trend-combomedium risk
Multi-Indicator Confluence (Crypto)
Does a multi-indicator confluence strategy work on crypto? Backtested on real Hyperliquid perps from 2024-08-15 to 2026-07-05, the default settings earned a Sharpe of 1.65 — +153.6% total return with a 25.9% worst drawdown across 3,070 trades, net of fees, slippage, and funding.
By Keel Research · Data as of 2026-07-05 · how this backtest was produced
Sharpe
1.65
default settings
Total return
+153.6%
Worst drawdown
−25.9%
Trades
3,070
2024-08-15 → 2026-07-05 · net of fees, slippage, and funding · price-only Sharpe 1.78 · these are the default settings, not a hand-picked best case.
At $1,000, this strategy’s worst historical dip was about $259. Free account, your own Hyperliquid keys — Keel only trades.
How it works
How it works
This is a three-filter trend strategy on the top-15 Hyperliquid perpetuals, four- hour bars. It goes long a coin only when all three agree: its EMA-50 is above its EMA-200 (an uptrend), its MACD histogram is positive (momentum), and its RSI sits in a healthy 45-70 zone (not overbought, not collapsing). If any filter disagrees, it holds cash. Positions are equal-weight, 1x gross, long-or-flat.
When it works best
Clean, established uptrends where trend, momentum, and strength all line up — the three filters keep it in the strongest names and out of anything ambiguous. Over 2024-08-15 to 2026-07-05 the default three-filter settings returned +153.6% with a Sharpe of 1.65 and a maximum drawdown of −25.9%, net of fees, slippage, and funding.
When it struggles
Choppy or reversing markets, where the filters flip in and out and entries get whipsawed. There is also an honest finding worth flagging: stacking more indicators is not free. Dropping the RSI filter — running just trend and momentum — actually scored a little higher in the test window, because the three measures overlap, so the third one screens out some good trades along with the noise. The default keeps all three because the full three-filter setup is the one people ask about, but the leaner two-filter version was the stronger performer.
How it’s built
The exact strategy behind this backtest — no black box. Switch to code to see or copy the full definition.
Configuration
Globalsconfig
target_timeframe4h
UniverseassetsTop by Volume · HL Perps · 15 assets
Executiontrading
rebalanceEvery Bar
Factories
ema_gatefactory
px_vs_50factory
Pipeline5 steps
PriceDataLoaderData Loader
timeframe15min
TargetTimeframeResamplerData Transform
Parallel2 branches
entries
Parallel2 branches
const
ConstantForecastForecast Mapper
value1
setup
Parallel3 branches
regime_up
ema_gatecall
AboveThresholdFilterSignal Transform
threshold0
momentum_on
MACDIndicator
fast_period12slow_period26signal_period9
AboveThresholdFilterSignal Transform
threshold0
not_exhausted
RSIIndicator
period14
RangeSelectorSignal Transform
low_threshold45high_threshold70
MaskAndSignal Composer
ApplyMaskSignal Composer
score_signalconstfilter_signalsetup
conf_entrystore
exits
Parallel2 branches
const
ConstantForecastForecast Mapper
value1
bail
Parallel2 branches
macd_flip
MACDIndicator
fast_period12slow_period26signal_period9
BelowThresholdFilterSignal Transform
threshold0
below_50
px_vs_50call
BelowThresholdFilterSignal Transform
threshold0
MaskOrSignal Composer
ApplyMaskSignal Composer
score_signalconstfilter_signalbail
conf_exitstore
PositionStateMachinePosition Manager
entry_slotconf_entryexit_slotconf_exit
EqualWeightSizerPosition Sizer
target_leverage1max_weight0.2
