Introduction
At any given moment, Bitcoin might trade at $83,200 on Binance and $83,650 on Kraken. That $450 difference on a single coin is the foundation of crypto arbitrage: buy where it's cheaper, sell where it's more expensive, keep the difference after fees.
It sounds almost too simple — and in a way, it is. The mechanics are straightforward. The challenge lies in the details: fees, transfer speeds, liquidity, network availability, and execution timing. These factors are exactly what separate traders who profit consistently from those who break even or lose.
This guide covers everything from scratch: what crypto arbitrage is, why price gaps exist, which strategies work in 2026, how to calculate real profit, what risks to watch for, and how to get started.
Why Do Prices Differ Across Exchanges?
Before diving into strategies, it's worth understanding where price gaps actually come from. If you know why they exist, you'll be better at finding and evaluating them.
Every crypto exchange is an independent market with its own order book, its own set of buyers and sellers, and no central price-setting authority. Price is determined purely by supply and demand within each platform at each moment in time.
Regional demand differences. Asian exchanges (OKX, Gate.io, HTX) serve a predominantly Asian user base, while Western exchanges (Coinbase, Kraken) serve European and American traders. Demand patterns differ across regions, especially during local trading hours.
News and reaction speed. When major news breaks, some exchanges — and their users — react faster than others. For a few minutes or even seconds, prices diverge noticeably across platforms.
Liquidity depth. High-volume exchanges have deep order books that absorb trades without significant price impact. On smaller exchanges, a single large order can move the price noticeably, creating a temporary gap versus the broader market.
Technical delays. Data feed latencies, API rate limits, and occasional technical issues create brief price anomalies on specific exchanges.
This is why price gaps never fully disappear — new ones form continuously as markets react to information at different speeds and with different liquidity profiles.
How Crypto Arbitrage Works: Step-by-Step Example
Let's walk through a complete real-world example using ETH/USDT.
The opportunity:
- Binance: ETH = $2,480
- KuCoin: ETH = $2,497
- Spread: $17 (0.69%) Step 1: Spot the opportunity
A scanner like SpreadScan shows that the ETH/USDT spread between Binance and KuCoin is currently 0.69%. That's above the average noise level — worth investigating.
Step 2: Calculate all costs
Before executing anything, calculate every cost:
- Binance taker fee: 0.10% of trade value
- KuCoin taker fee: 0.10% of trade value
- ETH withdrawal from Binance: ~$1.50 on Ethereum mainnet, or ~$0.01 on Arbitrum/Optimism
- Total trading fees: ~0.20% Remaining spread after fees: 0.69% − 0.20% = 0.49% net profit potential
On a $5,000 position, that's approximately $24.50.
Step 3: Check network availability
This step is critical and often overlooked by beginners. Before committing, verify:
- KuCoin deposits for ETH are open (network not under maintenance)
- A fast, cheap network is available (Arbitrum or Optimism instead of Ethereum mainnet)
- Estimated transfer time is acceptable (L2 networks typically take 1–3 minutes) Step 4: Execute
- Buy 2 ETH on Binance for $4,960 (2 × $2,480)
- Withdraw ETH to KuCoin via Arbitrum network (~$0.01 fee, ~3 minutes)
- Sell 2 ETH on KuCoin for $4,994 (2 × $2,497) Step 5: Final result
- Revenue: $4,994
- Costs: $4,960 + $4.96 (trading fees 0.1% × 2) + $0.01 (network) = $4,964.97
- Net profit: $29.03 (~0.58%) That's one cycle. Finding 3–5 similar opportunities per day across different pairs, the cumulative return becomes meaningful.
Types of Crypto Arbitrage
There are several fundamentally different approaches, each requiring different capital, skills, and risk tolerance.
1. Cross-Exchange (Spatial) Arbitrage
How it works: same coin, two different exchanges. Buy low on one, sell high on the other.
This is the most common type and where most people start. The main requirement: you either need capital pre-positioned on both exchanges, or you need to transfer assets between them faster than the spread disappears.
Two execution approaches:
With coin transfer: buy ETH on Exchange A, withdraw to Exchange B, sell. Risk: the price on B may drop while your transfer is in flight.
Balance arbitrage (no transfer): hold USDT on Exchange A and ETH on Exchange B simultaneously. When you spot a spread — buy on A (USDT → ETH) and simultaneously sell on B (ETH → USDT). No transfer needed, both trades execute instantly. Afterward, you rebalance at your convenience. This is the professional approach but requires capital locked on multiple exchanges.
Where to find opportunities: SpreadScan monitors spreads across 600+ pairs on 17 exchanges in real time.
2. Triangular Arbitrage
How it works: three coins, one exchange, a closed loop.
