Last Updated: 2026-10-25
Risk and scope: This is a documentation-based review worksheet, not a funded test or a recommendation to copy a trader. Futures and leverage can cause substantial losses. Eligibility, available settings and account terms must be checked for your exact product and region.
The useful question is not how many “elite” traders a directory contains. It is whether you can explain the follower's exposure, execution and net result. The earlier article's personal portfolio, latency measurements and alleged performance-driven Smart Copy algorithm were not supported by inspectable records. They are not used here.
Do not mix Classic and New Copy Trading
Bitget's copy-order limits guide, checked October 25, 2026, distinguishes Classic modes from New Copy Trading's fixed-ratio and fixed-margin modes. It describes orders being capped, partly copied or rejected; an unevenly copied hedge can leave directional exposure.
The New Futures Copy Trading guide describes separate copy accounts, a high-water-mark profit-sharing model, and a stop-copy choice between market closure and following the trader's closing strategy. It also describes cross margin in that version. These statements do not establish the behavior of a different version or account.
Before proceeding, save the product/version, guide date and the actual confirmation screen. If the interface and documentation disagree, resolve that discrepancy with the provider rather than substituting a setting from another tutorial. A label such as “Smart” does not prove that sizing adapts to a winning streak or reduces drawdown.
Five settings that need an explicit answer
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Setting | Record | Failure to investigate |
|---|---|---|
| Sizing mode | Whether the input is a ratio, margin per order or total allocation | Mistaking a $20 margin input for a $20 position |
| Margin and collateral | The account whose equity supports open positions | Assuming isolated risk when shared collateral is used |
| Instruments and leverage | Permitted pairs, new-listing behavior and available controls | New correlated positions increasing aggregate exposure |
| Order limits | Requested versus filled quantity and any rejection reason | One copied leg being smaller than its intended hedge |
| Stopping and settlement | Selected closure option, remaining positions and unsettled charges | Treating “unfollowed” as proof that exposure and settlement are finished |
After a setting changes, compare the intended state with the resulting account state. Neither a badge nor a control name establishes a maximum possible loss. A stop can encounter slippage; a platform limit may prevent an entry rather than close an existing position.
A cost example with one consistent unit
Assume a fictional position is bought for $1,000 notional and sold for $1,020. Use a hypothetical 0.06% fee on each fill and $0.90 funding paid. These are teaching inputs, not Bitget's current tariff.
Scroll sideways to see every column. Keyboard: focus the table region and use the arrow keys.
| Item | Calculation | Amount |
|---|---|---|
| Gross price gain | $1,020 − $1,000 | $20.000 |
| Entry fee | $1,000 × 0.0006 | −$0.600 |
| Exit fee | $1,020 × 0.0006 | −$0.612 |
| Funding paid | Assumed separate charge | −$0.900 |
| Net before profit sharing | Sum above | $17.888 |
If an entirely hypothetical contract charged 10% on that positive net amount, the charge would be $1.7888 and $16.0992 would remain. Do not assume this is the provider's actual settlement base or timing. Read the applicable high-water-mark rules and reconcile actual transactions, including prior losses and deposits. If a reported return already includes a charge, subtracting it again is wrong.
Reducing a fee from 0.06% to 0.051% saves (0.0006 − 0.00051) × $1,000 = $0.09 on $1,000 of filled turnover, not $0.90. Buying or holding a token to obtain a discount adds price exposure; it is not free money. Separate token gains/losses from any fee savings.
Review the follower record, not an anonymous portfolio table
Keep opening and closing equity, dated deposits/withdrawals, every filled and failed copy, open positions, fees, funding and settled profit share. A fall in assets under management can reflect withdrawals, losses or both. It is not by itself evidence that other users are abandoning a trader.
Weighted returns require comparable periods and known capital flows. A position started in January and another started in April cannot become a single five-month portfolio return by averaging their displayed percentages. Aggregate drawdown requires a time series of combined equity. Compare the leader and follower only after aligning periods and what their metrics include.
Use the failure-mode guide for stop, leverage and execution examples, and the fee worksheet for a full-turnover comparison. Missing history or an unclear control is a reason to leave the assessment unresolved, not to invent a score or increase the allocation.
Disclosure: AITR may earn commissions from marked referral links elsewhere. This page makes no claim of tested returns, a best trader, a recommended allocation or guaranteed loss protection.