No decision without data.
Every hypothesis we test begins in the academic literature and ends in the data, where the decision is made.
Our research protocol.
The gates an idea must pass before it becomes a rule, from the literature to untouched data.
-
Empirical findings from several academic studies are mapped and combined into one coherent market mechanism.
-
That mechanism is turned into a falsifiable hypothesis, with fully defined entry, exit, and sizing rules.
-
The rules are evaluated on intraday data from the development period to measure their expectancy, distribution of results, and consistency.
-
The results are tested against parameter changes and sub-period splits, so the edge does not depend on a single configuration.
-
Significance is judged against thresholds corrected for multiple testing, and the uncertainty is estimated with the bootstrap.
-
Final performance is evaluated on data not used during development, with no adjustment to the rules.
- Harvey, C. R., Liu, Y., & Zhu, H. (2016). …and the cross-section of expected returns. The Review of Financial Studies, 29(1), 5–68.
- Harvey, C. R., & Liu, Y. (2015). Backtesting. The Journal of Portfolio Management, 42(1), 13–28.
- Bailey, D. H., & López de Prado, M. (2014). The deflated Sharpe ratio: Correcting for selection bias, backtest overfitting, and non-normality. The Journal of Portfolio Management, 40(5), 94–107.
- Bailey, D. H., Borwein, J. M., López de Prado, M., & Zhu, Q. J. (2014). Pseudo-mathematics and financial charlatanism: The effects of backtest overfitting on out-of-sample performance. Notices of the American Mathematical Society, 61(5), 458–471.
- White, H. (2000). A reality check for data snooping. Econometrica, 68(5), 1097–1126.
- Sullivan, R., Timmermann, A., & White, H. (1999). Data-snooping, technical trading rule performance, and the bootstrap. The Journal of Finance, 54(5), 1647–1691.
- Efron, B., & Tibshirani, R. J. (1993). An introduction to the bootstrap. Chapman & Hall.
- Künsch, H. R. (1989). The jackknife and the bootstrap for general stationary observations. The Annals of Statistics, 17(3), 1217–1241.
- Newey, W. K., & West, K. D. (1987). A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica, 55(3), 703–708.
- Petersen, M. A. (2009). Estimating standard errors in finance panel data sets: Comparing approaches. The Review of Financial Studies, 22(1), 435–480.
- Lo, A. W. (2002). The statistics of Sharpe ratios. Financial Analysts Journal, 58(4), 36–52.
- Nosek, B. A., Ebersole, C. R., DeHaven, A. C., & Mellor, D. T. (2018). The preregistration revolution. Proceedings of the National Academy of Sciences, 115(11), 2600–2606.
- Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2011). False-positive psychology: Undisclosed flexibility in data collection and analysis allows presenting anything as significant. Psychological Science, 22(11), 1359–1366.
- Pardo, R. (2008). The evaluation and optimization of trading strategies (2nd ed.). John Wiley & Sons.
- Park, C.-H., & Irwin, S. H. (2007). What do we know about the profitability of technical analysis? Journal of Economic Surveys, 21(4), 786–826.
- Grossman, S. J., & Stiglitz, J. E. (1980). On the impossibility of informationally efficient markets. American Economic Review, 70(3), 393–408.
- Lo, A. W. (2004). The adaptive markets hypothesis: Market efficiency from an evolutionary perspective. The Journal of Portfolio Management, 30(5), 15–29.
- Neely, C. J., Weller, P. A., & Ulrich, J. M. (2009). The adaptive markets hypothesis: Evidence from the foreign exchange market. Journal of Financial and Quantitative Analysis, 44(2), 467–488.
- McLean, R. D., & Pontiff, J. (2016). Does academic research destroy stock return predictability? The Journal of Finance, 71(1), 5–32.
- Harris, C. R., Millman, K. J., van der Walt, S. J., et al. (2020). Array programming with NumPy. Nature, 585, 357–362.
- McKinney, W. (2010). Data structures for statistical computing in Python. Proceedings of the 9th Python in Science Conference, 56–61.
- Seabold, S., & Perktold, J. (2010). Statsmodels: Econometric and statistical modeling with Python. Proceedings of the 9th Python in Science Conference, 92–96.
- Hunter, J. D. (2007). Matplotlib: A 2D graphics environment. Computing in Science & Engineering, 9(3), 90–95.
The portfolio we built.
Seven strategies with low correlation and equal risk allocation. Backtest, Jan 2006 – Sep 2026, after costs, non-compounding.
Every component, open to inspection.
Pick a strategy to see its backtest performance. Every strategy uses 1% risk per trade, after costs.
All the research, from hypothesis to code.
Every document that explains how the strategies were built, plus the EAs and scripts to run them in MetaTrader 5 and TradingView.
These seven strategies are internal research by Trade By Data and have never been published on social media or any other platform.
- Seven strategies that passed the Trade By Data research protocol, for gold, silver, USD/JPY, and the Nasdaq.
- Complete rules for each strategy: entry, exit, stop loss, and sizing, with example trades on the chart.
- Full performance: equity, drawdown, and annual and monthly returns for each strategy and the portfolio, after costs.
- Portfolio risk allocation: the risk per trade of each strategy when they run together.
-
- The literature and hypotheses behind it
- Data and test design
- In-sample results and parameter robustness tests
- Statistical tests: significance, the bootstrap, and multiple testing correction
- Out-of-sample validation
- The weaknesses and limitations found
- MetaTrader 5 EA files for every strategy, with their source code, presets, and installation guides.
- TradingView strategy and indicator scripts (Pine Script) for every strategy, with their source code, alerts, and user guides.
Price
Strategies & Research - Trade By Data
- 7 strategies with full rules and performance
- MetaTrader 5 EAs with source code, presets, and installation guides
- TradingView strategy and indicator scripts with source code, alerts, and user guides
- Research-process documents for all seven strategies, from the literature to out-of-sample validation
No subscription. The files are yours to download right after checkout.
Secure checkout by Gumroad. All purchases are final.
*Make sure you use an active email address at checkout.
FAQ
You can run the EAs or TradingView scripts directly, use the rules and findings as a starting point for your own strategies, or study the research documentation to understand how each strategy was built and tested from start to finish.
No. You run the EAs in MetaTrader 5 or the scripts in TradingView yourself, with full control over the risk per trade.
These seven strategies are internal research by Trade By Data and have never been published on social media or any other platform.
MetaTrader 5 with a hedging account, at a broker that offers gold, silver, USD/JPY, and Nasdaq-100 CFDs.
Start from the drawdown you can tolerate. In the historical simulation, the portfolio’s maximum drawdown was about 6% on Conservative, 12% on Moderate, and 24% on Aggressive. The risk per trade of each strategy at each profile is listed in the portfolio section.
Yes, because each EA and script runs on its own. The portfolio figures on this page apply when all seven strategies run together with the listed risk allocation.
Non-compounding separates the quality of the edge from the effect of position size, so risk can be compared directly across strategies and profiles. Every EA and TradingView strategy script still has a compounding option if you want position size to follow your running equity.
No. The rules and parameters you receive are the versions that were frozen after passing the research protocol.
Yes. Every EA and TradingView script comes with its source code, so you can check its logic yourself.
Right after checkout, Gumroad emails you a link to download the files, so please make sure the email address you use is active.
All purchases are final.