Who and why

About

I test trading ideas on Indian stocks, on my own, alongside a full-time job. The setup has four parts: something that records the order book for a fixed list of stocks every trading day, ten years of daily history rebuilt so that it still contains companies that have since shut down, a replay engine, and the scoring code that tells me whether an idea predicts anything.

All of it runs by itself on one cloud machine. This website has no connection to that machine at all.

What is not on offer

I would rather say this plainly than have you guess. Each one of these would need something I do not have.

• No track record. Every number here comes from a simulation, either a backtest or paper trading with pretend money. I have never used real money. The one strategy that passed started its forward test in August 2026 and the first result is due around February 2027. Even then, six months of data cannot prove anything much.
• No advice, signals or tips. I am not a registered investment adviser and nothing here is a recommendation to buy or sell anything. If you trade because of something you read here, that is your decision alone.
• No data on offer, and here is the honest version. Part of why I keep this archive is that one day I might be able to sell access to something built on it. I would rather say that than have you work it out later. But I have not sorted out whether I am allowed to pass on exchange data, and I may never be able to. So there is no price, no download and no signup list, and there will not be one until that question has a real answer. If I ever do sell anything, it is far more likely to be my own work, the corrections and the curation, than the exchange's data.
• No settings you could copy. I publish the method, the stories and every idea that failed, with the numbers, because those are finished. How the one live strategy is built stays private while I am still testing it.
• No clients, no fund, no endorsements. You will not find testimonials here because I have no customers, and no logos because I have no affiliations.

Why publish at all

Because the failures are the interesting part and almost nobody puts them up. What you normally see is the small number of things that worked, written up afterwards by people with a reason to make them sound good. You never get to see how many attempts it took.

The Ledger is my attempt at the opposite: the full count, including every idea that looked good early and then fell apart on data it had not seen. If it helps one other person building something similar, that is enough for me.

Contact

Corrections are very welcome. If something here is wrong I would rather know, so please tell me. The quickest way is LinkedIn.

The code behind this site is not public. Part of it holds the settings for the strategy I am still testing, which is the one thing this site does not publish. My other work is at github.com/arkagimt.