Sixty-seven lessons have given you components. None of them has given you an edge, because an edge is not something you can be given — a strategy that works for someone else has not been tested by you, on your broker, at your execution quality, with your temperament.
This is the last lesson because it is the only one with an output. Everything before it was preparation for these twelve weeks.
Nobody finishes a course as a professional. This is the work that might get you there.
The twelve-week project
Twelve weeks is not arbitrary. It is roughly the minimum in which 100 trades can be generated honestly, and 100 trades is the minimum from which anything can be concluded.
---
Weeks 1–2 · Choose and specify
Pick one instrument, one timeframe pair, one setup. The urge to hedge by choosing three is the urge that produces three untested systems.
Write the seven components from the planning lesson. Then apply the stranger test: could someone else, reading only your document, mark up ten charts the same way you would?
Deliverable: a written plan where every setup condition is objectively checkable. Not "price shows strength" — *"the 4H closes above the prior swing high with a body exceeding 0.7× ATR(14)."*
Most people take longer than expected here, and that is the lesson: the vagueness you find is vagueness that was always in your trading.
---
Weeks 3–4 · Backtest, honestly
Bar replay only. Two years minimum, including a trending period, a range and a volatility spike.
Log every trade: date, setup, entry, stop, target, R outcome, and whether every rule was met.
Deliverable: 100+ backtested trades and six numbers — expectancy, win rate, average win, average loss, worst losing streak, worst drawdown.
If expectancy is negative, stop here. Go back to week 1 with a different setup. This is a success: you have discovered in four weeks, at no cost, what most traders discover in two years and a blown account.
---
Week 5 · Stress the numbers
The statistics lesson, applied to your own data.
Confidence interval.SE = std dev ÷ √n. What is the 95% range around your expectancy? Does it include zero?
Monte Carlo. Shuffle your trades 1,000 times. What is the 95th-percentile drawdown?
Costs. Subtract your actual spread and commission per trade. Recalculate.
Size for the 95th percentile drawdown, not the one you happened to observe.
Deliverable: a risk percentage derived from arithmetic. For most systems this lands between 0.5% and 1%.
---
Weeks 6–7 · Forward test
30 trades on demo, executed live in real time. This catches what a backtest cannot: hesitation, missed entries, rules that turn out to be ambiguous under pressure.
Deliverable: 30 demo trades with an adherence field on every one.
If adherence is below 90%, the problem is the plan, not you. A rule you cannot follow in real time is a rule that is not written clearly enough. Fix the document, not your discipline.
---
Weeks 8–12 · Live, at half size
50 trades at half your calculated risk. Full journal on every one.
Deliverable: 50 live trades, and the split that matters — expectancy on plan-following trades versus improvised ones.
---
The review that ends the project
Four questions. Answer them from data, not impression:
Question
How you answer it
Does the edge survive live execution?
Live expectancy versus backtest expectancy
Was the gap explained by costs?
Execution audit from the broker lesson
Can I actually follow this?
Adherence percentage across 50 live trades
Am I sized for the bad case?
Worst live drawdown versus Monte Carlo forecast
Four yeses: raise to full size. Continue to 300 trades before considering anything else.
Any no: the specific answer tells you what to fix, and none of them means "find a new strategy." An adherence problem is a document problem. A cost problem is a broker problem. Only a negative expectancy at 90%+ adherence is a strategy problem.
What this actually produces
Not a guaranteed income. Twelve weeks of disciplined work can still end with "this edge does not survive my costs."
That is a real result and it is worth having. It cost you twelve weeks and a small live sample instead of two years and an account.
What it does produce, in every case:
A written system that exists outside your head
Your own numbers, not someone else's claims
Evidence about whether you can execute under pressure
A measured risk percentage instead of a chosen one
The ability to tell variance from a real problem
That last one is the whole thing. It is what makes the difference between a trader who abandons a working system in month three and one who is still running it in year three.
The honest close
This curriculum was written with care, and its claims are checkable. It cannot make you a professional trader.
Only trading can do that, and only if you keep the records that let you learn from it. Most people who fail do so not because they lacked information, but because they never generated enough of their own data to tell what was working.
You now have everything you need to start generating it.
Where anything in these sixty-eight lessons conflicts with what your own journal shows over a real sample — your journal is right.