Example route: USDT → BTC → ETH → USDT
If the exchange rates between these three pairs are slightly misaligned, you end up with more USDT than you started with. The difference is pure profit.
Real example:
- USDT → BTC: buy 0.1 BTC for $8,300
- BTC → ETH: exchange 0.1 BTC for 3.35 ETH
- ETH → USDT: sell 3.35 ETH for $8,341 Gross profit: $41 (~0.49%)
After three trading fees (3 × 0.1% = 0.3%), net profit is approximately 0.19%, or about $15.80.
Key advantage: no cross-exchange transfers needed. Everything happens on one platform in seconds. This makes triangular arbitrage especially attractive for traders who want to avoid transfer risks and wait times.
Challenge: opportunities are fleeting — often lasting just seconds. SpreadScan's triangular scanner continuously searches for profitable routes across exchanges.
3. P2P Arbitrage
How it works: exploit the difference between P2P platform rates and the spot market.
P2P markets (Binance P2P, Bybit P2P) let users set their own prices, which often diverge from the spot market rate — especially in non-major fiat currencies like RUB, KZT, AED, or BRL.
Example: USD/RUB spot rate is 88.5 RUB/USD. On Binance P2P, you find a seller offering USDT at 87.2 RUB/USD. Buy USDT via P2P at the lower rate → sell on spot → pocket the ~1.5% difference.
P2P arbitrage is often more accessible for beginners since execution doesn't require millisecond timing — P2P trades take minutes, not fractions of a second.
4. CEX-DEX Arbitrage
How it works: exploit price gaps between a centralized exchange (CEX) and a decentralized exchange (DEX) liquidity pool.
For example: ETH trades at $2,480 on Binance but at $2,491 in a Uniswap pool on Arbitrum. Difference: 0.44%. Buy on Binance, bridge to wallet, sell on Uniswap.
The main challenge is gas fees (especially on Ethereum mainnet) and execution speed. This type of arbitrage is most often automated. SpreadScan tracks CEX-DEX spreads across 7 blockchains.
How to Calculate Real Profit (Full Formula)
One of the most common beginner mistakes: looking only at the spread and forgetting about costs. Here's the complete calculation framework.
Net profit formula:
Profit = (Sell price − Buy price) − Buy fee − Sell fee − Network fee − Slippage
Full example with numbers (position: $10,000, pair: SOL/USDT):
| Parameter | Value |
|---|---|
| Buy price (Gate.io) | $143.20 |
| Sell price (OKX) | $144.85 |
| Gross spread | 1.15% |
| Trading fee — Gate.io (taker) | 0.10% = $10.00 |
| Trading fee — OKX (taker) | 0.10% = $10.15 |
| Network fee (Solana) | ~$0.01 |
| Slippage estimate (~0.05%) | ~$5.00 |
| Total costs | ~$25.16 |
| Net profit | ~$89.84 (0.90%) |
Even with a solid 1.15% gross spread, actual costs reduce the take-home significantly. This is why only working with spreads meaningfully above your total cost structure matters — most experienced traders set a minimum threshold of 0.5–0.7% net spread before considering a trade.
Key Risks in Crypto Arbitrage (And How to Manage Them)
Arbitrage isn't risk-free despite its reputation for being "market-neutral." Here are the main risks and practical ways to handle each.
Execution Risk
Between the moment you see a spread and the moment your trades fill, prices can change. On volatile markets, a 1% spread can vanish in seconds.
How to manage: use balance arbitrage (no transfers required), trade high-liquidity pairs (BTC, ETH, SOL), avoid high-volatility periods like major news events or exchange listings.
Transfer Risk
When doing cross-exchange arbitrage that involves moving coins, transfer time is your biggest enemy. ETH on the Ethereum mainnet can take 10–15 minutes to confirm — more than enough time for a spread to disappear.
How to manage: use fast networks (Solana: <1 second, Arbitrum/Optimism: 1–3 minutes), always verify network availability on both exchanges before initiating a transfer.
Liquidity Risk
An attractive spread in a scanner can be illusory. If the exchange doesn't have real depth at the displayed price, you'll buy or sell with significant slippage that erases your profit.
How to manage: always check order book depth, don't take a position size larger than 10–15% of the visible book depth at your target price.
Network Availability Risk
An exchange may suspend withdrawals for a specific coin or network — often during technical maintenance. Your coins can get stuck for hours.
How to manage: check withdrawal status on both exchanges before executing. SpreadScan displays network availability and withdrawal status in real time.
Regulatory and Tax Risk
In most jurisdictions, arbitrage profits are taxed as trading income. Some countries have specific crypto tax reporting requirements. Keep a detailed trade log from day one.
