Free Lesson
Machine learning for trading: choosing the next experiment
30 min
Oct 28, 2026 11:00 AM
By continuing, you agree to Maven's Terms and Privacy Policy.
What you'll learn
Lay out a strategy as a sequence of choices
List each choice from market and data timing to labels, model, portfolio and costs behind a result.
Separate a result from what it means
Write down every explanation a score is consistent with before you decide what to change next.
Build the next experiment so its answer counts
Change one choice, hold the rest against a baseline fixed in advance, and count the variants tried.
Why this topic matters
A machine learning trading strategy is a chain of choices: the market, the data and its timing, the labels, the features, the validation protocol, the model, the portfolio and the costs. A backtest reflects all of them at once, so a result rarely says on its own what to change. This lesson works through that chain and shows how to go from one result to the next experiment worth running.
