Learn quantitative strategy concepts

Practical explanations of backtesting, bias, risk, and strategy evaluation, connected to reproducible QTSurfer workflows.

Building a strategy is only part of quantitative trading. You also need to understand what an experiment measures, which assumptions shaped the result, and how apparently strong performance can mislead you.

Learn provides practical explanations of the concepts behind QTSurfer. Each article connects the idea to a reproducible workflow rather than stopping at a dictionary definition.

Start here

• Quantitative strategy — what makes a strategy quantitative, its five components, and the research loop from hypothesis to validation.
• Backtesting — what a historical simulation can tell you, what it cannot, and which assumptions must travel with every result.
• Look-ahead bias — how future information can leak into a strategy and make a backtest impossible to reproduce in live trading.
• Overfitting — why the best result of a search is biased, and how to measure how much of it the search itself produced.
• Survivorship bias — how the universe, the catalogue, and the published record quietly exclude the failures.

Reading a result

• Equity curve — what the account-value series records, how to normalise it, and which shapes signal fragility.
• Drawdown — the fall from a previous peak, its depth and duration, and why it changes how a return should be read.
• Sharpe ratio — return per unit of variability, the conventions that make two values comparable, and exactly how QTSurfer computes it.
• Slippage — the cost between the decided and the obtained price, and how to account for it in a backtest.

Exploring parameters

• Parameter sweep — grid, random, and Latin hypercube sampling, leaderboards, plateaus, and sensitivity views.
• Walk-forward analysis — sequential out-of-sample validation of a sweep and what parameter drift reveals.

Strategies

• EMA crossover — the reference trend-following rule: lag, whipsaw, useful filters, and how to implement and sweep it.
• Strategy revision — immutable versions of a strategy’s code, and why every result should point at one.
• Algorithmic trading — from decision to order: what a trading system needs beyond the signal, and the backtest-to-live gap.

Data

• Historical market data — tickers versus candles, cadence, coverage and gaps, and how QTSurfer stores and serves exchange history.