A trading bot, in public
Every trade we make is on this site, including the ones that lose.
Swarm is a trend-following bot for crypto futures. Every four hours it asks 16 simple trend questions about ten large coins, holds the four with the clearest answer, and puts a stop-loss on each. This site is its public record: the trades, the drawdowns, and the research behind every rule, with the dull parts left in.
Where we are, honestly
What is real
The bot, the exchanges and the orders. The Binance account runs on Binance's own Demo Trading platform, so fills, stop orders and errors are the exchange's, not ours. Every number below comes from the bot's records, rebuilt here every 10 minutes.
What is not yet
The money. Both accounts are demo or paper, and they have only run since 27 September 2026. We won't ask anyone to copy us on the strength of a few weeks, or of a backtest. When a real account opens, it will be labelled LIVE and linked to its public page on the exchange.
What you can check
The rules are written down on the strategy page. Each change we made is explained in research, including the ones that made the backtest look worse, and the journal publishes a report every week, good or bad. Short answers to the usual questions are on the questions page.
The accounts
Binance account DEMO
No public exchange page for this account yetHyperliquid account PAPER
No public exchange page for this account yetLatest research
Each rule the bot follows was tested on six years of data first. These are the write-ups, including the tests that failed.
- We could cut 39% of our orders. We decided not to.Resizing less often cut orders from 687 to 420 a year with the same long-run return, but recent years were weaker. Wider still looked great, but only because of 2021.
- Our cheap orders were costing us money on every exitWaiting 45 seconds for a cheaper fee made sense for entries. On exits it never filled, the price moved away, and it cost 9–22 bps to save 2.
- If altcoins just follow Bitcoin, why hold four of them?They mostly do: 77% of 4-hour candles move with BTC. But holding fewer coins cut returns to 19–37% a year, and one stop-loss would cost 2–3 times more.
- Does the profit lock need to react in minutes? We replayed it on 5-minute candles.In six years, positions reached +100% 14 times. Checking every 5 minutes instead of every 4 hours changed nothing: same 3 exits, same returns to 0.01%.