Polymarket Copy Trading in 2026: Every Copied Trade Pays Three Costs the Leader Didn't

Copy trading on Polymarket copies a trader's decisions, not their returns. Every copied fill pays three costs the leader did not: price drift during your detection lag, the taker fee on your entry, and the worse fill the leader's own order just created. On Polymarket's thinner books the third dominates. Worked from the published fee formula, a copier who follows a 100-contract buy at 50 cents in a crypto market is down about $3.00 on a $50 stake, 6%, before the market moves at all. Politics markets cost about 4.5%; geopolitics, which Polymarket does not charge fees on, about 2.5%.

The tools differ mainly in custody and automation. Polycopy is non-custodial, free to follow and paper-trade, $30 a month to run live bots, and charges 0.5% per executed trade. LuckyLobster runs agents on managed wallets. Laika Labs and Alphascope are leaderboard-and-alert products that leave execution to you. The Vultax paper-trading desk (preview) is the honest version of a copy product: it fills simulated orders against live prices so the three costs can be measured before any money is at risk.

Last verified 5 September 2026 from each vendor's own pages.

Four products that let you follow a Polymarket wallet, plus the simulator.

Polymarket's taker fee is C × feeRate × p × (1 − p), where C is contracts and p the price; makers pay nothing. The rate is 0.07 on crypto, 0.05 on sports, economics, culture and weather, 0.04 on finance, politics, mentions and tech, and zero on geopolitical markets. At p = 0.50 the fee is at its maximum, so the worked example below is the worst case at the most common price.

Suppose the leader buys 100 Yes contracts at 50 cents in a crypto market. Their stake is $50. Your copy, placed as a market order a few seconds later, pays three things the leader did not.

• The leader paid none of these if they posted a limit order - "Makers are never charged fees", so a leader who rests orders keeps the 3.5% you paid to take.
• The costs repeat on exit - the same three land again when the position is closed or auto-closed, so a round trip in a crypto market costs a copier roughly 12% of stake at 50 cents.
• Speed helps only if it is measured - a sub-50 ms mirror still takes liquidity and still pays the fee; it reduces drift, not the other two costs.

The case for copying rests on persistence, and persistence exists. Across 9,949 Polymarket wallets active in both June and July 2026, 67.4% of June's profitable wallets were profitable again in July and the rank correlation between months was 0.83. But two facts from the same data cut against the copier.

First, persistence lives among wallets that trade often; one-trade wallets were profitable 12.2% of the time. The wallets worth copying are the busy ones, and busy leaders multiply the three costs by their trade count. Second, gains are concentrated: the top 1% of winning wallets, 36 addresses, took 43.9% of all realised gains over 18 June to 14 July, and the top 10% took 79.5%. Outside that decile a leaderboard rank is mostly noise, and inside it the leaders are the always-on wallets that fill hundreds of trades a day, each of which would charge you the table above.

Rules that follow from the numbers.

• Paper-trade the leader for a month first - Polycopy's free bots and the Vultax paper desk (preview) both do this; if the simulated copy is not clearly positive net of the fee formula, the live copy will not be either.
• Copy with limit orders, not market orders - resting at the leader's price or better converts the taker fee to zero and removes the worse-fill cost; you will miss some entries, and the entries you miss are the ones that moved against you fastest.
• Prefer fee-free categories - geopolitical markets carry no Polymarket fee, so the copier's structural disadvantage there is the platform fee and the fill alone.
• Follow decisions, not wallets - the useful output of a leader is the market they chose and the side; entering later, with your own size, at a resting price, keeps the information and drops most of the cost.
• Never copy a one-trade wallet - the persistence evidence does not reach it.

The strip at the top of this page is rewritten every ten minutes from Polymarket's public leaderboard by day, week and month, its data API's trades of $10,000 or more, its global open interest, and Vultax's record of every fill it observes, including the always-on wallet cohort. The Vi IQ beneath it scores the six domains the Vultax terminal scores for a pair, computed for this page's subject from published inputs, with a domain that cannot be computed shown as unavailable rather than filled in. The figures in the body of the article are the measurement it was written on and are dated; the strip is what the same measurement says now.

The strip shows the median trade Vultax observed in the last day, the share of wallets that traded exactly once, the number of trades of $10,000 or more in the last hour and how many wallets traded in the last day. The first two describe the crowd a copier joins; the third is the flow a copier is late to; the fourth is the size of the pool.

The three costs in this article have a fourth cousin in 2026: the leader may not be who the copier thinks. Columbia researchers flagged about a quarter of Polymarket's volume as wash trading and roughly one wallet in seven, many with no profit, which is what a wallet farming a ranking looks like from the outside; and the CFTC's February 2026 advisory cites insider trading by a candidate on his own race and by a video editor on his channel's contracts. A copier of the first pays to follow noise; a copier of the second follows conduct the Commission prosecutes.

The venues are also changing what can be copied. Polymarket's perpetual futures, launched 3 September with up to 20x leverage and closed to U.S. traders, mean a leader's profit may soon include leveraged positions that a copier on the U.S. exchange cannot hold; and with Kalshi at about 90% of combined contract volume in late August, the leaders on the two venues are increasingly different people.

Tool facts were read from each vendor's own pages on 5 September 2026; where a vendor site blocked automated reading the cell says so and leans on third-party guides, which is weaker evidence. The fee arithmetic uses Polymarket's published formula and rates as of that date and a 0.5% platform fee taken from Polycopy's pricing page; other platforms' fees are not stated. The persistence and concentration figures come from the linked Vultax studies and carry their own short windows. Nothing here is a recommendation to copy any wallet or to trade. Challenge a cell at [email protected]; material changes are logged in the changelog.