What “execution quality” means for clone firms
Execution quality describes how closely a cloned account’s order execution matches the referenced (source) strategy’s execution, given the same general market environment. For clone setups, that match is never perfect because the clone has its own order entry, order routing, and execution timing.
A practical assessment focuses on measurable execution outcomes rather than labels like “good” or “bad.” Typical execution factors include:
- Fill quality (how close fill prices are to intended prices)
- Timing (how quickly orders are placed and updated)
- Order completeness (full vs partial fills)
- Cost drag (spreads, commissions, fees, and execution slippage)
- Consistency of mapping (how the system translates signals or orders into the clone’s actual orders)
The key idea is to separate stable system mechanics from variable conditions that can change between the source and the clone.
Mechanism: which inputs should be evaluated
To assess execution quality, define the comparison you will measure. A simple approach is to compare order-level facts between:
- the source/reference execution (the strategy being cloned), and
- the clone’s execution (the account receiving the mirrored actions).
When data is available, evaluate these inputs:
- Order timing alignment: Compare event timestamps for order submission and execution. If timestamps differ, you can quantify the resulting opportunity for price movement.
- Price capture: Compare intended execution price (or decision price) to actual fill price for both source and clone.
- Slippage decomposition: Separate slippage caused by market movement from slippage caused by system delays or routing differences.
- Fill rate and partial fills: Measure how often orders are partially filled and whether the clone completes execution in the same time window as the source.
- Cost components: Track total execution costs per trade, including spread/mark-up effects where observable and any per-order fees.
Assumption for examples: if you do not have both source and clone order-level data, you must use whatever aggregated trade logs you have and accept reduced precision. Document what is missing before concluding anything.
Evidence and example scenario you can measure
Consider a realistic scenario: a clone system receives a “buy” instruction at time T, but the clone account’s order is entered at a later time T+Δ. In a fast-moving market, price can move between T and T+Δ.
What you can measure:
- Latency impact: Compute Δ from your available timestamps (submission time or first observable action).
- Fill difference: Compare source vs clone fill prices for the same intended action.
- Cost difference: Convert price differences into an approximate cost impact using the trade size and a consistent formula (state your assumption: e.g., linear conversion for a given contract size, or use P&L deltas from logs if provided).
- Consistency: Repeat across many trades, not just one. Look for systematic bias (e.g., clone fills are always worse by a similar amount) rather than isolated deviations.
Material limitation: even if the clone matches the source closely under one market regime, execution quality can deteriorate under different volatility, liquidity, or news conditions because slippage and delays respond to market microstructure.
Limitations, failure modes, and verification checkpoints
At least one material failure mode is common in clones: execution divergence due to timing and order-handling differences.
Common failure modes to look for:
- Delay-driven slippage: the clone executes after the source, causing worse prices.
- Partial fill and remainders: the clone’s order size or liquidity differs, changing fill completion time.
- Inconsistent order mapping: different order types (market vs limit), time-in-force, or rounding rules can lead to different actual outcomes.
- Data misalignment: missing timestamps or inconsistent trade IDs can make source vs clone comparisons misleading.
- Cost asymmetry: even if the fill price is similar, spread/fees can differ, changing net results.
Verification checkpoints:
- Use a time-bounded test set (a defined period) and compare source vs clone at the same event granularity you can reliably extract.
- Check robustness across market conditions (quiet vs volatile). If your assessment relies on one calm period, it may not generalize.
- Treat historical relationships as descriptive only. Similar past execution patterns do not establish future execution quality.
Finally, avoid equating execution quality with guaranteed outcomes. Execution quality is about measurable matching of orders and fills—not about predicted profits or safety.