The Edge That Wasn't
Algorithmic Trading, 81 Experiments, 5,556 Live Trades, and the Answer the Data Gave
Most books with "algorithmic trading" on the cover sell a system that prints money. This one hands you the losses. The central artifact of the book is a table of broker-verified results — its own — and the frozen snapshot of every recorded trade ships with it, so any per-strategy row reproduces from the raw file. Foot the table yourself.
It is the record of fx-core, a containerized retail FX research project. No single strategy ever ran on more than about a hundred dollars, and most on far less; roughly two hundred dollars was funded across the thirteen broker sub-accounts over the project's life, peaking near a hundred and ninety-nine. The methodology is the subject, not the size of the account.
One question runs through it: is there a repeatable, statistically real edge available to a retail trader in spot FX, and if so, is it large enough — after costs, after tail risk, after the capital required to hold open positions through their drawdowns — to beat a stock-index fund? Roughly seventy real experiments answer it, and one strategy survives.
What the apparatus taught outlasts the trading. Pre-registration, a sealed out-of-sample set touched exactly once, a working taxonomy of lookahead bias, the coin-flip control that separates skill from luck, and the cost floors that sink most edges: a full research-methods spine, demonstrated on live results rather than lectured. It is meant to be useful to a data scientist or engineer who will never place a trade.
Inside
- A live and paper strategy catalogue with a frozen snapshot of every recorded trade — the receipt behind every number in the book.
- The 9 rules of the apparatus — causality, sealed out-of-sample, Monte-Carlo, fill modeling, account modeling — and the specific failure each rule was written to prevent.
- The lookahead error that produced 55,000 pips of bad signals across sixteen strategies, and the root-cause analysis that followed.
- The negative-result library: thousands of backtests across twelve currency pairs and five and a half years of data, mapped by the structural reason each one failed.
- The momentum book that passed Monte-Carlo at a 100% rate, then was shown by a closeout-aware simulation to liquidate six of six accounts under finite margin.
- Zone Recovery: the martingale whose tail cannot be capped without killing the edge.
- The three walls — direction is unpredictable, volatility is forecastable, spread is the toll — and the narrow escape they leave open.
- The regime-gated mean-reversion that is the project's surviving high-water mark: t = 2.25, p = 0.025.
- A 161-entry indicator encyclopedia with a visual reference panel computed on real EUR/USD M5 data.