Copy-trading costs: measure the fill path before claiming a return gap

A dated correction withdrawing unsupported mirrored-trade performance findings and specifying the fill register needed for a cost study.

Editorial correction: copy-trading costs need the fill register

Editorial correction · 9 September 2026. The original February 14 article said that it audited 12,480 mirrored trades from 27 providers and reported a +38.6% provider return versus +9.4% follower net return. It also presented delay, slippage, fee, drawdown and sizing results as if they were measured in that cohort. The page did not link provider identities, mirrored trade rows, source and follower timestamps, fill records, cost schedules, portfolio paths or calculation output. Those records could not be recovered from the public page or its broad references.

We withdraw the cohort size, return gap, cost medians, regime table, modeled paths, timing observations and numerical copy-trading rules as empirical findings. The original URL and publication date remain; this correction is a separate dated event. The page does not establish that any provider or copy-trading strategy earned or lost a particular amount.

What a cost analysis should separate

The research question remains useful if the units are made explicit:

Layer	Proposed record	Common ambiguity to resolve
Source action	Provider signal/fill URL, timestamp, instrument and size	A public idea is not necessarily an executable order
Follower action	Follower order, fill, timestamp, size and partial-fill status	“Copied” can mean requested, filled or fully closed
Market friction	Spread, slippage, commission, funding and fee schedule	Costs vary by venue, direction, size and time
Timing	Source publication, source fill, relay and follower fill times	Delay must be measured against a stated clock and price
Risk path	Position size, leverage, cash, exits and liquidation treatment	Different sizing can change both return and drawdown
Comparison	Matched benchmark and gross/net convention	Provider-reported performance may not use the same benchmark or costs

Gross source performance minus follower performance is not automatically a causal decomposition. A defensible analysis would calculate the realized follower path from its own fills, rerun it under predeclared cost scenarios, and identify which rows are measured, modeled or missing. It should report uncertainty rather than turn a selected sample into a universal warning.

Minimum register for a replacement study

Complete provider and follower-account universe, inclusion rule and the treatment of inactive, deleted or partially mirrored records.
One row per source action and follower action, with UTC timestamps, prices, quantities, instrument, direction, fees and fill status.
A fixed benchmark, return-period convention, currency, financing treatment, position-sizing rule and handling for missing or rejected orders.
Raw price and fee inputs, realized equity paths, drawdown definition, exclusion log, calculation code and an independent check.

Until those records exist, delay, slippage and fees are mechanisms to measure, not measured results from this article. No fixed multiplier, cost ceiling or risk percentage is validated here.

Source context and limits

The original references included the ESMA Investor Corner, SEC Investor Alerts and the Barber–Lee–Liu–Odean paper Just How Much Do Individual Investors Lose by Trading?. These provide regulatory or general-investor context; the paper studies a different population and none contains this page's 12,480 trade rows, 27 provider universe, follower fills or return calculations. They do not validate the withdrawn comparison.

Change record

Original publication: 14 February 2026. The unsupported 12,480-trade, 27-provider and +38.6% versus +9.4% findings were withdrawn and the proposed fill-register method clarified: 9 September 2026. The original wording remains in version history and is not promoted as current research. AI-assisted editorial review; no independent human expert review.