Reading the classics
with a compiler.
StratCrate exists for one question: when an investing book gives you a rule, does the rule actually do what the book says it does? The only honest way to answer is to write it down precisely and run it.
What this is
A library of backtests. Each issue takes one rule from one investing classic, translates it into an explicit program, runs it over a decade of market data, and publishes the result along with every assumption that shaped it.
The aim is not to find winning strategies. It is to make famous ideas legible — to show you which parts of a well-argued book survive contact with data, and which parts were only persuasive.
Why I built it
Most people who want to invest seriously hit the same wall: the books are convincing, but there is no way to tell which parts are true. Backtests are usually locked behind expensive data and programming skill. That gap is where narratives win and money gets lost.
I started writing these for myself — to check whether the rules I was reading actually worked. This site is that notebook, made public.
What this is not
This is not investment advice, not a fund, and not a signal service. Nothing published here is a recommendation to buy or sell any security. There are no positions, no portfolios, and no performance figures representing real money.
Principles
- Reproducibility over narrative — a result that can't be recomputed isn't a result.
- Publish the losses — known biases and failed rules are written up alongside the ones that worked.
- One rule at a time — a claim small enough to be wrong is a claim worth testing.
Contact
Corrections, data notes, and suggestions for the crate are welcome at [email protected].