Overfitting

Overfitting — When a strategy is so tuned to historical data that it fails on live trading.

Quick definition: When a strategy is so tuned to historical data that it fails on live trading.
Category: Concepts

Full Definition

Overfitting (or "curve-fitting") happens when a strategy has so many parameters or so much optimisation that it perfectly fits historical noise instead of real market structure. Symptoms: amazing backtest, terrible live results. Combat overfitting with: walk-forward analysis, Monte Carlo simulation, parameter sensitivity tests, and out-of-sample validation.

Related Terms

Walk-Forward Analysis · Backtesting

Test Your Strategy on Real Prop Firm Rules

Now that you understand overfitting, take the next step. JPTC's Prop Firm Checker tests your backtest against FTMO, FundedNext, FXIFY, TopStep and Apex rules — see whether your strategy would pass before paying a cent.

Open Prop Firm Checker →