Continuous Improvement Framework

The best traders start with a good-enough strategy and improve it over months and years. They make small, data-driven adjustments and measure whether those adjustments work.

The PDCA Cycle Applied to Trading

PDCA (Plan, Do, Check, Adjust) is the foundation of systematic improvement. One change at a time, measured over sufficient trades.

PDCA in Action

Plan: "Most profits captured by day 7, but average holding is 12 days. Hypothesis: 10-day time exit will improve profit factor."

Do: Trade with 10-day time exit for 30 trades, all other rules unchanged.

Check: Profit factor improved 1.4 to 1.7. Win rate dropped 52% to 48% (some targets missed), but average R increased 0.4R to 0.6R (late-stage losses eliminated).

Adjust: Adopt permanently. Next cycle: test trailing 10 EMA after day 5 instead of hard time exit.

A/B Testing Strategy Modifications

A/B testing runs two versions simultaneously to determine which performs better. In trading, this requires discipline and careful variable isolation.

How to A/B Test

What You Can A/B Test

Variable	Control	Variant
Stop-loss type	Fixed percentage (2%)	ATR-based (1.5x ATR)
Profit target	Fixed R (2R)	Trailing stop (10 EMA)
Entry timing	Limit order at support	Next-day open after reversal candle
Holding period	No time limit	Exit after 10 days if no resolution
Market filter	No filter	Only enter when S&P 500 above 50 SMA

Paper Test First

If the variant involves significantly different risk characteristics (e.g., wider stops), A/B test it in paper trading first. Gather initial data without real-money losses.

What NOT to A/B Test

Journaling for Pattern Recognition

Your is a dataset. The more detailed your journal, the more patterns you can discover.

Monthly Pattern Mining

The Insight Log

Maintain a separate "Insights" section in your journal for observations that are not yet actionable but might become patterns:

Over time, these observations accumulate into hypotheses you can formally test using PDCA.

When to Modify vs. Abandon a Strategy

The 100-Trade Rule

Before abandoning a strategy, ensure you have at least 100 trades of live data. A 55% win rate strategy can produce a losing record over 50 trades roughly 14% of the time. Over 100 trades, the chance drops below 3%. Switching strategies after 10-20 losing trades means you will never give any system enough data to prove itself.

Building Your Knowledge Base

Over years of trading, you accumulate valuable knowledge. Most of it gets lost because it was never written down.

After three years, your knowledge base is worth more than any trading course because it is specific to your trading, your markets, and your psychology.

Try This

20 XP

Start a personal trading knowledge base today. Create four sections: Strategies, Market Observations, Psychology, and Lessons Learned. Write at least one entry in each section based on what you have learned in this course. Commit to adding one entry per week.

Key Takeaways

PDCA (Plan, Do, Check, Adjust) drives systematic improvement. One change at a time, measured over 20-30+ trades
A/B testing isolates variables and produces clear data on what works
Journal-based pattern mining reveals insights that metrics alone cannot surface
Modify when core logic is sound; abandon only after 100+ trades and 3+ failed modifications
A personal knowledge base captures years of accumulated wisdom that would otherwise be lost

Disclaimer

This educational content is for informational purposes only and does not constitute financial advice. Past performance is not indicative of future results. Trading involves risk of loss. You should consult a qualified financial advisor before making investment decisions. Swingfolio is a trade journaling tool, not a financial advisory service.