Research scope
- Roughly 9 years of US stock market history
- 30,000+ historical gap events of 40% or more
- Focus on extreme gap behaviour and intraday reversal context
- Designed specifically around backside short logic in micro-cap gappers
Research framework
Methodology behind the GBS model.
GBS was built to study a specific market behaviour: the tendency of some extreme micro-cap gappers to lose upside momentum and reverse after highly extended moves. The goal was not to create a one-line pattern detector, but to build a more systematic framework for identifying reversal context from a large historical sample.
This page explains the public methodology at a conceptual level. It is intended to show how the framework was designed and validated, while keeping the exact implementation and parameter recipe proprietary.
On this page
What is being studied
Extreme gappers are a recurring market phenomenon.
Micro-cap and low-float stocks can experience disorderly price expansion over a very short period of time. Those moves can be dramatic, but they are not rare enough to be treated as one-off anomalies. They appear repeatedly across trading sessions.
GBS was built around the observation that some of these extended moves eventually begin to show similar signs of exhaustion before a backside reversal develops. The research question was not whether every gapper reverses, but whether there are reliable combinations of conditions that materially improve the odds.
Model logic
GBS does not rely on a single trigger.
Parameter alignment
The model looks for alignment across multiple conditions rather than treating one input as decisive. A reversal zone appears only when the overall context resembles historical cases that behaved in a similar way.
Probabilistic framing
GBS is not trying to predict an exact turning point with certainty. It is trying to mark an area where the reversal thesis becomes more favorable from a historical probability perspective.
Selectivity over frequency
Many active gappers are intentionally filtered out. The framework is designed to pass on incomplete setups rather than force a reversal zone onto every stock that is moving.
Validation philosophy
Built around walkforward evidence, not a single optimized backtest.
Why a standalone framework was necessary
TradingView is a strong charting platform, but it was not sufficient for the kind of portfolio-level walkforward and execution modelling required for this research. Because of that, the validation framework had to be built independently from scratch.
That allowed the model to be studied in a more realistic way than a simple visual backtest or in-sample optimization pass.
What mattered in validation
- Walkforward testing rather than one static historical fit
- Realistic trading frictions such as slippage
- Locate-related costs rather than idealized assumptions
- Ongoing testing against changing market behaviour
Why it changes
Static indicators decay when the market regime changes.
Micro-cap gapper behaviour is not frozen in time. Volume regimes change, intraday range expansion changes, and the kinds of moves that once looked extreme can become routine later on.
That is one of the reasons many older static indicators lose relevance. They were built for a different market regime and never updated to reflect new behaviour.
Monthly re-optimization
GBS is reviewed and re-optimized on a recurring basis to keep the framework aligned with current market conditions rather than treating the model as finished forever.
Why that matters
The objective is not to chase noise. It is to prevent the model from being anchored to stale assumptions in a market that changes much faster than most retail tools account for.
What is public vs private
Enough transparency to judge the framework, without publishing the formula.
Shared publicly
- The market behaviour being studied
- The large-sample historical research basis
- The walkforward-oriented validation philosophy
- The role of re-optimization and realistic execution assumptions
Kept proprietary
- The exact parameter set and weighting logic
- The internal qualification thresholds
- The precise logic that activates and ends a reversal zone
- The optimization details of the standalone research framework
Limitations
What this methodology does not claim.
Not every reversal zone will work
The model identifies favorable context, not certainty. Some setups will fail, and some price moves will invalidate the thesis quickly.
Historical evidence is not future certainty
Even with careful validation, live trading can differ from historical studies because of execution quality, borrow constraints, halts, and changing market conditions.
The trader still controls implementation
Stock selection, risk, size, execution quality, and overall trade management remain critical. The framework supports decisions; it does not replace responsibility.