Topic

Validation

Every lesson in the library on this topic, from beginner to advanced.

Search and filter within this list

Coming soon: 62

Data quality: gaps, spikes and duplicates
Backtest, paper trading and forward testing
Two data providers disagree: what to check
Data quality checklist
Sample size: how many trades is enough?
Backtest sanity checklist
Overfitting in one picture
In-sample vs out-of-sample
Parameter sensitivity: plateaus vs peaks
Walk-forward analysis: testing as time passes
Stress-testing a rule
Benchmarks: choosing the right yardstick
Code as a transparent calculation aid
A macro chart can use numbers nobody had then
Checking an AI-generated indicator formula
Statistical significance without the jargon
Sample size calculator for strategy metrics
Calendar and seasonal rules
Machine learning for markets: what it can and can't do
Confidence intervals for strategy metrics
A small p-value is not a strategy
Recipe: a multi-timeframe filter experiment
Bootstrapping: resampling your own results
Look-ahead bias: the backtest that read tomorrow
Survivorship bias: the symbols that vanished
Data checks before any backtest
Baselines: the model to beat is simple
Benchmarks and null strategies: beat something simple
Fill assumptions: close, next open or intrabar?
Leakage: how ML studies accidentally cheat
Regime analysis: when did the rule work?
Backtest autopsy: five ways a great result was wrong
Overfitting simulator: the best of N random rules
Keep a record of every strategy you rejected
What walk-forward efficiency does and doesn't say
Backtest bias quiz: spot the flaw
Multiple testing: why many discoveries fail
The deflated Sharpe ratio, made intuitive
The look-ahead bug our engine refuses to run
Many strategies, one dataset: selection bias
Business cycles are labelled with hindsight
Data vintages: when the numbers get revised
Reading published pattern statistics critically
Test a pattern definition separately from outcomes
Pivots that move: why patterns change later
Test Fibonacci confluence before believing it
Optimisation and its traps
Monte Carlo simulation: what could have happened
Probability of backtest overfitting (PBO)
Screens have biases too: survivorship and timing
Precision, recall and calibration for market models
A backtesting protocol checklist
Text data and LLMs: timestamps and leakage
Ablation: prove which part of a rule matters
From screen results to a tested hypothesis
Time-series cross-validation done right
Signal decay: how fast does an edge fade?
Purged and nested validation
Trees and boosting on tabular market data
Spot the leak: an ML study quiz
Monitoring a strategy with rules set in advance
Model drift and monitoring

Validation

Beginner

What a backtest can prove, and what it can't