A 42-opportunity arb session led by a 10.06% MOVR spread between KuCoin and Bitget, plus a rare double-spread on 龙虾 across four venues. Boring Boris breaks down the top setups, fee math, and which exchange pairs to watch tomorrow.
Forty-two arbitrage opportunities crossed the desk today, and the top of the board was loud. MOVR opened a 10.06% gap between KuCoin ($0.777000 bid-side) and Bitget ($0.815800 offer-side) — the kind of spread that either means a fat, fast fill or a liquidity trap dressed up as free money. Nothing else in the session cracked double digits, but the next tier was dense: two separate 龙虾 (Longxia) spreads within roughly an hour of each other, both routed through different venue pairs, followed by a cluster of mid-cap alts — HUMA, HEMI, DEXE, and a token labeled simply "4" — all posting spreads north of 7.5%. That's a busy tape by any measure, and it's the kind of session where the difference between a trader who profits and one who gets run over comes down entirely to execution speed and how honestly you price your fees. Let's walk the board.
MOVR — 10.06% spread. Buy on KuCoin at $0.777000, sell on Bitget at $0.815800. This was the standout print of the day and the only spread to clear double digits. MOVR (Moonriver) is a mid-liquidity Polkadot-parachain token — tradable in size on both venues, but not deep enough that a large clip won't move the book on either leg. The size of the gap is a red flag as much as an opportunity: a 10%+ spread on a token this liquid on two major CEXs doesn't usually last long or fill clean at the quoted price, which tells me this was likely a fast, order-book-driven dislocation rather than a stable structural gap — probably triggered by a large one-sided fill or thin resting liquidity on one side. Executable, but only for traders who already had capital pre-positioned on both KuCoin and Bitget; anyone needing to move funds between exchanges first almost certainly missed the window. Withdrawal/deposit lag on MOVR between these two venues would have closed most or all of the edge before a transfer-based arb could complete.
龙虾 — 8.48% spread. Buy on Gate Futures at $0.064361, sell on Binance Futures at $0.066633. A futures-to-futures spread, which is actually the cleaner arb structure — no spot withdrawal delay, since the play here is basis/funding arbitrage or a rapid open-close pair rather than a physical transfer. The catch with 龙虾 specifically is that it's a thin, low-cap name; futures open interest on both Gate and Binance for a token like this tends to be shallow, so slippage on entry and exit can eat a meaningful chunk of that 8.48% before you've even accounted for fees. Executable in small size for a trader already running bots on both venues with pre-funded margin; not executable at scale.
龙虾 — 7.85% spread. Buy on Bitget at $0.067079, sell on Exchange24 (HTX) at $0.072342. The second 龙虾 print of the day, and notably this one is spot-to-spot across two different venues than opportunity #2 — a sign that 龙虾 was genuinely mispriced across the ecosystem today, not just a one-off glitch on a single pair. Exchange24/HTX withdrawal times and listing depth for a token this obscure are the real risk here; HTX in particular has had spotty withdrawal processing on lower-cap assets historically. This spread is more believable as a genuine, sustained dislocation than opportunity #1, but the physical-transfer version of this trade (rather than pre-positioned capital on both sides) is a coin-flip on whether the spread survives the transfer time.
HUMA — 7.80% spread. Buy on Exchange51 (Aster) at $0.020732, sell on OKX at $0.022350. Aster is a newer venue in our mapping (added late July), and new/smaller exchanges routinely show inflated spreads against majors like OKX simply because their order books are thinner and slower to arbitrage away by the broader market. That's not automatically bad for a trader who can act fast — it just means the spread is more likely to be real and less likely to be a stale-quote artifact. The risk is entirely on the Aster side: withdrawal reliability and processing time on a smaller exchange is the single biggest variable, and I'd treat any transfer-based version of this trade as higher risk than the raw percentage suggests.
HEMI — 7.71% spread. Buy on KuCoin at $0.011121, sell on Bybit at $0.011585. Both venues are top-tier, well-capitalized exchanges, which makes this one of the more genuinely tradable spreads on the board despite the modest headline number. HEMI is a newer, sub-cent token, so percentage spreads naturally run hotter than they would on a more mature asset — but KuCoin-to-Bybit withdrawal times for most ERC-20/native tokens are reasonably fast (typically minutes, not hours), which meaningfully improves the odds of capturing a chunk of this before it closes. Of the five, this is the one I'd rate most executable for a mid-size trader without pre-positioned dual-exchange capital.
