Always start in dry-run
freqtrade's dry-run mode trades a simulated wallet against live prices. It costs nothing and reveals how the strategy behaves in current conditions rather than historical ones.
# config.json
"dry_run": true,
"dry_run_wallet": 1000,Watch it for at least 2-4 weeks
- Compare dry-run results against the backtest for the same period — a big gap means overfit or lookahead.
- Check trade frequency matches expectations; silence can mean a broken entry condition.
Guardrails before live
- Risk only money you can lose entirely. Start with the minimum viable stake.
- Set max_open_trades conservatively; every open trade is concurrent risk.
- Use exchange API keys with trade-only permission — never withdrawal.
- Verify the stop-loss actually fires on the exchange, not only in the bot.
Know what a strategy can't tell you
No backtest predicts the future. Research evidence — Vetta's included — tells you how a strategy behaved under specific past conditions, never what it will earn. Size positions accordingly.
Educational content, not financial advice. Past performance does not predict future results.