copy trading
Copy trading answers a real problem: most people have no edge and know it. The way platforms present the choice introduces a different problem, because the ranking is optimised for signing you up rather than for keeping you solvent.
Two mechanisms, both structural
01
Variance selection
Over any short window the highest returns belong to the highest risk takers. Ranking by 30-day return is a lottery draw with a leaderboard attached.
02
Survivorship
Accounts that blow up leave the list. The visible population is permanently biased toward people who have not met their losing streak yet.
What the leader is paid, and for what
Compensation determines behaviour more reliably than intent does. Volume share pays the leader on turnover regardless of your result. A performance fee pays on profit but rarely claws back losses, and often resets after a reset. A follower-count bonus pays for visibility. Only one of the three is aligned with you keeping money, and it is the least common.
| Metric shown | Ask instead |
|---|---|
| Return over 30 days | Return over 24 months including the worst quarter |
| Win rate | Average win against average loss, and the single largest loss |
| Number of copiers | Maximum drawdown and the date it happened |
| Risk score out of ten | How the score is computed and whether it uses open or closed equity |
A short filter
- 01Require at least 18 months of history covering one difficult market period.
- 02Reject any curve with no losing month. That pattern means open losses, not skill.
- 03Find out whether the leader trades their own capital, and how much.
- 04Get the compensation structure in writing from the platform, not from the leader.
You are not outsourcing the decision. You are outsourcing it to someone whose incentives you have not read.
Written by Aram Latifi. Ex-quant developer, now writing about the plumbing of retail trading. No affiliate links on this site.
