TrendRider Strategy

### ⭐ If you clone this repo, please give it a star — it's the only payment this strategy asks for.

>

GitHub tells me people actually clone and run this, but only a handful star it. One click, one second, and it's the signal I need to keep pushing updates publicly. Thanks.

Open-source Freqtrade strategy from the TrendRider crypto trading bot.

See it run live (real money, dry-run mode) → trendrider.net/live

The /live page exposes the bot's actual SQLite — including losing exit reasons. This repo lets you run the same strategy yourself.

---

Philosophy

Ride established trends with a wide stoploss. Crypto swings 2–4% per hour; tight stops get noise-killed. Default SL is 6% with ATR-aware widening.

Confidence scoring layers six signals (EMA cross, RSI, ADX, volume, BB position, MACD histogram) into a 0–100 score. Entries require ≥ a tunable threshold.

Cascading early loss cut (V4) closes positions that fail to recover within 4h, instead of waiting for the hard SL.

Backtest highlights (BTC/ETH/SOL · 1h · 30 days)

| Metric | V3 baseline | V4 (this repo) | |---|---|---| | Total profit | — | +69% vs V3 | | Max drawdown | — | −77% vs V3 | | Timeframe | 1h | 1h |

V4 was validated against V5 (breakeven exit) and V6 (relaxed 4h cut) — both rejected via backtest. V4 is the local optimum as of 2026-04-14.

For current live performance see trendrider.net/live.

Install

# 1. Install Freqtrade (requires Python 3.10+)
pip install freqtrade
2. Clone this repo
git clone https://github.com/darkvolg/trendrider-strategy.git
cd trendrider-strategy
3. Init Freqtrade user_data
freqtrade create-userdir --userdir user_data
4. Drop in the strategy
cp TrendRiderStrategy.py user_data/strategies/
5. Use the sample config (dry-run by default)
cp config.example.json user_data/config.json
6. Download data + backtest
freqtrade download-data --exchange bybit --pairs BTC/USDT:USDT ETH/USDT:USDT SOL/USDT:USDT --timeframes 1h --days 30
freqtrade backtesting --strategy TrendRiderStrategy --timerange=20260315-20260414

Going live

The shipped config runs in dry-run mode (no real orders). To trade for real:

• Set "dry_run": false in config.example.json
• Add Bybit API key/secret (futures, isolated margin)
• Start with the smallest possible stake_amount
• Run for at least 7 days dry-run first

You can verify the strategy behaves identically to the live bot by comparing your dry-run trades to trendrider.net/live.

What's not in this repo

The production bot adds a few private layers that aren't open-sourced:

• Fear & Greed Index API integration
• Bybit funding rate / OI signals
• On-chain whale alerts
• Telegram notifications + Cornix routing
• SQLite price-alert layer

All of those are stubbed out here with neutral defaults — the public strategy still trades, just without the external boosts.

Support the project

This strategy is free and will stay free. If it's useful to you:

• ⭐ Star the repo — the single biggest signal that this work matters
• 🐛 Open an issue if backtest numbers don't match yours
• 🔀 Fork it, tune it, share what you find

Every star visibly unblocks listings on awesome-lists and helps other traders discover open, honest strategies.

Optional: V4 Pro Pack

The base strategy is free forever (MIT). A Pro Pack ($19) is available with tuned production hyperopt params, 30-day backtest report (V3 vs V4 comparison, profit factor 1.03 → 1.41, drawdown −6.12% → −1.42%), setup guide, and V1→V4 changelog — for those who want to skip their own hyperopt run. Pay via @CryptoBot

Full post-mortem of the V3 → V4 fix: trendrider.net/blog/freqtrade-bot-14-days-breakeven-v4-fix-2026

License

MIT — use it, fork it, paper-trade it. Don't blame me if you lose money.

Links

• Live stats: trendrider.net/live
• Blog: trendrider.net/blog
• Freqtrade docs: freqtrade.io