Great investing books,
re-run as programs.
Every two weeks, we take one rule from an investing classic, write it as a program, and publish the backtest. You get to see how the idea actually held up — not just how well it was argued.
Next issue
Being written now. The full article lands with the results.
Peter Lynch on the peg ratio rule
One Up on Wall Street · 1989 · 9 min read
Lynch argued that a stock is cheap when its price-to-earnings ratio is below its growth rate. That single comparison became one of the most quoted rules in retail investing. We wrote it as a program and ran it over a decade of US large caps.
Rank every stock by PEG = PE / (growth × 100). Buy the cheapest. Rebalance monthly. Sell on the same rule.
- Period: 2015 – 2024
- Universe: US large caps
- Rebalance: Monthly
- Positions: 20, equal weight
Every issue, the same three steps
The format is fixed so issues stay comparable over time.
One book, one idea
Each issue isolates a single rule an author actually stated — not the whole book, not a vague philosophy. One rule, quoted and pinned down.
Turned into a program
The rule is translated into explicit code: universe, entry condition, position sizing, exit. Every assumption is written down so you can disagree with it.
Tested on real data
The program runs over a decade of market data across hundreds of names. Results are published as they came out, including the ones that did not work.
In the crate
Published so far, and what is being written next.
Lynch argued that a stock is cheap when its price-to-earnings ratio is below its growth rate. That single comparison became one of the most quoted rules in retail investing. We wrote it as a program and ran it over a decade of US large caps.
Graham gave the defensive investor seven hard filters and the enterprising investor a looser version. We test what survives modern markets and what was quietly killed by accounting changes.
Dalio divides the world into growth and inflation, each up or down, and holds a portfolio in every one of the four quadrants. The idea is simple to state and surprisingly hard to rebuild with public data.
One book, one program, every two weeks
Join the list and get each new issue the day it is published.