Statistical power
The chance that a test or check finds an effect or problem that is really there. A small check has low power: it will usually find nothing, whether or not something exists.
In plain English
Power depends on how big the check is, how big the effect is and how much noise there is. Twenty yards of road, a dozen patients or two months of trades give low power, so 'we found nothing' from them means very little. Before trusting a clean result, ask what the check could have found if the problem were real.
Uncle Gary explains
Uncle Gary's road committee inspected 20 yards of a 2,000-yard road with ten potholes on it. A check that small finds nothing about 90% of the time. The committee reported 'no evidence of potholes'.
Why it matters for your money
A backtest with a handful of trades will rarely show a real loss or a real edge. Low power is why small tests of trading ideas so often 'find nothing' or 'find something' by luck.
Where we tested it
- Lesson 15: Type I and Type II Errors: Why Small Tests Miss Things
- The Committee Found No Evidence of Potholes. It Checked 20 Yards.
Related terms
All glossary terms. Educational content, not financial advice.