Best momentum lookback and selection settings for crypto

You can adjust Drift before rebalancing, Momentum lookback (days), Momentum selection cutoff. The config explorer on this page lets you compare every setting we tested — including the ones that underperformed — with the exact backtested result for each.

momentumhigh risk

Momentum Strategy on Hyperliquid

Name: Momentum Strategy on Hyperliquid — backtest results (Hyperliquid, 2024-08-15→2026-07-05)
Creator: Keel

Does a momentum strategy work on Hyperliquid perps? Backtested on real Hyperliquid perps from 2024-08-15 to 2026-07-05, the default settings earned a Sharpe of 0.98 — +90.5% total return with a 39.0% worst drawdown across 949 trades, net of fees, slippage, and funding.

By Keel Research · Data as of 2026-07-05 · how this backtest was produced

Sharpe

0.98

default settings

Total return

+90.5%

Worst drawdown

−39.0%

Trades

949

2024-08-15 → 2026-07-05 · net of fees, slippage, and funding · price-only Sharpe 1.07 · these are the default settings, not a hand-picked best case.

Run this live →Fork & customize

At $1,000, this strategy’s worst historical dip was about $390. Free account, your own Hyperliquid keys — Keel only trades.

How it works

This is cross-sectional momentum: perps that have outperformed their peers over the past two weeks tend to keep leading for the next several days. In crypto the effect is amplified by attention and flow — rallies pull in volume, and volume pulls in more flow. Each day the strategy ranks the top-50 Hyperliquid perps by two-week momentum and buys the leading quarter, but only names trading on genuinely elevated volume, which filters out stale drifts. Holdings are long-only and sized inverse to volatility so each contributes roughly equal risk, and a market-wide trend filter moves the whole book to cash when the broad market rolls over.

When it works best

Broad risk-on regimes with real dispersion between coins — the book compounds hardest when a handful of names break away and lead for days or weeks, the alt-rotation pattern crypto tends to produce in bull markets. The trend filter keeps it invested while the market is rising and out of the way when it isn't.

When it struggles

Regime flips are the pain point: the trend filter exits a bar after the market turns, so the sharp momentum-crash bars right at the top are where the drawdowns come from — the maximum drawdown over the window was −39.0%, with long underwater stretches. Being long-only, it also pays funding while crowded on the popular side, and choppy, leaderless markets produce whipsaw entries and fees without follow-through.

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_timeframe1dbar_offset12h

UniverseassetsTop by Volume · HL Perps · 50 assets

Executiontrading

rebalanceBufferedbuffer_threshold0.2buffer_moderelative+1

Factories

selectionfactory

Pipeline5 steps

PriceDataLoaderData Loader

timeframe15min

TargetTimeframeResamplerData Transform

Parallel3 branches

selection

selectioncall

market_regime

EWMAIndicator

window2

MarketTrendRegimeFilterRegime Detector

fast_period5slow_period17

vol

ReturnVolatilityIndicator

window36d

return_volstore

RegimeGateForecast Composer

signal_keyselectionindex_keymarket_regimeconditionindex > 0

VolWeightSizerPosition Sizer

vol_slotreturn_voltarget_leverage1max_weight0.15

Globals(target_timeframe="1d", bar_offset="12h")

Universe(mode="top_volume", top_n=50, market="perp", resolved=[], resolved_at="")

Execution(
    rebalance="buffered",
    buffer_threshold=0.2,
    buffer_mode="relative",
    rebalance_method="to_edge",
)

selection = Pipeline([
    {
        "momentum_filter": [
            ROC(period=14),
            QuantileRankTransform(),
            QuantileRangeSelector(low_quantile=75, high_quantile=100),
        ],
        "rel_vol_gate": [
            RelativeVolume(lookback=20),
            QuantileRankTransform(),
            QuantileRangeSelector(low_quantile=50, high_quantile=100),
        ],
    },
    MaskAnd(),
], name="momentum_selection")

