The strategy, without the marketing

The site describes a Polymarket arbitrage engine. This page sets out exactly what that engine is able to do, what it has actually done, and where the yield is coming from in the meantime. Every figure is read from the contract, including the ones that do not flatter us.

Watching Polymarket, live

A read-only scanner — no keys, cannot place an order — checks every active two-outcome market's order book for a complete set priced under a dollar, against real depth and each market's own fee schedule. It runs on a timer and this is its last pass, not a summary written after the fact.

A different kind of arbitrage — CEX prices, for reference

This is the classic buy-low-sell-high spread between two centralized exchanges, streamed live from their own public order books. It has nothing to do with the Polymarket engine above, and the contract cannot act on it — there is no Binance or KuCoin key anywhere in this system. It is shown because visitors ask about it, and the honest answer is a live number, not a claim.

BTC/USDT — Binance vs KuCoin

Live public order-book prices, for reference only. ArbiSmart's contract has no connection to Binance or KuCoin — this isn't a position the protocol holds or a trade it takes, just what the two order books show right now, gross of either exchange's trading fees.

ExchangeBidAskStatus
BinanceConnecting…
KuCoinConnecting…

The arithmetic, in full

Anyone can work this out from the rates and fees we publish. Doing it here first, with the live figures, is the only version of this page that is worth reading.

For every dollar deposited

Entry fee off the top, the plan's rate over its full term, claim fee off the yield. The last column is the one that matters.

PlanPaid outWe keepShortfall
Starter$2.59$0.31$1.59
Growth$3.09$0.34$2.09
Advanced$3.47$0.20$2.47
Elite$3.39$0.18$2.39

The shortfall is what must come from somewhere other than the deposit itself — strategy profit, or later deposits. It is the same ratio at any size: a hundred dollars and ten million behave identically, because the figures are proportions. Growth does not dilute it, because the growth is what creates it.

How long an idle pool lasts

If no new money arrives and everyone claims as it accrues, the pool drains at rate × (1 − claim fee) a day. Divide one by that and you get the number of days it survives.

Starter — term 180 days93 days
Growth — term 150 days62 days
Advanced — term 120 days44 days
Elite — term 90 days35 days

Every plan empties before its own term ends

A pool holding only Elite positions is drained on day 35 of a 90-day term. This is not a liquidity risk that good management avoids — it is what a fixed rate above what the capital earns means, arithmetically. Continuing past that day requires money that has not arrived yet.

What rate a real strategy could fund

Move the slider to whatever annual return you think a strategy can genuinely earn, and compare the daily rate that would pay for against the ones above.

Payable daily rate

0.0329%

after a 20% performance fee

Lowest plan promises

1.2000%

36× the payable rate

Push the slider to 50% — better than almost any fund sustains — and the payable rate is still a small fraction of the lowest plan. This gap is not closed by a better strategy. It is closed by a lower rate, which is what a variable-yield design does: it pays what was earned and promises nothing in advance.

Where that leaves us today

Principal staked0 USDT
Yield promised on it, at the lowest rate0 USDT
Total owed by the end of those terms0 USDT
Held by the contract right now0 USDT

Principal is fully backed — the contract holds what it owes as capital. The gap is the promised yield, and nothing has earned it yet. Read live from the contract at this block, including on the days it does not flatter us.

What the contract can do

Three operations, all of them real calls to Polymarket's own deployed contracts. The events for each are emitted only after the external call has already succeeded, so an event on-chain means the action genuinely happened.

Split collateral into a complete set

One dollar becomes one YES token and one NO token of the same market, through a direct call to Polymarket's permissionless Conditional Tokens contract.

Merge a set back to collateral

The reverse, at any time before the market resolves. Used to unwind a position rather than to realise a gain.

Redeem after a market resolves

Once the outcome is settled, the winning tokens are exchanged for collateral. Fully autonomous and fully on-chain.

What it cannot do, and why that matters

The contract cannot buy or sell on Polymarket's order book. This is not a feature we have not got round to — it is a permission Polymarket does not grant.

Polymarket's own CTFExchange.fillOrder, fillOrders and matchOrders are gated behind an onlyOperator modifier in their deployed source. Only addresses their admins have explicitly granted the Operator role may call them, and that role is not given to arbitrary third-party contracts. Ours does not have it.

Why that removes the profit

A complete set always costs exactly one dollar to create and always returns exactly one dollar when redeemed. The profit in this kind of arbitrage comes from buying a set for less than a dollar, or selling the outcomes for more — and both of those are order-book trades. Split and redeem on their own, which is what the contract can do unaided, net to zero before gas.

There is a supported route to closing this gap: an off-chain component using Polymarket's CLOB API, together with the contract implementing EIP-1271 so it can act as a smart-contract order maker. It is deliberately not implemented, because getting the signature validation wrong is worse than not shipping it. Until it is built and verified, the engine cannot trade.

All of this is written into the contract's own source under the heading “IMPORTANT, HONEST LIMITATION”, which is published and verified on Sourcify. You do not have to take this page's word for any of it.

Why publish this

Because you would find it anyway — the numbers are on-chain and the limitation is in the source. A project that only publishes the figures that flatter it has told you nothing about the figures it did not publish. When the strategy does start, its results will appear on this page from the same contract reads, and they will be worth something precisely because these ones were shown first.