Step-by-Step Plan for Beginners
If you're new to arbitrage, here's a practical roadmap — no fluff.
Step 1: Learn the tools (1–2 days)
Sign up for SpreadScan and explore the interface. The free tier gives access to public spreads. Watch how they change throughout the day — which pairs show consistent gaps, what times are most active, how quickly opportunities disappear.
Step 2: Open accounts on 2–3 exchanges (1–2 days)
Recommended starting set: Binance + Bybit + OKX. These are the largest exchanges by volume, with broad coin selection, good liquidity, and support for most popular withdrawal networks.
Complete KYC (identity verification) immediately — without it, withdrawal limits will be severely restricted.
Step 3: Paper trade for one week
Don't risk real money yet. Each day, open SpreadScan and log every opportunity you'd take: what the spread was, what the estimated profit would be after all fees, and what actually happened to the price 10 minutes later.
After a week, you'll have a realistic picture: how many opportunities appear per day, which pairs are most active, what your realistic return would have been.
Step 4: First real trades with minimal capital
Start with $200–500. The goal of your first 10–20 trades is not to make money — it's to understand the mechanics in practice: how quickly orders fill, how withdrawals work, where delays happen.
Step 5: Scale methodically
Only after 20–30 successful small trades does it make sense to increase position sizes. By then you'll have real data on your actual costs, execution times, and which strategies work best in your hands.
Common Mistakes Beginners Make
Forgetting fixed withdrawal fees. Many beginners calculate only percentage trading fees (0.1% + 0.1%) and overlook the flat withdrawal fee. On small trades, this can eliminate all profit. For example, a USDT withdrawal on TRC20 costs 1 USDT — on a $5 spread trade, that's 20% of your profit gone immediately.
Trading illiquid altcoins. A large spread on a low-volume coin is almost always a trap. You won't be able to fill your order at the displayed price.
Not checking network availability. Imagine buying a coin and discovering that withdrawal through your target network is suspended. Your capital is stuck.
Chasing pumps and dumps. When a coin is moving violently, spreads between exchanges look massive — but this isn't arbitrage, it's directional risk. The price can reverse sharply while you're mid-trade.
No trade journal. Without records, you can't improve. Log every trade: entry prices, exit prices, fees, any issues. The data is valuable.
Realistic Earnings Expectations
What can you actually make? Here's an honest breakdown:
- Beginner with $1,000–2,000, manual arbitrage: $50–150/month (realistic with consistent daily monitoring)
- Experienced trader with $10,000+, multiple strategies: $300–800/month
- Professional with automation and $50,000+: significantly more, but this requires substantial technical setup Important context: arbitrage is not passive income. It requires active time for monitoring and execution. Consistent profitability comes from treating it like a business — systematic, process-driven, with good record-keeping.
Conclusion
Crypto arbitrage is a legitimate, market-neutral strategy for profiting from price inefficiencies. It doesn't require predicting whether the market will go up or down — it exploits the fact that the same asset trades at different prices across different venues simultaneously.
That said, success requires being realistic about costs and risks. Spreads are small, fees compound, and opportunities are time-sensitive. The edge isn't in finding the biggest spreads — it's in consistently executing on well-analyzed opportunities while keeping costs under control.
SpreadScan is built for exactly this: aggregating real-time data from 17 exchanges, accounting for network fees and availability, and presenting a complete picture of each opportunity — not just the raw spread number.
Start with SpreadScan for free →
Risk disclaimer: Cryptocurrency trading involves significant risk and may result in loss of funds. This article is for educational purposes only and does not constitute financial advice.
Frequently Asked Questions
Is crypto arbitrage legal? Yes, in the vast majority of jurisdictions. It's a legal trading strategy. Profits are typically taxed as trading income. Check the tax rules specific to your country.
How much capital do I need to start? Technically you can start with $200–300, but meaningful returns require at least $1,000–2,000. Below that, flat withdrawal fees consume too large a share of profit.
Can I automate crypto arbitrage? Yes — many experienced traders build bots for automated execution. But starting manually is strongly recommended. It builds understanding of the real market mechanics that you'll need to design and troubleshoot automated systems later.
Does arbitrage work in a bear market? Yes. Cross-exchange and triangular arbitrage are market-direction neutral — they exploit price differences between venues, not price movement itself. Spreads actually tend to be larger during volatile periods.
Why don't arbitrage spreads disappear completely? Arbitrageurs do compress them — but new ones form continuously, driven by regional demand differences, news reaction lags, liquidity asymmetries, and technical delays. Markets are never perfectly efficient.
What's the minimum spread worth trading? As a rule of thumb, the gross spread should exceed your total cost structure (trading fees + network fees + estimated slippage) by at least 0.3–0.5% to leave a meaningful net profit after execution risk.