86 cross-exchange arbitrage opportunities detected on May 16, 2026. Top spread: UP at 20.99% between Gate Futures and OKX. Coinbase consistently priced above Binance and Bybit across multiple assets. Detailed breakdown of executable setups, fee-adjusted profit estimates, and risk factors for professional arb desks.
May 16, 2026 delivered a genuinely busy session for arbitrage desks. The scanner logged 86 total cross-exchange opportunities across the monitored universe — a number that sits comfortably above the weekly average and suggests meaningful fragmentation between venues, particularly between Coinbase and the Asian-origin books (Binance, Bybit, Gate). If you were watching screens today you had material to work with. If you were not, this is the debrief.
The headline figure is a 20.99% spread on UP — a futures-to-spot dislocation between Gate Futures (buy at $0.203900) and OKX (sell at $0.220400). That kind of spread on a futures-involved leg demands immediate scrutiny: funding rates, open interest, and withdrawal pipeline all need to be checked before you size anything meaningful. But the number is real, and for a desk with existing OKX spot inventory and Gate Futures access, the setup was live.
Below UP, the table stays thick with double-digit spreads. OP printed 13.50% between Bybit Spot and Coinbase, then again at 11.35% between Binance and Coinbase — two separate legs on the same asset within the same session, which tells you Coinbase's OP order book was consistently elevated relative to both major Asian venues. XLM showed 12.47% and ATOM 11.88%, both curiously as intra-Coinbase spreads — almost certainly spot-versus-perpetual or different contract maturities on the same platform. BEAM, TST, and the AI token trilogy (three separate entries at 10.94%, 10.79%, and 10.53%) rounded out a top ten that any arb scanner operator would call a productive day.
The pattern reading: Coinbase is running elevated across a cluster of mid-cap assets versus its Asian counterparts. This is not unusual — Coinbase's retail depth tends to spike when U.S. session momentum kicks in and order flow is directional rather than arb-balanced. But the persistence across OP (twice), AI (three times), XLM, ATOM, and BEAM in a single day suggests this is structural for the session, not a one-off glitch. Desks positioned to move from Binance or Bybit into Coinbase had a repeatable, multi-asset setup today.
Opportunity 1 — UP, 20.99% spread. Buy side: Gate Futures at $0.203900. Sell side: OKX at $0.220400. This is the session's outlier, and outliers require the most skepticism. A 20.99% spread on a futures leg immediately raises the question of whether the Gate contract is a dated future with time decay baked in, or a perpetual sitting under abnormal funding pressure. If it is a perpetual, the funding rate on the buy side could be materially negative — meaning you are paying to hold the long while executing the transfer pipeline. OKX withdrawal for UP carries standard L1 or L2 timing depending on the network; if this token routes through a congested chain, a 20-minute window can become a 90-minute window, and by then the OKX sell-side spread has compressed. Execution verdict: executable only for desks with pre-funded OKX spot and the ability to open the Gate Futures hedge simultaneously. Pure cash-and-carry execution here, not a simple withdrawal-and-sell. Size cautiously — low-cap futures books fill fast and slippage on UP at Gate is non-trivial.
Opportunity 2 — OP, 13.50% spread. Buy side: Bybit Spot at $0.134800. Sell side: Coinbase at $0.153000. Optimism is a widely-held, reasonably liquid mid-cap with withdrawal infrastructure on most major exchanges. The Bybit-to-Coinbase leg is the cleanest pipeline of the top five — Optimism runs on its own L2 with bridge times that have tightened considerably, and native OP token transfers via the Optimism mainnet are typically sub-30 minutes when gas is cooperative. The 13.50% spread means that even with 0.2% trading fees on each side and a conservative withdrawal fee estimate, you are looking at approximately 12.9% net before slippage — roughly $1,290 on a $10,000 position. The question is always Coinbase-side depth: selling $10K of OP into Coinbase at $0.153 is achievable without significant slippage given typical Coinbase OP order book depth, but sizing into $50K+ starts to move the market against you. This is a Tier 1 executable for standard arb position sizes.
