Walk Forward Optimization Is Not Automatically Out of Sample

Everyone sells walk-forward validation as proof a strategy is out-of-sample. It is not. I ran my own walk-forward pipeline on pure random-walk noise and it still produced a healthy positive stitched Sharpe, because out-of-sample is a property of provenance, not partitioning. Here is the random-walk control that tells you whether your pipeline, or your edge, produced the number.

By Rayan Azhari · 9 min read · Essays

Latest Essays

Essays

Better Than Assad Is the Minimum, Not the Dream

One argument recurs constantly in Syrian conversation today, and it is true in much of what it claims: "there are problems, but things are better than they were under Assad". The problem begins the moment it stops describing reality and starts answering every question. This piece separates the question of how far we have come from the question of where we are going, distinguishes rights that need capacity from rights that need only a decision, and sets out a floor of demands that needs no budget and no ten years.

2026-08-17

The Deflated Sharpe Ratio: Why Your Grid-Search Winner Is Probably Noise

The expected best Sharpe of a parameter sweep climbs as the grid grows, even when every strategy in it is worthless. Here is the deflated Sharpe ratio explained as a practitioner sees it, with a small N to noise-ceiling lookup table you can apply to your own sweep tonight, and the one rule that killed my proudest grid-search winner: N is the pool, not the podium.

2026-08-14 · 9 min

Article 10: Time to Retire ECG-19? Updating Office Benchmarks With Evidence From 6,000 Greater London Buildings

A series mining the PhD thesis London and UK Office Buildings (Azhari, 2025). Key takeaway. The UK mainstream office benchmarks ECG-19 and CIBSE TM46 rest on a 1992 study, and the Greater London data shows actual offices nowhere near those values, so a refresh is overdue.

2026-08-11 · 9 min

One Arrow, Drawn Once: The Architecture That Keeps Research Out of Production

The most dangerous code in a trading repo is the throwaway research notebook. One structural rule, dependencies that flow in a single direction, keeps it out of production and turns invisible coupling into a number you can grep.

2026-08-07 · 9 min

Article 09: NABERS for Britain? Lessons from Australia on Operational Rating

A series mining the PhD thesis “London and UK Office Buildings: Investigating Energy Use and Landlord-Tenant Influences” (Azhari, 2025). Key takeaway. Australia NABERS rates buildings on what they actually use and makes ratings mandatory for office disclosure. The UK is part-way there with NABERS-UK and the DEC, but missing the policy backbone.

2026-08-04 · 9 min

Buy the Engine, Build the Edge: NautilusTrader, backtrader, or Roll Your Own

Choosing a quant trading framework is not a feature comparison between NautilusTrader, backtrader and a DIY loop. It comes down to one property, and a rubric you can apply to any engine: does your backtest run the same code as live?

2026-07-31 · 9 min

Essays in your inbox