Start here
Most trading ideas sold to you are empty waters. Alpha Survey publishes one research case each week: a trading idea close to the way it is sold, tested with the pass or fail criteria written down before looking and real trading costs, and what we found. The rule we test is our own falsifiable version of the pitch, not the product itself: when a scanner or indicator was not part of the test, the entry says so. We are quant traders: rules, not opinions or emotions. This is a study, for information only, not investment advice.
How to read a log entry
- How it is usually sold. The chart they show you and the promise, with the indicators or scanners it comes with.
- Where we cast. The exact rule we tested: universe, signal, holding period, costs.
- Why it looked promising. The mechanism that could have made it work, and who would be on the losing side.
- What we found. The gate table: every criterion fixed in advance, and which ones failed. A p-value there is the share of outcomes under our null model at least as extreme as what we saw; it is not the probability that the idea is luck.
- What we learned. The lessons, each with a concrete picture.
- What would bring us back. What does not reopen the idea, and what could, one day.
Every term is explained in the glossary; how we judge is in the method.
The first three entries
▶Testing the shorting premium: what a rebalancing bug taught us
Empty because: a short leg that loses once positions are held for the month, and a fee ranking that does worse than random
▶The post-earnings bounce did not survive our cost assumptions
Empty because: net return of zero in one variant and of the opposite sign in the other
▶Short-term momentum in large caps: no robust edge in our test
Empty because: negative net return, and no convincing evidence that the ranking added value before costs
Follow the survey
New entry every Tuesday. YouTube (or the playlist, in order), the newsletter, or the RSS feed. Were you sold an idea? Send us the pitch and we may sound those waters next.