When people hear “algorithmic trading”, they often imagine systems predicting price moves. In institutional markets, a large share of algorithmic activity has a more modest goal: executing a decided trade at the lowest total cost. That is the job of execution algorithms.

The problem they solve

A large order placed all at once consumes available liquidity and pushes the price against the trader — market impact. Splitting it into smaller pieces reduces impact but extends exposure to price movement while the order is being worked. Every execution algorithm is a different answer to this trade-off.

TWAP: time-weighted

OpenCloseTrading day (illustrative) TWAP — equal slices VWAP — follows typical volume (U-shape)
Figure 1. Illustrative schedules for the same order. Bars show the share of the order executed in each interval; values are schematic.

A TWAP algorithm divides the order evenly across a time window — for example, equal slices every five minutes for two hours. It is simple, predictable and suitable when volume patterns are unknown. Its weakness is that it ignores when liquidity is actually available.

VWAP: volume-weighted

A VWAP algorithm schedules slices in proportion to expected trading volume, trading more when the market is typically active, such as near the open and close. Its benchmark — the volume-weighted average price over the period — is widely used to judge execution quality.

Implementation shortfall

Implementation shortfall algorithms measure performance against the price at the moment the decision to trade was made. They tend to trade faster when prices move unfavourably and more patiently when conditions are calm, explicitly balancing impact against the risk of delay.

Every execution algorithm is a different answer to the same question: how fast is too fast?

Measuring the result

Transaction cost analysis compares achieved prices against benchmarks and decomposes costs into spread, impact and timing. For systematic traders, it closes the loop between backtest assumptions and real-world execution.

Relevance for individual traders

Retail-sized orders in liquid markets rarely need sophisticated execution. But anyone trading less liquid instruments, or scaling a strategy, benefits from understanding these trade-offs — and from modelling realistic costs in backtests, as discussed in our guide to backtesting pitfalls.

Beyond the classic three

Modern execution suites offer many variants. Percentage-of-volume (POV) algorithms participate at a fixed fraction of market volume, speeding up when the market is active. Liquidity-seeking algorithms search across venues, including those that do not display orders publicly, to find large counterparties. Close and open algorithms target auction prices. Each is a different weighting of the same trade-off between impact, timing risk and opportunity cost.

Limit orders and the risk of not trading

Passive execution using limit orders can earn the spread rather than pay it, but introduces the risk that the order is not filled at all — particularly when the price moves away. For a systematic strategy, unfilled orders mean the live portfolio diverges from the backtested one. This opportunity cost is easy to overlook and is one reason simulated results often exceed live ones.

Execution in backtests

Most backtesting engines offer simple fill models: next bar's open, the close, or a fixed slippage in basis points. More realistic models scale slippage with order size relative to volume. Testing a strategy under several fill assumptions shows how dependent its results are on execution quality — a useful proxy for how fragile it will be live.

Quorient Editorial Desk

Articles credited to the Editorial Desk are researched and written for Quorient and reviewed by the editor, Callum Reid, before publication, in line with our Editorial Policy. Spotted an error? Request a correction.