Many backtests fail in live trading for a reason that is less glamorous than regime change or sophisticated statistics: the trade was never large enough to pay its own bills.

Every strategy competes against friction. Spread, slippage, commission, financing and execution rules remove part of the gross opportunity before the trader sees the result.

Two closely related parts of this decision are MFE and MAE Explained: What Trade Geometry Reveals That P&L Hides and Backtest vs Live Trading: 7 Reasons the Same Strategy Can Behave Differently.

Spread is variable in the real market

Using one fixed spread for every historical trade can be convenient and misleading. Spreads widen around news, illiquid sessions and stressed markets. A strategy that trades precisely when liquidity deteriorates may experience materially different economics from the average spread assumption.

Slippage is not random decoration

Fast entries, stop orders and thin markets can produce slippage that is directionally painful. The strategy may tend to buy when prices are accelerating upward and sell when they are accelerating downward, which means the fill model should reflect the conditions around execution.

Commission matters most when the target is small

A $5 commission is irrelevant to one strategy and fatal to another. The correct comparison is not the absolute fee. It is the fee relative to the movement the strategy expects to capture and the number of times it trades.

Financing changes longer-horizon economics

Overnight financing, swap and borrow costs can matter when positions persist. A strategy that looks profitable before financing can become much less attractive when the holding period and instrument-specific charges are modelled honestly.

Same-bar assumptions can flatter stops and targets

If a bar touches both the stop and target, the backtest needs a defensible rule for which occurred first. High-resolution price data can help. Simply choosing the favourable sequence creates a subtle form of optimism that compounds over many trades.

Capacity enters later

A strategy may work at small size and degrade when orders become large relative to available liquidity. Capacity is a different question from retail execution, but the principle is the same: the economic result depends on what can actually be filled, not what an idealized backtest says should have been filled.

Cost headroom is a useful idea

When QuantWeaver evaluates a candidate opportunity, one useful question is how much movement remains after realistic costs. If the opportunity barely clears friction under ordinary assumptions, there may be no margin for error. If it retains substantial headroom under stressed assumptions, the evidence is more interesting.