Opportunity 3 — XLM, 12.47% spread. Buy side: Coinbase at $0.152132. Sell side: Coinbase at $0.171100. Both legs on the same exchange, which immediately signals a spot-versus-perpetual or spot-versus-futures basis play rather than a classical cross-exchange arb. Stellar Lumens carries a Coinbase perpetual or futures contract, and the $0.171100 sell price represents the elevated contract price relative to spot. This is a cash-and-carry setup: buy XLM spot at $0.152132, simultaneously short the Coinbase perp or futures at $0.171100, and wait for convergence. The 12.47% basis is rich but not absurd — similar basis levels have held for hours on mid-cap assets during momentum sessions. Risk factors: funding rates on the short futures leg eat into profit every 8 hours; if the basis widens before it converges, margin requirements increase. For desks running delta-neutral books, this is a clean setup. For pure spot arb desks without derivatives access, this opportunity does not apply.
Opportunity 4 — ATOM, 11.88% spread. Buy side: Coinbase at $1.936000. Sell side: Coinbase at $2.166000. Like XLM, ATOM presents as an intra-Coinbase spread — almost certainly the same spot-versus-futures or spot-versus-perp structure. Cosmos is a more actively traded asset than XLM with better derivatives liquidity, which makes the funding rate calculation more predictable. At $2.166 on the sell side versus $1.936 on the buy, the per-unit profit before fees is $0.230 — on a 1,000 ATOM position (roughly $1,936 notional), that is $230 gross, or approximately 11.88% gross margin. After estimated fees and funding drag over a 24-hour hold, net margin targets 9-10%. This is a slower, carry-oriented trade rather than a rapid withdrawal-and-sell. Suitable for desks comfortable with intraday carry risk on Cosmos.
Opportunity 5 — BEAM, 11.49% spread. Buy side: Bybit Spot at $0.001767. Sell side: Coinbase at $0.001970. BEAM is a gaming ecosystem token that has seen periodic Coinbase premium spikes driven by U.S. retail attention. The buy at Bybit Spot ($0.001767) and sell at Coinbase ($0.001970) represents a clean cross-exchange classical arb — buy on one, withdraw, sell on the other. The challenge with BEAM is withdrawal network specifics: depending on which chain BEAM routes on, transfer times vary. Coinbase BEAM listings tend to specify a particular network, and if that network is experiencing congestion or Coinbase is in a withdrawal-verification window, your transfer timeline extends. The 11.49% spread is large enough to absorb moderate delays, but gas costs on some networks can be disproportionately high for low-unit-price tokens (you might be moving millions of tokens to capture the spread, and the gas cost on the token count can matter). Position the trade with pre-funded Coinbase fallback if possible. Verdict: executable with attention to network routing and gas.
The dominant pattern today is Coinbase premium versus Asian venues — and it ran consistently, not episodically. OP appeared twice as a Coinbase-elevated spread (versus Bybit at 13.50% and versus Binance at 11.35%), and the AI token appeared three times as a Binance-to-Coinbase setup (10.94%, 10.79%, 10.53%). That is five separate entries across just two assets, and all five had Coinbase as the sell side. When the same exchange prints premium repeatedly across multiple assets in a single session, you are not looking at random noise — you are looking at structural U.S. session demand imbalance. Coinbase's retail order flow was absorbing supply at elevated prices, and the Asian venues were not repricing fast enough to close the gap. This is one of the most reliable cross-regional arb conditions.
The Bybit-to-Coinbase pipeline specifically was the workhorse today, appearing in both the OP and BEAM setups. Bybit has competitive withdrawal infrastructure and relatively predictable KYC-to-withdrawal timing for verified accounts. For desks that have established Bybit-Coinbase as a primary corridor, today's session was a reward for that infrastructure investment. The Binance-to-Coinbase corridor also activated in multiple assets (OP and AI), confirming that the Coinbase premium was broad-based and not venue-specific on the buy side.
The Gate Futures-to-OKX spread on UP is an outlier from a different pattern category. Gate Futures versus a major CEX spot is a derivatives-basis play with a different risk profile than spot-to-spot arb. Gate's futures books can carry localized sentiment that diverges dramatically from consensus spot prices — particularly for low-cap assets with thin open interest. OKX, by contrast, tends to run a tighter spot book that tracks broader market consensus. When Gate Futures are priced significantly below OKX spot on the same asset, the most common driver is either forced liquidation pressure on the Gate book or a structural basis from a dated contract. Both scenarios require the arb desk to understand the derivatives mechanics before sizing.
The intra-Coinbase spreads (XLM and ATOM) represent a third pattern: internal basis arbitrage. These are not traditional cross-exchange opportunities but rather spot-versus-derivatives setups that require access to Coinbase Advanced derivatives or Coinbase Prime. This pattern appears when perpetual funding rates get dislocated from spot during momentum sessions — longs pile into perps, pushing the perp price above spot, and the basis opens up. It is entirely executable but requires a different toolkit than pure spot arb: you need derivatives access, margin, and the discipline to hold the position through funding cycles.
