Key takeaways
- Master and follower matches are inferred candidates, not broker-declared links.
- A timestamp gap measures an observed difference but does not prove its cause.
- Size rounding and fees can change follower outcomes even when candidate trades align.
A follower can finish with a different result from the master because the observable trades differ in timing, execution price, rounded size or fees. Before attributing the gap, remember that master/follower pairs are inferred candidates, not a copy relationship declared by the broker.
1. Fill lag
TradeStats forms candidates only from normalized symbol, direction and open-time proximity after applying the estimated clock offset. It searches within a 120-second window with 5-second jitter, then uses maximum-cardinality one-to-one selection so each master and follower trade can appear in at most one pair. Volume is not an input to candidate selection.
That method may still mispair similar trades or leave a valid copy unmatched. A timestamp gap between inferred candidates does not prove a copier, broker, VPS or liquidity cause.
2. Entry/exit slippage
Once a candidate pair is accepted, the master and follower entry or exit prices can be compared in the same normalized symbol. Signed price difference shows whether the follower price was better or worse for that candidate, but the result inherits any uncertainty in the pairing and symbol normalization. The execution-quality guide explains the separate requested-price slippage method used for broker fills.
3. Size ratio and rounding
Worked example: lot-step rounding
At a 0.4× follower ratio, a 0.01-lot master trade implies 0.004 lot. If the follower account only accepts 0.01-lot steps, the copier must round, skip or accumulate the order. That decision changes exposure even if timing and price are identical. TradeStats calculates the volume or size ratio only after matching; the ratio describes an accepted inferred pair and does not help select it. Compare size ratios across many inferred pairs rather than assuming every trade can preserve the target ratio.
4. Fees
A follower can pay a performance, subscription, commission or financing cost that is absent from the master's displayed result. Compare gross trading outcome and follower net outcome separately, using the copy-trading fee guide to identify the applicable charging model.
Measuring all four at once
A useful report shows matched and unmatched counts, the matching window, inferred timing gaps, price differences, size ratios and fees. It should not turn one observed difference into a causal verdict.
Method and limitation note: the pairing algorithm produces inferred candidates from observable normalized symbol, direction and clock-adjusted time patterns. It may mispair or miss trades, especially when several similar orders occur close together. Use the evidence alongside the track-record verification checklist.
Frequently asked questions
How much lag is normal for copy trading?
There is no universal healthy range. Strategy, copier design, broker, symbol, session and clock quality all matter. Compare a distribution of inferred pairs and inspect unmatched trades rather than using one fixed cutoff.
Does a master-to-follower timestamp gap prove the copier was slow?
No. It is an observed gap between inferred candidates. Clock offset, broker timestamps, routing, mapping and an incorrect pairing can all contribute, so the gap alone does not identify one cause.
Can the matching process pair the wrong trades?
Yes. Matching is inferred from observable fields rather than a broker-provided copy ID. Similar trades close together can mispair, and the report should expose that limitation.