Tracking and analytics for a portfolio of trading bots

The solution

1. A spreadsheet as a live dashboard. Google Sheets with logic in Apps Script: exchange rates are pulled from the exchange APIs on a schedule (with a backup source in case one fails), and each bot's status is recalculated automatically. A little mobile page triggers a refresh in a single tap.
2. Reliable history collection via the exchange API. A Python script pulls the data for every bot each day and writes it to an archive. It's resilient to network failures and can tell a genuine bot closure from a temporary connection error. It runs automatically on a dedicated server, with access keys kept isolated.
3. A calculator for setting up new bots. For a chosen coin and the current price, it works out the range boundaries for the width you want — or the other way round.
4. Analytics grounded in data, not gut feel. Regular reports: strategies compared through controlled experiments, closure reasons weighed as a factor in profit — with a clear marker of what's fact and what's still a hypothesis.

The result

The daily manual check of dozens of positions is replaced by a dashboard that updates itself and shows what needs attention right now. Decisions are made from statistics across hundreds of trades, not from memory. Each layer of the system can be improved on its own, with no risk of breaking the rest.

Tech: Google Apps Script, Python, exchange REST API, Google Sheets API, a remote server, isolated access keys.