Pipeline([
    PriceDataLoader(timeframe="15min"),
    TargetTimeframeResampler(),
    {
        "selection": [selection],
        "market_regime": [EWMA(window=2), MarketTrendRegimeFilter(fast_period=5, slow_period=17)],
        "vol": [ReturnVolatility(window="36d"), Store("return_vol")],
    },
    RegimeGate(signal_key="selection", index_key="market_regime", condition="index > 0"),
    VolWeightSizer(vol_slot="return_vol", target_leverage=1.0, max_weight=0.15),
], name="momentum_hyperliquid")

Globals(target_timeframe="1d", bar_offset="12h")Universe(mode="top_volume", top_n=50, market="perp", resolved=[], resolved_at="")Execution(    rebalance="buffered",    buffer_threshold=0.2,    buffer_mode="relative",    rebalance_method="to_edge",)selection = Pipeline([    {        "momentum_filter": [            ROC(period=14),            QuantileRankTransform(),            QuantileRangeSelector(low_quantile=75, high_quantile=100),        ],        "rel_vol_gate": [            RelativeVolume(lookback=20),            QuantileRankTransform(),            QuantileRangeSelector(low_quantile=50, high_quantile=100),        ],    },    MaskAnd(),], name="momentum_selection")Pipeline([    PriceDataLoader(timeframe="15min"),    TargetTimeframeResampler(),    {        "selection": [selection],        "market_regime": [EWMA(window=2), MarketTrendRegimeFilter(fast_period=5, slow_period=17)],        "vol": [ReturnVolatility(window="36d"), Store("return_vol")],    },    RegimeGate(signal_key="selection", index_key="market_regime", condition="index > 0"),    VolWeightSizer(vol_slot="return_vol", target_leverage=1.0, max_weight=0.15),], name="momentum_hyperliquid")

Explore the settings

precomputed · updates instantly

Adjust a setting to see the exact backtested result — including the ones that lost money.

Drift before rebalancing

Momentum lookback (days)

Momentum selection cutoff

Sharpe

0.98

Return

90.5%

Worst DD

−39.0%

Trades

949

Run these settings live →Fork & customize

At $1,000, these settings' worst historical dip was about $390. You land in the editor with this exact setup. Free account, your own keys.

Compare all 19 settings

Settings	Sharpe	Return	Worst DD	Trades
Drift 0.2 · Momentum 7 · Momentum 60	0.87	74.9%	−40.9%	1,687
Drift 0.2 · Momentum 7 · Momentum 75	0.95	85.9%	−38.6%	1,127
Drift 0.2 · Momentum 7 · Momentum 85	0.78	53.9%	−30.3%	688
Drift 0.2 · Momentum 10 · Momentum 60	0.62	40.4%	−41.3%	1,537
Drift 0.2 · Momentum 10 · Momentum 75	0.74	55.2%	−38.3%	1,054
Drift 0.2 · Momentum 10 · Momentum 85	0.73	47.9%	−34.2%	613
Drift 0.2 · Momentum 14 · Momentum 60	0.73	56.2%	−42.8%	1,481
Drift 0.2 · Momentum 14 · Momentum 75default	0.98	90.5%	−39.0%	949
Drift 0.2 · Momentum 14 · Momentum 85	0.86	62.7%	−29.6%	538
Drift 0.2 · Momentum 20 · Momentum 60	0.51	27.7%	−42.8%	1,427
Drift 0.2 · Momentum 20 · Momentum 75	0.64	40.0%	−37.1%	870
Drift 0.2 · Momentum 20 · Momentum 85	0.71	41.3%	−27.3%	510
Drift 0.2 · Momentum 30 · Momentum 60	0.46	22.2%	−41.8%	1,342
Drift 0.2 · Momentum 30 · Momentum 75	0.68	43.8%	−32.0%	845
Drift 0.2 · Momentum 30 · Momentum 85	0.73	40.9%	−22.2%	450
Drift 0.1 · Momentum 14 · Momentum 75	0.98	98.3%	−42.3%	1,110
Drift 0.3 · Momentum 14 · Momentum 75	0.94	75.6%	−35.0%	843
Drift 0.2 · Momentum 40 · Momentum 75	0.38	14.3%	−31.2%	782
Drift 0.2 · Momentum 60 · Momentum 75	-0.17	-21.4%	−44.2%	701

