Introduction
Backtesting is the method of simulating a trading strategy using historical data to generate and analyze results without risking any capital. The reason retrospective analysis is so key to a well-balanced trading strategy is that anything that worked well in the past, is likely to work well in the future.
Sources of Historical Data
Backtesting is reliant on historical data, which is often provided by brokerages or financial solution companies. Some of the most common places to conduct retrospective analysis for free are TradeIdeas, TradeStation, and TradingView just to name a few.
Common Measurements in Backtesting
Retrospective analyses typically provide results that encapsulate the trading strategy as a whole. Here are some of the common measurements given when backtesting:
- Net Profit/Loss
- Percent Profitable
- Volatility Measures
- Average Drawdown
- Annualized Return
Retrospective Analysis Rules
Retrospective analysis is extremely customizable, thus the results could be insignificant or made deceitful. It is important that you do not overfit when conducting retrospective analysis. Overfitting a retrospective analysis happens when the simulation is based on too small, or too specific a set of data. If a retrospective analysis model becomes overfit then it may lose its value as a predictive tool.
Proper Backtesting Analysis
A proper backtesting analysis takes into account at least three factors:
- Evaluating performance under bear, bull, and horizontal market conditions.
- Assessing strategy efficacy across various sectors.
- Examining strategy performance over different timeframes.
Retrospective Analysis Example & Case Studies
Here is a screenshot of one stock that PeakBot backtested it is Mont Blanc algorithmic trading strategy with. This retrospective analysis was performed on NVDA using 5 minute bars and a 10/22/2021 – 2/07/2022 timeframe using $1,000 position sizes.
Throughout the timeframe the strategy would have returned you 13.21% profit on a total of 34 trades with a 64.71% win rate.
A compilation of retrospective analyses is often referred to as a case study. Here is a Mont Blanc case study of 20 stocks throughout Q4 2021 and the automated strategies performance on those companies.
