One asset, one exchange pair, nineteen chances to print money — or nineteen chances to get stuck holding a bag on withdrawal. COTI's KuCoin-to-Bybit spread ran as high as 13.41% today, and Boring Boris walks through exactly what it would have cost you to actually capture it.
Nineteen arbitrage events crossed the desk today, and every single one of them wore the same jersey: COTI. Not a rotating cast of majors, not a mid-cap altseason grab-bag — just COTI, again and again, mispriced between KuCoin and Bybit with a stubbornness that should make anyone running a spread bot sit up straight. The best print of the day was 13.41%, buying on KuCoin at $0.013430 and selling on Bybit at $0.013970. That is not a rounding-error spread. That is the kind of number that either means free money is sitting on the table, or it means something about COTI's liquidity on one of these venues is structurally broken and the 'arbitrage' is an illusion built on an order book too thin to actually trade against.
The honest read: when one token generates ten-plus double-digit spread events in a single day against the same exchange pair, that is not random noise from ten different assets briefly diverging. That is a pattern. Either KuCoin's COTI market is systematically lagging Bybit's price discovery, or Bybit is running hot on COTI relative to the rest of the market, or — the boring but likely explanation — the spread is real on paper and evaporates the moment you try to move size through it. This report walks through the top five prints, the exchange pattern behind them, and the fee math that decides whether any of this was worth touching.
The day's headline print. Buy COTI on KuCoin at $0.013430, sell on Bybit at $0.013970 — a gap of 13.41% before any costs are applied. This is the kind of spread that gets screenshotted and posted in Discord servers, and for good reason: even after aggressive fee assumptions, there's real margin left over if the size is there. The problem is the size. COTI is not a deep-book asset on either exchange during normal hours, and a spread this wide almost never survives more than a few thousand dollars of notional before slippage eats it alive. Risk factors: KuCoin withdrawal queues for COTI can run anywhere from a few minutes to over an hour depending on network congestion, and by the time funds land on Bybit to actually sell into that $0.013970 print, the price has usually moved. My take: executable in theory for a fast bot with pre-positioned inventory on both sides (no withdrawal wait), but a manual trader chasing this after the fact was almost certainly too late.
Buy on KuCoin at $0.013620, sell on Bybit at $0.014260. Same pair, same direction, slightly tighter than the day's top print but still comfortably in double digits. What's notable here is the absolute price level — both legs sit noticeably higher than opportunity #1, which tells you this print likely occurred later in a short-term COTI rally, with KuCoin's book lagging the move upward on Bybit. That lag pattern is a classic 'slow exchange' signature: KuCoin isn't repricing as fast as Bybit during a directional push. Risk: chasing a rally-driven spread is riskier than chasing a flat one, because if COTI keeps running on Bybit while you're still waiting on a KuCoin withdrawal, your sell price moves favorably — but if it reverses, the spread compresses right as you need it to hold. My take: opportunistically executable only with capital already parked on both venues; not one to withdraw-and-chase.
Buy on KuCoin at $0.013360, sell on Bybit at $0.013950. This print sits close in absolute price terms to opportunity #1, which suggests these two events may be part of the same underlying dislocation window rather than two independent occurrences — likely captured minutes apart as the gap oscillated between roughly 12.5% and 13.4% before eventually closing. Volume data wasn't captured cleanly in this scan (more on that in the risk section below), so sizing has to be estimated from typical COTI order book depth, which on both these exchanges tends to thin out fast past low five figures. My take: same execution profile as #1 — good on paper, good only for pre-funded bots, not for anyone starting from a single wallet.
Buy on KuCoin at $0.013600, sell on Bybit at $0.014192. Another entry clustered near the higher price band alongside opportunity #2, reinforcing the read that COTI had at least two distinct windows today where the KuCoin-Bybit gap blew out — one around the $0.0133-0.0134 level and another around $0.0136-0.0142. That clustering matters for pattern recognition: if you're building a monitor for this pair, you're not looking for COTI to sit still, you're looking for it to move fast on one exchange before the other catches up. Risk factors mirror the others: withdrawal lag is the single biggest killer of this trade, full stop. My take: executable for automated systems with API-triggered withdrawals and pre-approved whitelisted addresses to shave transfer time; a manual trader is fighting a losing race against the clock here.
The one genuinely different setup in the top five: buy on Gate Futures at $0.013150, sell on Bybit spot at $0.013844. This is a cross-product arb, not a simple spot-to-spot transfer play, and that changes the risk profile entirely — in a good way. A futures-to-spot structure means no withdrawal wait is strictly necessary if you're running it as a basis trade rather than a physical transfer: short the futures leg's implied mispricing, hedge or unwind against spot, and you've sidestepped the transfer-time problem that plagues opportunities #1 through #4. The tradeoff is funding rate exposure and futures-specific liquidation risk if the position is levered. My take: this is the most interesting entry on the list precisely because it doesn't depend on beating a withdrawal clock — worth a second look for anyone running cross-venue basis strategies rather than pure spot arb.
