About
This is an experiment I am funding to find out how good the NFAI engine really is. It runs three books in the open, tracks them against the market every day, and shows the results without spin. The goal is to learn and to make the engine better, not to sell anyone anything.
The NFAI engine picks the names, sizes them, and decides what to do with each one. I funded the books and let it run. Every day the site shows where things stand so the engine is held to a real scoreboard instead of a story.
Right now there are three books. NFAI is the main one. Marauri is a tiny mirror that runs the same logic at a fraction of the size. SPY is the benchmark, the line the engine is trying to beat over time. All values are in USD.
The engine scores a universe of names, builds a book it can defend, and then manages risk on autopilot. In plain terms it:
The exact thresholds and the scoring weights are the part that makes it worth running, so those stay under the hood. What you see here is the output and the scoreboard.
Prices are daily closes. After the US market closes, the engine pulls the latest close for every holding and for SPY, recomputes everything, and updates the site. Because the data is end of day, the numbers move once a day, not by the second. Every page shows when it last updated.
The whole point falls apart if the numbers are dressed up, so they are not. When a position is down, it shows red. When the book trails SPY, the chart shows it trailing. Good days and bad days get the same treatment.
These are simulated books run for research and education. Nothing here is a recommendation to buy or sell anything, and no one should copy these trades. It is one experiment, testing one engine. Past results do not predict future ones.