The pattern today skews toward newer/smaller venues bleeding value to the majors. Exchange51 (Aster) and Exchange24 (HTX) both showed up as the cheap side of a spread against OKX and Bitget respectively — consistent with what you'd expect from exchanges still building out market-maker coverage and arb-bot presence. KuCoin also shows up twice as the buy-side venue (MOVR, HEMI), both times against a larger, more liquid counterpart (Bitget, Bybit) — suggesting KuCoin's book on smaller-cap listings is lagging price discovery relative to the top-tier venues. On the futures side, the Gate Futures / Binance Futures pair (龙虾) is worth flagging separately: futures spreads are structurally different from spot spreads since they don't require an asset transfer, only margin pre-positioning, which makes them faster to exploit but requires a bot or a trader already running both venues simultaneously. No Hyperliquid prints showed up in today's top spreads, which is itself informative — the majors-vs-majors gap (OKX/Binance, Bybit/Bitget) stayed tight, and essentially all the meaningful edge today was concentrated where a smaller or futures venue was mispricing against a bigger spot venue.
There's a clear inverse relationship on today's board between spread size and how long you can reasonably expect it to survive. The 10.06% MOVR gap and the two 龙虾 prints (7.85%-8.48%) are the largest spreads and also the ones most likely to have been fleeting — big dislocations on tokens with decent CEX presence get arbed shut fast, often within seconds to a couple of minutes, by bots faster than any manual trader. The smaller spreads in the 6.8%-7.2% range (LIGHT, HUMA/OKX Spot, TRUMP) are more likely to represent a slower-moving structural gap — thinner order books, less bot coverage — and therefore have a longer effective window, but you pay for that in size: you simply can't push much volume through a thin book without your own trade collapsing the spread you're trying to capture. My rule of thumb for sizing today's book: treat anything above 8% as a small, fast in-and-out clip (5-15% of your normal position size, because you're racing bots), and treat the 6.8%-7.5% cluster as your primary sizing zone, where you can run closer to full position size but need to ladder your fills rather than market-order the whole size in one shot to avoid eating your own edge in slippage.
Take the HEMI spread as the cleanest worked example: buy at $0.011121 on KuCoin, sell at $0.011585 on Bybit, a gross spread of 7.71%. Trading fees first — assume 0.1% taker on each leg (0.1% buy + 0.1% sell = 0.2% round trip), which is standard for both venues without VIP tiers. That knocks the spread to 7.51%. Next, withdrawal fee: moving HEMI from KuCoin to Bybit might cost a flat network fee — call it roughly 0.3%-0.8% of position value depending on trade size (flat fees hurt small trades disproportionately; on a $10,000 clip a $20-30 network fee is 0.2-0.3%, on a $1,000 clip the same fee is 2-3% and can wipe the trade). At a reasonable $10,000 position size, net profit after fees lands around 7.0%-7.2% — still very healthy. Now compare to the MOVR trade: 10.06% gross, minus 0.2% round-trip trading fees, minus a withdrawal leg that — if you don't already have capital pre-positioned on both KuCoin and Bitget — likely takes long enough that the spread closes before you finish transferring, turning a 10% paper spread into a real loss if the market reverts. The takeaway: below roughly 1.5%-2% gross spread, fees and slippage eat essentially all your edge on a standard-size clip, so that's the practical floor worth chasing on major-to-major pairs. On smaller/newer venues with higher withdrawal fees, I'd push that floor up to 3%-4% minimum before it's worth the operational risk.
Exchange24 (HTX) and Exchange51 (Aster) both appear on the sell/buy side of today's larger spreads, and both carry elevated withdrawal-reliability risk relative to Tier-1 venues — Aster as a newer platform with less battle-tested infrastructure, HTX with a history of inconsistent processing times on lower-cap tokens. Any arb trade routed through either exchange should be treated as transfer-risk-heavy, not just price-risk-heavy. 龙虾 and "4" are both thin, low-liquidity names — the percentage spreads on these can be structurally wide simply because the order books are shallow, meaning your actual fill price on size will slip well past the quoted top-of-book number regardless of which side you're on. If you're running any of today's spreads as a physical-transfer arb rather than dual-pre-positioned capital, build in a hard stopwatch: if the spread hasn't been captured within a couple of minutes of order placement, assume it's gone and don't chase it by widening your limit price.
Two 龙虾 spreads on the same day across four different exchanges is the strongest signal on the board — that token's price discovery is fragmented enough that it's worth keeping on a watchlist for a repeat tomorrow, particularly the Bitget/Exchange24 and Gate Futures/Binance Futures pairs. Aster (Exchange51) is worth monitoring generally as it continues onboarding liquidity — new-venue mispricings against OKX and Binance tend to persist for a few weeks post-listing before market makers close the gap. Best windows to watch: the first hour after major Asian-session opens, when smaller venues like HTX and Aster tend to lag price updates from the majors, and any period following a sharp BTC move, which is when spot books on thinner exchanges fall behind fastest. Keep KuCoin-vs-Bybit and KuCoin-vs-Bitget on the standard watch pair list — KuCoin's book depth on sub-$0.02 tokens has been the recurring soft spot all session.
Forty-two spreads, one double-digit print, and a lot of edge that existed for less time than it took to read this report. Size small on the fast ones, size up on the boring ones, and never forget the withdrawal clock is part of your P&L. Arbitrage Hunter — August 29, 2026.