjoy CS2 matches with a small bet on the side, compare prices, read the rules, set a budget and keep it fun.
Try Clash.gg free cases, no code needed
18+ only. If betting stops being fun, take a break and get help. Responsible gambling.