The pattern today is about as unambiguous as it gets: KuCoin sits on the low side, Bybit sits on the high side, and COTI is the asset where that gap keeps reopening. Nine of the ten logged events follow this exact structure — buy KuCoin, sell Bybit — with spreads ranging from 11.71% up to 13.41%. That's not a coincidence of ten separate assets each drifting apart; it's one persistent structural lag between two venues on one token. The likely mechanics: KuCoin's COTI market makers are either thinner or slower to reprice than Bybit's, so whenever COTI gets a burst of directional volume on Bybit (where liquidity and taker flow both tend to be deeper for this kind of small-cap), KuCoin's book takes longer to catch up. That creates a recurring, almost mean-reverting gap rather than a one-off dislocation.
The lone outlier — the Gate Futures leg into Bybit spot — hints at a second, related phenomenon: COTI's price discovery may be lagging across multiple venues relative to wherever the 'true' price is being set, with Bybit consistently acting as the reference point everyone else is chasing. If that's the case, a smarter version of this trade isn't 'buy KuCoin, sell Bybit' as a mechanical rule — it's 'identify wherever COTI is currently mispriced relative to Bybit, spot or futures, and trade that gap with whichever instrument gets you there fastest.' No Binance-OKX or Hyperliquid-CEX pairs showed up in today's data at all, which either means COTI simply isn't listed/active enough on those venues to generate a comparable signal, or their market makers are simply faster and the gap never opens there in the first place.
Every one of today's ten logged spreads falls in the 11.7%-13.4% band — tight clustering, which usually signals a shallow order book rather than a genuinely deep dislocation. Wide, consistent spreads on a small-cap token like COTI are a size warning, not a size invitation. The math on a token trading at roughly $0.0134 is straightforward: even a modest $10,000 position is over 745,000 COTI tokens, and moving that much through either KuCoin's or Bybit's book in a single clip will absolutely walk the price and compress the spread you're trying to capture. The realistic play here is small, fast, and repeated — not large and slow.
Position sizing recommendation: treat anything above low four figures per leg as size that needs to be worked in slices, not dumped as a market order. Slippage on a token this thin scales non-linearly — the first $1,000-2,000 might move the book 0.1-0.2%, but push past $5,000 and you're likely eating 1%+ just on execution, which starts cannibalizing the very spread you're chasing. The trade-off is real: chase the full 13.41% with size and you'll get a worse average fill than the headline number promises; stay small enough to avoid slippage and your absolute dollar profit shrinks even as your percentage capture improves. For an asset like COTI at these spread levels, the sweet spot is probably running the trade in $500-1,500 clips repeatedly rather than one large swing — assuming the withdrawal-time problem is solved (see below), speed of repetition beats size of any single fill.
Let's run the top print — 13.41%, buy KuCoin $0.013430, sell Bybit $0.013970 — through realistic costs on a $2,000 notional position:
That's the good case — the trade held its shape and size stayed small enough to avoid material slippage. Now run the same math on the smallest print in the set, 11.71% (buy KuCoin $0.013400, sell Bybit $0.013982): gross profit on $2,000 is $233.40, fees and network cost run roughly the same $4-5 combined, but slippage eats a proportionally larger bite of a smaller edge. Net still clears comfortably above 11%, which is the real takeaway: at these spread magnitudes, fees and withdrawal costs are almost a rounding error. The number that actually decides whether the trade works is slippage and — more than anything else — whether the spread is still open by the time you've moved funds between exchanges. As a rule of thumb for a token in COTI's liquidity tier, anything under roughly 2-3% gross spread isn't worth chasing once fees, slippage, and withdrawal timing risk are priced in; today's entire list, sitting at 11.7% and above, clears that bar with room to spare — assuming you can actually get filled at the quoted prices.
The single biggest red flag in today's dataset isn't the spreads themselves — it's what's missing around them. Pump volume, dump volume, buy pressure, and sell pressure all read $0.0M across the board. That means this scan captured price dislocation without capturing depth or flow, which is exactly the condition under which a headline spread looks amazing on a chart and turns into a bad fill in real trading. Treat every number in this report as an indicative spread, not a confirmed fillable size, until you've checked the live order books yourself.
Watch COTI on the KuCoin-Bybit pair first thing — a pattern this persistent across a single session (ten-plus events, same direction, same venues) often bleeds into the next session before market makers on the lagging side finally catch up. If KuCoin's COTI book was systematically behind Bybit's today, that lag doesn't necessarily resolve overnight; it resolves when someone on the KuCoin side finally repositions their quotes, and that can take a day or two of arbitrageurs draining the edge before it tightens. Best windows to watch: the first hour after any sharp COTI move on Bybit, since that's precisely the kind of directional push that exposed today's gaps. Beyond COTI, keep half an eye on other low-cap tokens that share KuCoin's market-making desk — if the lag is structural to how KuCoin prices thin assets rather than COTI-specific, similar gaps could show up elsewhere. And don't ignore the futures angle from opportunity #5 — if Gate's futures curve on COTI continues drifting from Bybit spot, that basis trade may be more repeatable and less time-sensitive than the spot-to-spot version that dominated today's list.
Ten prints, one token, one exchange pair, and a spread that never dipped below 11.7% all day. On paper, today was a gift. In practice, it was a gift wrapped in a withdrawal queue and an order book that couldn't have absorbed most of you trying to unwrap it at once. Trade the pattern, not the headline number — and check the depth before you check the percentage.
Arbitrage Hunter — August 5, 2026