What Is Cost drag?

Cost drag is the cumulative reduction in a trading strategy’s performance caused by transaction costs — commissions, bid-ask spreads, slippage, market impact, borrowing costs for shorts, and exchange fees. It represents the gap between a strategy’s gross (paper) returns and its net (implementable) returns. For high-turnover strategies, cost drag can easily consume 30–60% of gross alpha, turning a seemingly attractive backtest into a net-negative live system. This makes cost drag one of the most important diagnostics for separating robust strategies from fragile paper optima.

The components of cost drag interact in non-obvious ways. Bid-ask spread costs scale linearly with turnover, but market impact costs scale super-linearly with trade size relative to available volume — meaning that doubling position size more than doubles the impact cost. For small-cap or illiquid instruments, even modest rebalance volumes can move the market against the strategy. Models like Almgren–Chriss provide a framework for estimating these costs as a function of trade size, urgency, and daily volume. The practical implication is that strategies optimised for maximum gross Sharpe often have very different optimal parameters when costs are included, because the cost-minimising and alpha-maximising objectives are in tension.

In strategy development, cost drag analysis should be performed during the optimisation process, not as an afterthought. Reporting gross and net returns side by side, along with the turnover and average cost-per-trade assumptions, allows allocators to assess whether a strategy’s edge survives implementation. A strategy with a gross Sharpe of 2.0 but 400% annual turnover in illiquid mid-caps may net a Sharpe below 1.0. Conversely, a strategy with a gross Sharpe of 1.2 and 50% annual turnover in liquid large-caps may net a Sharpe of 1.1 — making it the superior live choice despite appearing weaker on paper. This is why cost drag, turnover, and rebalance volume should always be considered as a group.

Pros

  • Reveals the true implementability of a strategy by bridging the gross-to-net gap

  • Catches “paper optima” that look great in backtest but cannot survive real trading

  • Forces realistic assumptions about execution quality into the research process

  • Helps size positions and set rebalance frequency to maximise net (not gross) returns

Cons

  • Accurate cost estimation requires detailed market microstructure data (spreads, volumes, impact)

  • Cost models can themselves be overfit to historical conditions that may not persist

  • Costs vary significantly across brokers, venues, and market regimes — a single estimate may not generalise

  • Non-obvious interactions between impact, timing, and order routing make precise estimation difficult

See also