2018issue C0932-35
Critiquing breakout systems through robustness tests
The archive workflow presents a breakout-system as transferable across markets and holding periods, then uses robustness-testing to reject most early candidates as overfit material. Extra false-breakout filters are said to worsen that overfitting. Timing-layers and scenario-based dynamic-risk are the stated way to stand aside.
- System work is said to begin by combining incomplete ideas, conditions, risk techniques, and indicators, then rejecting most of them through robustness-testing.
- A breakout-system is one procedure for entry, exit, and standing aside, not a growing stack of false-breakout filters.
- Failed follow-through is framed as a timing problem across general, system-level, volatility, and market views, while dynamic-risk can cut size or skip a trade.
- After the beginner stage, attention moves to a small portfolio of low-correlation robust systems rather than a search for a currently winning rule set.
How breakout work is said to start
Breakout systems are presented as transferable across markets and timeframes, including both day and swing holding periods. System work is said to start by combining incomplete ideas, conditions, risk techniques, and indicators, then checking those candidates on historical data.
About 99% of early candidates are characterized as overfit material, so robustness-testing is the step that must separate unusable ideas from potentially viable systems. In the terms used here, robustness-testing is a demanding validation procedure used after ideation to separate overfit strategy candidates from systems that may remain usable.
Where systems are said to fail
The main stated reason systems fail is the absence of a comprehensive robustness-testing procedure. Commonly used validation methods are described as no longer sufficient. Most of the described development time, about 80%, is spent improving robustness-testing rather than searching for a currently winning rule set.
Later windows test the same procedure
Walk-forward-analysis is a sequential out-of-sample check that re-estimates rules on later data windows to test whether the same procedure still produces usable signals. TradersWeek editorial view: apply that later-window check to the same entry, exit, and abstention rules. A later window is not a reason to attach another filter.
False-breakout filters inflate the pile
A student survey is cited as finding false breakouts the top breakout-system problem, and filtering them out is said to increase overfitting instead of fixing the procedure. A false-breakout is a breakout signal that fails to follow through. Stacking extra filters is presented as a path to overfitting rather than a remedy.
Four timing-layers, not more entries
Reducing false breakouts is framed as a timing problem across four layers: general, system-level, volatility, and market. Those four timing views are the timing-layers used to reduce failed breakouts without adding signal filters. TradersWeek editorial view: if the procedure cannot stand aside, another entry condition is the wrong repair.
Scenario-based risk includes a skip
Risk is varied by scenario through an automated procedure: as much as 1.5 to 2% when conditions look favorable, about 0.5% or less when unpersuasive, and a full skip when highly unfavorable. That automated change in per-trade risk, including a full skip, is dynamic-risk driven by how favorable the current market scenario appears. TradersWeek editorial view: the skip belongs inside the breakout-system rather than beside it.
A small robust set after the beginner stage
Futures are preferred over equities and indexes because dissimilar contracts are described as easier to combine into lower-correlation portfolios. After the beginner stage, attention is directed to a portfolio of a few low-correlation robust systems, with three to five named as a sufficient starting set. That intermediate shift from single-strategy tinkering to a small set of low-correlation robust systems is the portfolio-stage.
All readings on this track · 51 readings
- 1986Degrees of freedom in trading system optimization
- 1988Walk-forward and neighborhood tests after optimization
- 1988Undisclosed rules block system robustness tests
- 1988Testing re-optimization calendars against random parameter controls
- 1989Binary search limits on multi-peak average grids
- 1989Parameter neighborhoods that survive a shift
- 1990Use profit mapping to keep a cycle and stop plateau
- 1990Why popular indicator optimization fails robustness
- 1991Retesting weighted indicator balances across horizons
- 1992Constructing forecast models with regression, walk-forward, and robustness
- 1992Diagnose regimes before you lock parameters
- 1992When stops change system timing
- 1993Walk-forward halt rules for forecast models
- 1994Walk-forward evaluation of genetic index rules
- 1995Input pruning as walk-forward system evaluation
- 1995Critiquing neural nets as incomplete trading systems
- 1996Rebuild the equity-path ratio before it ranks a designed system
- 1996Parameter grids can fit random walks
- 1996Walk-forward analysis belongs in the design of a mechanical trading system
- 1997When a holdout fails, discard the rule set
- 1997Test rewarded rule breaks before replacing the system
- 1997Walk-forward rules keep system research from rewriting live trades
- 1999Keep a channel-breakout to two lookbacks and test neighbor stability
- 1999Constant investment size in stock system evaluation
- 2000Forcing optimization maps mechanical system failure boundaries
- 2000Robust parameter selection with surface charts
- 2001A two-gate classroom test for a two-window momentum trend filter
- 2002How a two-sided continuation factor becomes a testable trend rule
- 2002Evaluating two-window trend intensity as a reversal rule
- 2003Discounting speculative bubbles in system robustness tests
- 2003Walk-forward evaluation of locked stochastic oscillator rules
- 2003Critiquing mechanical system design after extreme price regimes
- 2004Evaluating a two-window trend trigger
- 2005Grade backtested signals with holdouts and optimization plateaus
- 2006Reserved-sample evaluation of trading system design
- 2006Walk-forward critique of hindsight crossover systems
- 2008Condition-matched walk-forward evaluation for mechanical systems
- 2011Session-split evaluation of regular and overnight systems
- 2012Walk-forward evaluation as operator rehearsal
- 2013Two-window evaluation of mechanical trading systems
- 2013Walk-forward filter selection for repeated-median velocity
- 2014Walk-forward evaluation for fading-memory velocity systems
- 2015Test oscillator events before tuning rules
- 2016Walk-forward evaluation of a five-parameter parabolic stop-and-reversal
- 2016Walk-forward optimization without curve fitting
- 2017Optimization without overfitting in trend-system evaluation
- 2017Parameter stability is a better guide than a larger crossover grid
- 2018Point-in-time universes for system evaluation
- 2018Walk-forward robustness evaluation for optimized systems
- 2018Critiquing breakout systems through robustness tests
- 2018A critique of parameter fitting in system design