Sobreoptimización

Learn why overoptimizing trading strategies is a common pitfall. This guide explains the dangers of chasing maximum backtested returns and how to build robust,

Published · Updated · Methodology: Mixed

Part of: Algorithmic & Automated Trading

  • Methodology: Mixed
  • Content type: educational

Source video

Decoded from: El Error de Elegir los Mayores Beneficios: Evitando la Sobreoptimización by Bfunded EA — watch the original

Strategy overview

Overoptimization — curve fitting — is what happens when a strategy's parameters are tuned until they describe the past too well to say anything useful about the future. What distinguishes this entry is where its title places the error: not in the testing, but in the choosing. "El Error de Elegir los Mayores Beneficios" ("The Mistake of Picking the Biggest Profits") points at a specific moment in a workflow — the one where an optimization report is already on screen, sorted by result, and a human picks a row.

That moment deserves the attention. A parameter sweep does not produce one strategy; it produces hundreds, and the top of the ranking is by construction the combination on which noise fell most favourably. Selecting it uses the same data that ranked it, so the number that justified the choice has already been spent. The related question the title implies — whether a top result sits on a narrow peak or a broad plateau of neighbouring settings that behave similarly — is the difference between a parameter set and a lucky coordinate.

The fields on this entry are blank in a way that is coherent rather than incomplete: no indicators, no timeframe, and a Mixed methodology, because the failure mode described is indifferent to which indicator or bar size was optimized — it attacks whatever you tuned. For the same reason no trading rules were extracted here; there is no entry or exit to extract, since the material is a criterion for reading a results table, not a setup. The Spanish-language source comes from Bfunded EA, a channel oriented toward expert advisors and funded-account trading — the context where optimization is a routine build step and where a proprietary firm's evaluation phase ends up serving as the out-of-sample test, with consequences attached.

Topics

overoptimization · trading strategy · trading psychology · strategy development · backtesting pitfalls · realistic trading · robust strategy

Frequently asked questions

What is overoptimization (curve fitting) in trading?

It is tuning a strategy's parameters until its historical results look excellent, at the cost of fitting the noise in that specific data rather than any repeatable behaviour. The symptom is a backtest that degrades sharply on data it was not tuned on.

Why is choosing the highest-profit result from an optimization a mistake?

Because the top of a ranking is selected using the same data that produced the ranking. Across hundreds of parameter combinations, the best-performing one is also the one that benefited most from chance, so its headline profit is the least reliable estimate of what comes next.

How do you tell an overoptimized strategy from a genuinely good one?

Standard practice is to test on data held out from the tuning, and to look at the neighbourhood of a parameter set rather than a single point — settings whose nearby values perform similarly are more credible than an isolated peak surrounded by poor results.

Does this entry include a specific trading setup?

No. No rules, indicators, or timeframe were extracted for this entry, which is consistent with its subject: it addresses how to select among optimization results rather than describing an entry and exit. Strategy Decoder catalogues methodology material like this alongside the rule-based strategies it decodes from video sources.

About this strategy page

This trading strategy was decoded by Strategy Decoder's AI from a public YouTube trading video and turned into a structured, reviewable specification. In the interactive app this page shows the full entry and exit logic, risk management settings, the indicators involved with their parameters, AlgoWizard-compatible logic and a Pine Script export ready for TradingView backtesting — plus an automated backtest verdict when one has been computed for this strategy.

Strategy Decoder catalogs 2,229 decoded strategies. Each one is extracted with confidence scoring, cross-linked to the indicators it uses, and kept up to date as new videos are processed daily. Load this page with JavaScript enabled to use the interactive tools, or start from the strategy explorer to filter by methodology, market and timeframe.

Other versions of this strategy

More decoded strategies