How we evaluate copy traders
Short answer: we never judge a strategy by profit alone. Each trader we review on RoboForex's RoboForex Copy Trading is checked against eight evidence points: account identity, account type, history length, drawdown, fees, exposure, position sizing, and how the subscription could be exited. A high return is a research lead. It is never a recommendation.
1. Account identity and type
Different accounts can share a nickname. We always record the exact account number, the platform (MT4, MT5 or R StocksTrader), the account type, the account currency and the leverage setting shown on the profile. RoboForex's current subscription guides require a compatible receiving account. For MT5, that means a Hedge account. We check compatibility before anything else.
2. History length
We distinguish between a sprint (weeks), a season (several months) and a track record (a year or more across different market conditions). A large yield over 48 days is weaker evidence than a modest yield over two years. Our reviews show the account age next to every return figure.
- Under 90 days: never presented as more than a high-risk research case, however large the return. The short history is shown as a badge (for example, "short history").
- 90 days or more: the minimum before we will discuss a trader as a candidate at all.
Longer observation shows more behaviour, but no fixed number of months certifies a strategy.
3. Return vs drawdown, over the same window
Our first screen is simple: we read the return for the period next to the maximum drawdown for the same period. A large return only earns a closer look once the drop is clear. The score is then marked down if the account is very young, has few followers, or if its recent period contradicts its lifetime history.
Check the period label carefully. On RoboForex's rating, the period filter period=3 means "Month", not three months. We verified this on 8 October 2026.
We put the maximum drawdown next to the return for the same period, and we label monthly and lifetime figures separately. We also show the recovery maths: a 40% drawdown needs about +67% to get back to the previous peak. Historical maximum drawdown is not a ceiling on future losses.
4. Fees and the follower's net result
RoboForex traders can set a performance fee (the platform's own trader page shows a range up to 50% of subscribers' profit), and it is paid at the end of an investment period. We record the fee, the payout period and any minimum investment. We also remind readers that spreads, swaps, conversion and withdrawal costs can lower the result. A partner (affiliate) commission share is not a cost to the follower, and we never present it as performance.
5. Exposure and concentration
We look at which instruments drive the result (for example, 100% XAUUSD), how many positions are open, and floating losses. Three traders who all trade gold with high leverage may still be one concentrated bet. Having separate accounts does not remove correlated exposure.
6. Position sizing and copying mode
RoboForex offers three copying modes: Proportional (based on equity ratio × coefficient), Classic (trader volume × multiplier) and Fixed (a set lot size). Copied volumes that fall below an instrument's minimum, or off its volume step, are adjusted or rounded. We estimate the first copied order for a small allocation, because a $100 account that receives the same minimum lot as a $1,000 provider carries roughly ten times the relative risk.
7. Subscribers, balance and minimum allocation
The subscriber count tells us how much attention a trader gets, not how safe they are. We note the trader's own balance, the minimum investment the trader requires (the platform-wide floor is USD 100), and whether the strategy can realistically be reproduced at that size.
8. Exit and review rules
Before any subscription, the reader should write down a review date, a drawdown threshold that triggers a pause or a stop, how open positions will be handled, and a withdrawal rule (see Chapter 5 – Extraction).
The minimum evidence record and how we label it
| Area | Required record |
|---|---|
| Identity | Exact provider/account ID and source URL |
| Timing | Review date, plus the start and end of the measurement period |
| Performance | Definition of yield, equity and drawdown used |
| Cash flows | Relevant deposits, withdrawals and fee treatment |
| Replication | Platform, instruments, copying mode and allocation requirements |
| Behaviour | Sizing, concentration and material changes |
| Conclusion | What the evidence supports, and what remains unknown |
Every review carries an evidence status. The status describes the evidence. It is not an investment rating:
- Documented: the specific observation is supported by an accessible source. "Documented" does not mean safe.
- Incomplete: necessary data are missing, inaccessible or cannot be reconciled.
- Needs refresh: time has passed or conditions have changed, so the review may no longer describe the current offer.
Alongside the evidence status, we state the risk profile in plain words, for example "high risk: short history, single-instrument exposure, large recent drawdown". We never use "safe".
Facts, interpretations and predictions
"The profile shows this position size on this date" is an observation. "The strategy looks more concentrated than before" is an interpretation. "This trader will recover" is a prediction, and it does not become a fact because the trader recovered before. We keep these three visibly separate. We also review the adverse evidence (open losses, pauses, strategy changes, missing data). A case study keeps failed or closed silos rather than reporting only the survivors.
Limits of this method
- Figures are read from the public RoboForex profiles on the date shown. They can change minutes later.
- The broker-calculated yield is not the net return any individual copier received.
- Worked examples (the USD 100 walkthrough, the three-silo journal) are simulations and are labelled as such.
- This page describes our review standard. It does not claim that every trader has been assessed, or that a live monitoring service is running.
- A review is educational research, not personalised investment advice or a judgement on whether a strategy suits you.
Frequently asked questions
Do you rank the "best" RoboForex traders?
No. We publish dated watchlist snapshots and reviews with an evidence status. A ranking based on past yield cannot tell you which trader will perform next. When no current selection is published, that is simply the case. It is not an invitation to reuse an old list.
How often are trader figures updated?
Figures are not live. Every review shows the date its figures were checked and the measurement period. When time has passed or conditions have changed, the review is marked Needs refresh rather than silently re-dated.
Are you paid by the traders you review?
We may earn partner commission from RoboForex, including a share of some traders' commission through trader-specific partner links. A commercial incentive never justifies leaving out adverse evidence. See the Affiliate Disclosure.
Do you use AI?
Yes, for research, drafting, arithmetic checks and organisation. Sources, assumptions and corrections stay visible so you can check our work.
Capital at risk. Leveraged copy trading can result in the loss of all deposited funds. See the full Risk Disclosure.