HOW WE BUILT IT
The strategy rules stay private, but everything else is here: how the bot was built, how each strategy did on each coin, and why the final setup looks the way it does.
01 · Rules first
Before any code: no shorts, no leverage, spot only, no overfitting, and backtests on unfavourable costs.
02 · Four temperaments
From years of research we kept four strategies that behave differently: long-term trend, consensus trend, dip buying, short-term momentum.
03 · Every strategy on every coin
Same data, same costs, same period from Oct 2017. Weak results were kept, not hidden or removed.
04 · Combining
All four together, as 12 separate slots: 52.9% a year with a −35% worst fall.
05 · Money management
One shared pool with levelled trimming: 65.2% a year. The live bot runs this exact code; replaying history reproduces the backtest to the cent.
06 · Stress the costs
0.05% slippage checked against a real order book; futures with funding tested and rejected (47.5% vs 57.6%).
07 · The coin split
BTC 50%, ETH 30%, XRP 10%, SOL 10%. The coins with the deepest order books and steadiest prices get the most, so our orders stay small next to the market even as neulquant grows.
08 · Ready for your account READY
Tested end to end on a real exchange account: orders, retries, crash recovery and balance checks all work as designed. The same code that produced the backtest now trades for you.
$10,000 per test, Binance spot, 0.1% fee + 0.05% slippage per trade, filled at the next candle's open. Pick a strategy, then a coin.
| Coin | Final value | Per year | Max drawdown | Buy & hold / yr |
|---|