The speed-versus-size tradeoff in cross-exchange arbitrage is not abstract — it shapes every single execution decision, and today's opportunity set forces you to confront it directly. The AI token at 10.53-10.94% across three separate Binance-to-Coinbase entries is the clearest example of a speed-favoring setup. At $0.030-$0.034 per token, you are moving in a low-price asset with likely deep Binance order books but uncertain Coinbase depth. The spread is real, but Coinbase's AI token book may not absorb large sell orders without meaningful slippage. A desk that moves $5,000 quickly — buy on Binance, fast withdrawal, sell into Coinbase immediately — captures 10%+ net. A desk that tries to move $100,000 through the same corridor will find that the Coinbase book thins out well before $0.034 and the effective average exit price drops materially.
Position sizing recommendations based on today's dataset: For Coinbase-premium plays on mid-caps (OP, AI, BEAM), keep individual trade size under $25,000 notional per pass. Above that, slippage costs start to eat meaningfully into a 10-13% gross spread. For assets with verified deeper books (OP has better depth than AI), you can push toward $50,000 but should ladder your sell orders across multiple price levels rather than hitting the book with a single market order. For the UP futures play, size conservatively given the futures-book depth uncertainty at Gate — $5,000-$10,000 exploratory size to validate execution before scaling.
Slippage modeling for today's top setups: assume 0.3-0.5% slippage on the sell side for assets priced under $0.01 per token (BEAM at $0.001967, AI at $0.030-$0.034). At these unit prices, order books tend to be expressed in large token quantities that look deep but evaporate quickly once a single aggressive seller appears. For ATOM at $2.166, slippage is lower in percentage terms because the order book is priced per unit in dollar terms rather than fractional cents — expect 0.1-0.2% slippage on reasonable sizes. For OP at $0.153, the Coinbase book has been historically liquid enough that 0.15% slippage is achievable on $20,000 sell orders.
The general speed rule for cross-exchange spot arb: if the spread is above 8% net of fees, you can afford to take 30-45 minutes to execute carefully. If you are chasing a 3-5% spread, you need sub-10-minute execution or the window closes on you. Today's opportunities are all in the 10%+ range, which gives you meaningful time — but do not mistake a wide spread for an infinite window. The Binance-to-Coinbase gap on AI appeared three times today, which suggests either it recycled as traders moved capital or the spread is structurally sticky for this session. Either way, windows above 10% on reasonably liquid assets rarely survive past 60-90 minutes before market makers or other arb desks compress them.
Let us walk through three concrete fee-adjusted profit calculations from today's data, using a $10,000 notional position for each.
Minimum viable spread calculation: with 0.10% taker fees on each of two exchanges, plus 0.10-0.20% average withdrawal costs, plus 0.30% slippage buffer, your all-in cost floor is approximately 0.80-1.00% for an efficiently executed cross-exchange trade. A 2% gross spread barely clears costs. A 5% gross spread gives you 4% net — acceptable. The 10%+ spreads in today's data are exceptional and represent genuine, meaningful opportunity. Any spread below 3% in the current fee environment should be ignored unless you are running a zero-fee institutional arrangement with the exchange.
For desks with Coinbase Prime access (maker fees as low as 0.05-0.10%), the fee math improves dramatically. A 13.50% gross OP spread with 0.10% round-trip fees on both sides nets closer to 13.0%, and the minimum viable threshold drops below 1%. If you are not on Coinbase Prime and you are running Coinbase as your primary sell venue, the fee optimization alone is worth the operational overhead of a Prime account given today's spread environment.
Withdrawal delay risk is the primary operational threat to every opportunity in today's dataset. Cross-exchange spot arb lives and dies by how fast your tokens move from exchange A to exchange B. Any spread that looks good at T=0 is potentially worthless at T=90 minutes if your withdrawal is stuck. The specific alerts for today's setups: Coinbase withdrawal queues can extend during high-traffic sessions — if you are moving capital OUT of Coinbase (not into it), watch for delays. The Bybit-to-Coinbase corridor for OP and BEAM is your primary route today, and Bybit has periodic withdrawal verification holds for new destination addresses. If you have not pre-whitelisted your Coinbase deposit address on Bybit, do it now — holds on new addresses can run 24-48 hours.
