Every strategy in the team's paper program was replayed on HTX perpetual price history: about 3 years of 15-minute candles and 5.5 years of 4-hour candles.
- 15-minute strategies: almost all lost about 10% and then stalled, because once the $100 virtual account fell ~10% it dropped below the $5 minimum order size. Win rates of 4–42% did not cover costs.
- 4-hour strategies: mixed. Some (Donchian, EMA, Keltner) were modestly positive, others negative. Most would still need more trades to be convincing.
- Adding a daily trend filter (only allow longs when a daily Donchian breakout is active) improved several 4-hour variants but did not rescue the 15-minute ones.
- Takeaway: at a 0.6% round-trip cost, 15-minute trading is a structural loser for this setup. Any 4-hour edge is thin and needs longer out-of-sample testing.