The data

Monthly returns

Month	Return
2024-08	0.0%
2024-09	+32.9%
2024-10	-7.1%
2024-11	+35.8%
2024-12	+5.8%
2025-01	-1.8%
2025-02	0.0%
2025-03	-13.9%
2025-04	+34.1%
2025-05	-3.8%
2025-06	-7.5%
2025-07	+31.8%
2025-08	-5.6%
2025-09	0.0%
2025-10	-20.0%
2025-11	-7.2%
2025-12	-5.1%
2026-01	+8.4%
2026-02	0.0%
2026-03	+1.5%
2026-04	+9.6%
2026-05	+0.7%
2026-06	0.0%
2026-07	0.0%

Which assets it traded

Share of return	Avg allocation	Days held	max weight	% of time held	Asset
18.88	0.1172	94	0.15	13.6	TAO
16.4	0.1234	61	0.15	8.9	kPEPE
16.36	0.115	93	0.15	13.5	ENA
14.53	0.1458	51	0.15	7.4	XRP
13.79	0.1025	66	0.15	9.6	PENGU
10.56	0.1302	77	0.15	11.2	SUI
8.96	0.1285	27	0.15	3.9	XLM
8.93	0.1227	61	0.15	8.9	CRV
7.8	0.1376	38	0.15	5.5	ADA
7.61	0.1016	44	0.15	6.4	JTO
7.05	0.1132	74	0.15	10.7	WIF
6.92	0.1278	51	0.15	7.4	ONDO
6.09	0.1271	67	0.15	9.7	kBONK
5.29	0.1223	59	0.15	8.6	FET
3.69	0.1109	70	0.15	10.2	HYPE
3.69	0.121	68	0.15	9.9	PENDLE
3.58	0.1426	52	0.15	7.5	BCH
3.3	0.0801	37	0.15	5.4	ZEC
3.07	0.0874	64	0.15	9.3	SPX
2.56	0.0877	55	0.15	8	VVV
2.25	0.1481	24	0.15	3.5	AVAX
1.47	0.0754	56	0.1467	8.1	FARTCOIN
1.4	0.1248	67	0.15	9.7	AAVE
0.75	0.0967	10	0.1491	1.5	XPL
0.45	0.1448	52	0.15	7.5	DOGE
0.32	0.15	9	0.15	1.3	XMR
0.17	0.1493	45	0.15	6.5	BTC
0.06	0.15	32	0.15	4.6	PAXG
0.03	0.0866	12	0.1282	1.7	LIT
0	null	1	0	0.1	GRAM
0	null	1	0	0.1	CHIP
-0.42	0.1286	34	0.15	4.9	SOL
-0.68	0.1271	36	0.15	5.2	NEAR
-0.69	0.1222	62	0.15	9	UNI
-0.71	0.12	50	0.15	7.3	MORPHO
-0.74	0.1428	33	0.15	4.8	TRX
-1.22	0.1052	13	0.15	1.9	WLFI
-1.25	0.1388	42	0.15	6.1	BNB
-1.52	0.0855	11	0.1458	1.6	MET
-1.6	0.1106	25	0.15	3.6	PUMP
-2.27	0.0973	21	0.15	3	MON
-2.35	0.1271	50	0.15	7.3	TIA
-2.59	0.1345	41	0.15	6	ARB
-2.79	0.127	35	0.15	5.1	LTC
-3.26	0.1349	36	0.15	5.2	LINK
-3.49	0.1248	36	0.15	5.2	JUP
-4.07	0.1367	79	0.15	11.5	ETH
-4.21	0.0772	22	0.15	3.2	TRUMP
-5.75	0.118	50	0.15	7.3	WLD
-10.44	0.124	36	0.15	5.2	DYDX

How it did in rising vs falling markets

Market	Days	Return	Sharpe
Rising markets	333	+257.6%	2.59
Falling markets	353	-46.7%	-3.23

What this is: a historical backtest on real Hyperliquid market data, net of fees, slippage, and funding. Its worst historical drawdown was 39.0% — 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.

Momentum Strategy on Hyperliquid

How it works

How it works

When it works best

When it struggles

How it’s built

Configuration

Factories

Explore the settings

The data

Monthly returns

Which assets it traded

How it did in rising vs falling markets

Common questions

Related strategies