Globals(target_timeframe="4h")Universe(mode="top_volume", top_n=15, market="perp", resolved=[], resolved_at="")Execution(rebalance="every_bar")ema_gate = Pipeline([ { "fast": [EWMA(window=50)], "slow": [EWMA(window=200)], }, Crossover(fast_key="fast", slow_key="slow"),], name="ema_gate")px_vs_50 = Pipeline([ { "px": [EWMA(window=2)], "e50": [EWMA(window=50)], }, Crossover(fast_key="px", slow_key="e50"),], name="px_vs_50")Pipeline([ PriceDataLoader(timeframe="15min"), TargetTimeframeResampler(), { "entries": [ Pipeline([ { "const": [ConstantForecast(value=1.0)], "setup": [ Pipeline([ { "regime_up": [ema_gate, AboveThresholdFilter(threshold=0.0)], "momentum_on": [MACD(fast_period=12, slow_period=26, signal_period=9), AboveThresholdFilter(threshold=0.0)], "not_exhausted": [RSI(period=14), RangeSelector(low_threshold=45, high_threshold=70)], }, MaskAnd(), ], name="setup"), ], }, ApplyMask(score_signal="const", filter_signal="setup"), Store("conf_entry"), ], name="entries"), ], "exits": [ Pipeline([ { "const": [ConstantForecast(value=1.0)], "bail": [ Pipeline([ { "macd_flip": [MACD(fast_period=12, slow_period=26, signal_period=9), BelowThresholdFilter(threshold=0.0)], "below_50": [px_vs_50, BelowThresholdFilter(threshold=0.0)], }, MaskOr(), ], name="bail"), ], }, ApplyMask(score_signal="const", filter_signal="bail"), Store("conf_exit"), ], name="exits"), ], }, PositionStateMachine(entry_slot="conf_entry", exit_slot="conf_exit"), EqualWeightSizer(target_leverage=1.0, max_weight=0.2),], name="confluence_crypto")Explore the settings
precomputed · updates instantlyAdjust a setting to see the exact backtested result — including the ones that lost money.
RSI confirmation zone
tf
Universe size (top by volume)
Sharpe
1.65
Return
153.6%
Worst DD
−25.8%
Trades
3,070
At $1,000, these settings' worst historical dip was about $258. You land in the editor with this exact setup. Free account, your own keys.
Compare all 4 settings
| Settings | Sharpe | Return | Worst DD | Trades |
|---|---|---|---|---|
| RSI 45-70 · tf 4h · Universe 15default | 1.65 | 153.6% | −25.8% | 3,070 |
| RSI off · tf 4h · Universe 15 | 1.80 | 184.7% | −27.1% | 2,984 |
| RSI 45-70 · tf 4h · Universe 25 | 1.40 | 166.9% | −31.5% | 5,484 |
| RSI 45-70 · tf 1d · Universe 15 | -0.35 | -17.7% | −29.5% | 268 |
The data
Monthly returns
| Month | Return |
|---|---|
| 2024-08 | 0.0% |
| 2024-09 | +9.7% |
| 2024-10 | +9.3% |
| 2024-11 | +27.3% |
| 2024-12 | +0.4% |
| 2025-01 | -13.6% |
| 2025-02 | -1.8% |
| 2025-03 | 0.0% |
| 2025-04 | +2.9% |
| 2025-05 | +2.5% |
| 2025-06 | +2.7% |
| 2025-07 | +12.4% |
| 2025-08 | -8.1% |
| 2025-09 | +25.8% |
| 2025-10 | -2.6% |
| 2025-11 | +0.2% |
| 2025-12 | +4.7% |
| 2026-01 | +8.3% |
| 2026-02 | +3.9% |
| 2026-03 | +3.8% |
| 2026-04 | -4.0% |
| 2026-05 | +25.1% |
| 2026-06 | -6.5% |
| 2026-07 | +3.2% |
Which assets it traded
| Avg allocation | Days held | % of time held | Asset |
|---|---|---|---|
| 0.1928 | 563 | 13.62 | HYPE |
| 0.1894 | 552 | 13.35 | BTC |
| 0.1932 | 476 | 11.51 | VVV |
| 0.1854 | 471 | 11.39 | LINK |
| 0.1887 | 460 | 11.13 | SOL |
| 0.1867 | 455 | 11.01 | DOGE |
| 0.188 | 425 | 10.28 | ETH |
| 0.1858 | 424 | 10.26 | XRP |
| 0.1877 | 403 | 9.75 | SUI |
| 0.1851 | 344 | 8.32 | kPEPE |
| 0.1966 | 214 | 5.18 | ZEC |
| 0.1847 | 132 | 3.19 | PUMP |
| 0.1973 | 110 | 2.66 | XPL |
| 0.1987 | 90 | 2.18 | LIT |
| 0 | 0 | 0 | GRAM |
What this is: a historical backtest on real Hyperliquid market data, net of fees, slippage, and funding. Its worst historical drawdown was 25.9% — expect drawdowns of that order or worse. Past performance does not predict future results, and this is not investment advice. Size your account so a full drawdown is survivable.