Fills, stops & modeled P&L

A backtest is only as honest as its fill assumptions. This page states ours — including what we deliberately do not claim.

How simulated fills work

• Entries and exits are priced from the quoted market at that second — mid-based marks with slippage haircuts applied per leg.
• The haircuts are asymmetric: exits — especially stop exits, where you're crossing the spread under pressure — are haircut harder than entries. That asymmetry matches what live execution data shows.
• Stop exits are recorded at the stop level as the modeling convention: when the cost-to-close crosses the configured stop, the loss is booked there. Live stops can fill worse than the configured level (see the live numbers below) — treat backtested stop losses as the intended loss, not a guaranteed fill.

What "modeled P&L per contract" means

Catalog figures are modeled P&L per contract. We deliberately do not label them "after fees" or "including slippage" as a blanket claim — execution costs are modeled, but your commissions, your broker's routing, and the market's mood on your particular day are yours. The honest reading of any catalog number: this is what the strategy's rules produced against historical quoted markets under our stated modeling conventions.

How we keep the model honest: live measurement

CashFlow Engine trades these strategy families on its own automated accounts, and every live fill is measured against the quoted market:

• Each fill is compared to the quote midpoint at fill time, using a reference quote only when one exists within ±30 seconds — stale references are discarded, not approximated. Positive slippage = cost.
• Across 30,000+ measured fill records (roughly a year of live execution): entry fills averaged $3.40 per contract worse than mid (median $2.00), and 38.8% filled at or better than mid.
• Stops are the expensive part: realized stop fills came in worse than the configured stop target by a median of about $15 per contract, with a fat tail on fast days. This is precisely why the previous section tells you to read backtested stop losses as intentions.

This live-vs-modeled reconciliation is the discipline behind the fill model — and the full fill-by-fill comparison ships as Reality Check on Overdrive.

What this means for your expectations

• Simulated results sit somewhere between "theoretical ceiling" and "live reality" — closer to reality than naive mid-fills, but not a promise.
• The gap is strategy-dependent: high-frequency-of-stop configurations carry more execution drag than configurations that mostly settle.
• If an edge in the catalog is so thin that a few dollars of per-contract slippage erases it, it is not an edge — filter accordingly.