Last Updated: 2026-10-25
Risk warning: Both approaches can lose money. Automation does not transfer responsibility for losses or establish a reliable return. The figures below are synthetic arithmetic, not personal test results or financial advice.
Copy trading delegates trade selection to another trader or provider. A rule-based bot executes a specified rule. Either can run through an exchange’s own product or external software; that implementation determines permissions, custody and failure modes. The decision is which process you can inspect and control, not which label supposedly produces passive income.
Start with the action you must be able to take
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Requirement | Copy workflow | Rule-based workflow |
|---|---|---|
| Explain a new trade | Obtain the provider’s mandate and change history; motives may remain unknown | Read the rule, input data and trigger log |
| Limit exposure | Verify allocation, leverage and margin mode for the follower account | Verify position, order and aggregate exposure limits |
| Stop safely | Check whether stopping copying closes existing positions or only prevents new trades | Check cancellations, remaining inventory and a persistent halt state |
| Recover from a timeout | Reconcile follower orders/fills with the venue | Reconcile client IDs, fills and balances before retrying |
| Evaluate net results | Include follower slippage, profit share and open positions | Include subscription, execution costs and open inventory |
Do not assume the leader’s fill is available to followers, or that an accepted bot order filled. Product-specific behavior must be checked in current documentation. A read-only API cannot place orders; a trading-enabled API without withdrawal permission can still cause losses through trades. Neither configuration guarantees privacy or venue solvency.
Compare equal periods and cash flows
Suppose four fictional equal $800 allocations return 18.2%, 6.7%, −4.3% and 14.1% over the same period with no flows. The aggregate return is their mean, 8.675%, and profit is $277.60 on $3,200. It is not 11.4%.
A different fictional $4,500 allocation returning 8.6% earns $387. That is more dollars but a slightly lower percentage. It does not prove that bots outperform copied traders: the capital, risks, timing and strategy could all differ. The former purported personal comparison lacked the records needed to support that conclusion.
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Metric | Record needed |
|---|---|
| Return | Opening/closing equity, deposits, withdrawals, fees and common period |
| Drawdown | A chronological combined-equity series, not the worst component’s drawdown |
| Exposure | Positions and notional amounts at the same timestamp |
| Execution gap | Matching leader/signal and follower fill records, including missed/partial fills |
| Repeatability | Unselected later observations with the same rules, including failures |
Leverage adds notionals, not leverage labels
Three separate $1,000 allocations each holding $10,000 notional have $30,000 notional over $3,000 equity: 10× portfolio leverage, not 30×. If all are long the same asset, concentration is still high. Shared collateral, borrowed funds, changing equity and additional positions can change the calculation; inspect the actual account mode.
Two $500 allocations losing 42% and 38% lose $210 + $190 = $400, not $580. Smaller account size does not by itself justify a larger percentage position. Fixed fees and minimum orders may make a small allocation impractical, which is a reason to reconsider the setup rather than increase risk automatically.
A paper comparison that can fail honestly
Write the intended task, eligible product, permitted instruments, maximum exposure and stop conditions before looking at results. Log every selected signal and simulated fill with a timestamp and cost assumption. Mark stale data, unknown order states and unavailable fills as failures; do not replace them with a favorable assumed price.
Use a held-out period and a simple baseline. A short profitable period can be luck or market direction. A provider can change behavior, while a fixed bot can continue a rule after conditions change. Check open inventory and stressed scenarios rather than treating a smooth closed-profit curve as low risk.
Provider documentation is a place to verify mechanics, not evidence of profitability: 3Commas help center and the relevant venue’s current product terms. Use the fee worksheet to keep costs on the same turnover basis.
Affiliate disclosure: AITR may earn commissions from marked referral links. This comparison contains no provider ranking or referral recommendation, and makes no claim of a funded test.