One of the most important lessons in trading system development is also one of the least understood.
The Default Reaction
When a trading strategy underperforms, the instinct is almost universal:
“Maybe I should use a 20-day MA instead of 30…”
“Maybe I need a better entry trigger…”
This thinking is logical—and necessary. It addresses a critical question: “Is the signal itself valid and robust?” But here’s the problem: It only addresses part of the overall process.
The Reality: Same Signal, Completely Different Outcomes
At the Owl Bundle User Group (OBUG), we study and apply Dr. Ken Long’s systematic trading methodology and use EdgeRater’s Multi-Factor Analysis to test strategies across large universes of symbols. What we consistently observe is a repeatable pattern: Signal effectiveness can vary significantly depending on the instrument.
For example:
A mean reversion signal shows strong, consistent performance in:
ASML
MSFT
The exact same signal shows weak or unstable performance in:
CPB
T
Nothing changed in the rules. Only the instrument changed.
The Core Insight Most Traders Miss
Trading rules identify opportunity — Symbol selection plays a major role in determining whether that opportunity has a consistent edge.
This is the difference between:
A strategy that is statistically valid
A strategy that is profitably deployable
What Most Traders Are Actually Missing
Most strategy development focuses on:
Parameter optimization
Walkforward testing
Out-of-sample validation
These steps are essential—they validate that a signal is real. But they often under-emphasize the next step: Determining where that already-valid signal actually works best
This is where performance differences emerge—and where much of the edge is realized.
Two Complementary Approaches
1. Signal Validation
Define rules
Optimize parameters
Validate across time
Purpose: Confirm the signal has a real edge
2. Deployment Optimization
Apply the validated signal across many symbols
Identify where it performs consistently
Deploy capital selectively
Purpose: Capture the edge where it expresses most clearly
Why This Happens (Structural, Not Random)
This variation is often linked to structural characteristics of the instrument.
Different symbols respond differently due to:
1. Institutional Flow
Large-cap growth → sharp, reflexive moves
Defensive stocks → slower, muted reactions
2. Volatility Elasticity
High-beta names → strong mean reversion response
Low-volatility names → weaker response
3. Liquidity Depth
Deep markets → cleaner, more reliable behavior
Thin markets → noise and instability
The Hidden Truth About Edge
Most traders assume: “Edge comes from the strategy.”
A more accurate framing is: Edge comes from the interaction between the strategy, the instrument, and how it is deployed.
When testing across large universes, results are averaged and strong performers are diluted by weaker ones. In many cases, a subset of symbols may contribute disproportionately to overall performance.
A More Effective Workflow
Instead of asking: “How do I improve the strategy?”
Ask: “Where does this strategy perform most robustly?”
Step 1: Validate the Signal
Confirm robustness across time
Avoid overfitting
Step 2: Test Broadly
Apply the same rules across many symbols
Let the data reveal differences
Step 3: Identify Consistency
Focus on symbols with:
Stable expectancy
Controlled drawdowns
Repeatable behavior
Step 4: Build a Curated Universe
This becomes your tradable list
Strategy and instrument are now aligned
A Simple Analogy
Think of your strategy as a fishing technique:
Signal-focused → refining the rod and bait
Deployment-focused → finding the lake where the fish actually are
Same method, completely different outcome.
Net-Net
In some cases, your strategy may not be broken — your universe may be misaligned.
Parameter optimization validates the signal
Symbol selection determines where it works
Deployment determines results
Inside OBUG
At the Owl Bundle User Group (OBUG), this is the type of work we are actively studying and testing.
Using EdgeRater’s backtest and Multi-Factor Analysis, we explore this process by:
Identifying robust trading signals
Testing them across broad universes of symbols
Isolating where those signals show consistent, repeatable edge
Building curated symbol lists aligned with Dr Ken Long’s strategies, including:
Critical States Systems
Swing Systems
At OBUG, we approach this as an ongoing research and development process. If that aligns with your interests — Join us inside OBUG