3Commas RSI Dip + EMA Trend – Backtest

Hypothesis

A tradable, long-only translation of 3Commas’ TradingView Pine v6 strategy, calibrated by its author for XMRUSDT.BYBIT on a 15-minute chart. Preserve the source defaults: initial capital 10,000 USDT; strategy.cash default quantity 1,500; pyramiding 5; commission 0.05% per side; slippage 3 ticks; process orders on close true; calc_on_every_tick false. Base margin is 60 USDT, requested leverage 25x, and the base order is market by default (useLimitBase=false). Four safety orders use first-order margin 60 USDT,……Show moreShow less

A tradable, long-only translation of 3Commas’ TradingView Pine v6 strategy, calibrated by its author for XMRUSDT.BYBIT on a 15-minute chart. Preserve the source defaults: initial capital 10,000 USDT; strategy.cash default quantity 1,500; pyramiding 5; commission 0.05% per side; slippage 3 ticks; process orders on close true; calc_on_every_tick false. Base margin is 60 USDT, requested leverage 25x, and the base order is market by default (useLimitBase=false). Four safety orders use first-order margin 60 USDT, first step 1%, step multiplier 1.2, and size multiplier 1.05. Entry uses 5-minute RSI(12) crossing below 35 and 15-minute EMA(50)>EMA(100). Exit defaults are 1% take-profit activation, 0.1% trailing deviation, 9% close-based stop, and a 259,200-second maximum hold. The enabled entry window is January 31, 2024 00:00 UTC through January 6, 2026 23:59 UTC. The factory instead starts with 100,000 USDT, charges Bybit maker 0.020% and taker 0.055% per fill, and caps leverage at 20x; therefore its execution cannot exactly reproduce the source’s 25x setting. For equity-relative factory sizing, the source’s 60/10,000 margin fraction is 0.6% of current equity per base and first safety order. The source does not directly model 3Commas Risk Reduction or Reinvesting. Webhooks and visual elements do not affect trades. Any claimed performance is unverified.

The primary chart is now 15 minutes, with RSI updated from point-in-time 5-minute bars. The entry window ends on January 6, 2026. A safety level advances only after position quantity confirms a fill. The factory caps leverage at 20x, so sizing cannot reproduce the source's 25x exposure. The reported positive average trade return percentage conflicts with negative average trade PnL and must be reconciled in the backtest ledger; this strategy does not rely on that percentage.

verification_loop: Verification failed (Layer 3 — sandbox backtest): No trades produced Bar type used: XMRUSDT.BYBIT-15-MINUTE-LAST-EXTERNAL, Bars processed: 19969 Diagnostics: should_enter() returned a side 0 times over 19969 evaluated bars -> your ENTRY CONDITION never triggered. Loosen the entry logic / thresholds. Ensure your strategy produces trades with the given data and parameters.

Implementation

Long-only XMRUSDT Bybit translation using a 5-minute RSI cross and 15-minute EMA trend, with four price-triggered safety orders and close-based exits.

Verification Results

Verification failed (Layer 3 — sandbox backtest): No trades produced Bar type used: XMRUSDT.BYBIT-15-MINUTE-LAST-EXTERNAL, Bars processed: 19969 Diagnostics: should_enter() returned a side 0 times over 19969 evaluated bars -> your ENTRY CONDITION never triggered. Loosen the entry logic / thresholds. Ensure your strategy produces trades with the given data and parameters.

Iteration History

Verification failed (Layer 3 — sandbox backtest): No trades produced Bar type used: XMRUSDT.BYBIT-5-MINUTE-LAST-EXTERNAL, Bars processed: 19873 Diagnostics: should_enter() returned a side 0 times over 19873 evaluated bars -> your ENTRY CONDITION never triggered. Loosen the entry logic / thresholds. Ensure your strategy produces trades with the given data and parameters.

Verification failed (Layer 4 — QA review) [class=hypothesis_mismatch]: - [edge_concern] The 70-day sandbox lost 13.4% despite a 98.8% win rate. Its positive average trade return percentage conflicts with negative average trade PnL; reconcile the ledger before relying on that percentage. - [critical] The hypothesis specifies a 15-minute host chart, but config.bar_type is 5 minutes. Checking timestamps every 15 minutes does not make the configured primary timeframe match the hypothesis. (line 67) — fix: Use a……Show moreShow less

Verification failed (Layer 4 — QA review) [class=hypothesis_mismatch]: - [edge_concern] The 70-day sandbox lost 13.4% despite a 98.8% win rate. Its positive average trade return percentage conflicts with negative average trade PnL; reconcile the ledger before relying on that percentage. - [critical] The hypothesis specifies a 15-minute host chart, but config.bar_type is 5 minutes. Checking timestamps every 15 minutes does not make the configured primary timeframe match the hypothesis. (line 67) — fix: Use a 15-minute primary bar type with a synchronized 5-minute RSI input. - [critical] window_extended_end_utc changes the stated entry-window end from January 6 to September 30, 2026. The sandbox therefore tests entries outside the hypothesis window. (line 35) — fix: Use the stated window_end_utc, or preregister a revised hypothesis with the later date. - [critical] safety_filled advances after calling _submit_entry even when that helper declines to submit an order. A rejected safety order is then treated as filled, skipping a ladder level. (line 163) — fix: Advance the count only after a confirmed fill.

Outcome Summary

Check that the combined entry conditions trigger on the available data before further evaluation.

Verification failed because the entry condition never triggered. The run was abandoned after three iterations; later review stages were not reached.

It sought long XMRUSDT.BYBIT entries when 5-minute RSI crossed below 35 while the 15-minute EMA(50) exceeded EMA(100), using safety orders and trailing exits.

The Layer 3 sandbox backtest processed 19,969 bars and produced no trades. No return or Sharpe result was provided.

Analysis

RSI Dip + EMA Trend Long DCA - Strategy by 3Commas