Assessing Execution Quality for a Dealing Desk (DD)

What “execution quality” means for a Dealing Desk

Execution quality describes how closely trade outcomes match the terms a buyer expected when placing an order. For a Dealing Desk (DD), the key idea is that the execution process translates your order request into actual fills (executed price, quantity, and timing). A good assessment separates stable mechanics—how the system behaves—from variable external conditions such as market volatility and liquidity.

To assess execution quality, focus on measurable elements:

Fill price vs. a reference price at the moment you observe execution (e.g., the price you saw, or another consistent benchmark).
Fill completeness (full vs. partial fills) and time-to-fill.
Slippage: the difference between expected and actual fill price.
Total transaction cost as experienced (spread and any explicit/implicit fees).

“Dealing Desk” is not a single universal implementation; across providers, the practical behavior of order handling can differ even if they share the same label. So your goal is to measure outcomes you can observe, not rely on assumptions about internal decision-making.

How execution quality works: the main inputs you can test

A useful way to evaluate DD execution is to model each order as having expected terms and realized terms.

Define a reference and timing assumption Execution quality depends on when you measure. Examples of references you might use (choose one and apply consistently):

The quoted price shown at order submission.
The last available price at the moment the platform records the fill.
A benchmark from your own data capture.

Assumption needed for calculations: your reference price must correspond to the same event across orders. If it does not, your slippage estimate becomes a measurement artifact.

Compute realized cost and slippage For each order, compare expected vs. realized outcomes using a simple structure:

Slippage (price basis) = realized fill price − reference price (direction depends on buy/sell).
Effective spread / cost can be approximated by realized fill behavior relative to your reference and by adding any visible fees.

Assumption for examples: you know whether the reference is bid/ask and whether the fill direction aligns with that side.

Track fill behavior and order handling Execution quality is not only price. Record:

Whether you get a full fill or partial fill.
Whether fills occur at the requested time or after delays.
Whether your order is modified in response to market moves (for example, delays, re-pricing, or rejected fills).

This helps distinguish “price quality” from “execution reliability.”

Evidence you can use: a practical, comparable evaluation

Because you cannot assume future behavior from past patterns, evaluation should rely on comparisons that are as controlled as possible.

Start with a dataset approach:

Gather many executed orders over multiple days.
Group orders by approximate market state (for example, calm vs. volatile periods), using observable proxies like wide vs. narrow movement in your own reference series.
Keep order size and order type consistent where possible.

Then compute these metrics:

Average and distribution of slippage (not just the mean). A few extreme outcomes can distort averages.
Fill rate: proportion of orders fully filled.
Time-to-fill distribution: speed variability matters, especially during rapid price changes.
“Tail risk” checks: the frequency of large slippage events.

Realistic scenario-impact example (with explicit assumptions): Assume you submit 100 orders with the same reference definition. Suppose 10 orders occur during fast swings where your reference changes quickly. If those 10 show much larger slippage and partial fills, that may indicate a limitation of execution stability under volatility rather than a consistent baseline issue. The key is that your conclusions apply to the scenario you measured, not to all future markets.

Limitations and failure modes you should explicitly account for

A complete assessment must include at least one material limitation.

Market regime and liquidity change Execution quality can vary with volatility and liquidity. If you measure mainly in one regime, your results may not generalize. Historical relationships do not establish future results.
Measurement ambiguity If your reference price is not aligned to the true expected terms (bid vs. ask, submit time vs. fill time), you may misinterpret “slippage” as execution weakness when it is partly a timing or quote-side mismatch.