Low liquidity warning on BEAM and TST. BEAM at $0.001767 is a micro-price token with wide bid-ask spreads on secondary levels. The 11.49% spread looks attractive, but the Coinbase order book for BEAM is thinner than it appears — the top-of-book depth may only support $5,000-$8,000 of sell volume before the price steps down materially. Do not assume the $0.001970 price is achievable for your full position; model your exit at $0.001940-$0.001950 conservatively and re-run the profit calculation. TST (Binance Futures to Hyperliquid, 11.29%) carries additional complexity: Hyperliquid is a decentralized perp exchange, and executing a sell on Hyperliquid requires on-chain interaction, gas costs, and familiarity with Hyperliquid's bridge mechanics. This is not a CEX-to-CEX trade — it is a CEX-futures-to-DEX trade, and the execution complexity is meaningfully higher.
Gate Futures-specific risk on the UP opportunity: Gate's futures platform has historically shown wider-than-expected slippage on low-cap assets during high-volatility sessions. If you are attempting to buy UP futures at $0.203900 in meaningful size, expect the actual fill price to be $0.206-$0.208 depending on order book depth. Model your buy at 1-2% above the quoted price for conservatism. Also verify that Gate Futures allows withdrawal in the underlying token — some Gate futures contracts settle in USDT only, which eliminates the delivery leg of a physical arb and requires a separate spot purchase.
Intra-Coinbase spread alert for XLM and ATOM: these are not traditional withdrawal-pipeline plays. They require margin account access and derivatives capability on Coinbase. If your operation is structured as a spot-only arb desk, XLM and ATOM from today's dataset are not accessible to you without modifications to your account infrastructure. Do not attempt to capture these by buying spot on one venue and selling spot on Coinbase when the spread is internal to Coinbase — the prices quoted reflect different instrument types, not a simple withdrawal arbitrage.
Based on today's patterns, the most likely continuation setups for May 17, 2026 center on three persistent themes: the Coinbase premium on mid-cap assets, the Bybit-Binance corridor into Coinbase, and any carryover of the UP futures basis.
OP is the highest-probability repeat candidate. It appeared twice today from two different buy venues (Bybit and Binance), both times with Coinbase as the elevated sell side. This suggests that OP's Coinbase order book is structurally absorbing buy flow above Asian venue prices during the current macro/news environment. Watch OP on Bybit and Binance at the Asian open (typically 0200-0400 UTC) and again at the U.S. open (1400-1600 UTC) — these are the windows when the Coinbase premium historically builds. If OP opens tomorrow with the same bid-ask structure, the Bybit-to-Coinbase trade is repeatable.
The AI token (three entries today) is worth monitoring closely tomorrow, but with reduced conviction. Three entries on the same asset in a single day can mean the corridor is structurally profitable or it can mean you are seeing stale data from the same window being recycled in the scanner. Verify that each AI entry represents a genuinely different time window before treating this as three independent opportunities. If AI recurs tomorrow at similar spread levels, it becomes a legitimate recurring-corridor trade.
Best monitoring windows for tomorrow: Pre-U.S.-open (1200-1400 UTC) when Asian close meets European mid-session tends to produce the most durable cross-regional spreads. The Bybit-to-Coinbase plays have historically been strongest during the 60-90 minute overlap before U.S. markets open, when Coinbase retail order flow accelerates and Asian venue market makers have not yet fully repriced. Set price alerts on OP (Bybit below $0.138, Coinbase above $0.148), BEAM (Bybit below $0.0018, Coinbase above $0.0019), and AI (Binance below $0.032, Coinbase above $0.035).
Exchange pairs to monitor tomorrow in priority order: Bybit Spot versus Coinbase (primary corridor today, likely to repeat), Binance versus Coinbase (secondary corridor, confirmed on OP and AI), Gate Futures versus OKX (monitor UP specifically — if the basis persists tomorrow, it is structural and worth a larger position). Hyperliquid versus Binance Futures (the TST setup is worth watching if you have Hyperliquid infrastructure — the DEX premium occasionally runs for multiple sessions during momentum periods).
Eighty-six opportunities is a good day. A 20.99% headline spread is a great number. The real story is the systematic Coinbase premium running across OP, AI, BEAM, XLM, and ATOM in a single session — that is not random noise, that is a corridor. Desks that had the Bybit-Coinbase and Binance-Coinbase pipelines warmed up and ready made money today. Desks that did not are going to be chasing this setup tomorrow. Infrastructure is alpha. Speed is alpha. Everything else is just math.
Arbitrage Hunter — May 16, 2026