Strategies and backtests

Hyperliquid Strategies: 6 Automated Setups Traders Use, Backtested

A moving-average trend, a breakout, a volatility stop, dip buying, a grid and weekly buys: the rule behind each, how it runs as an automation, and how each did on six Hyperliquid markets over the same 180 days, losses included.

[

Updated November 15, 2026

]
•
[

22

 

min to read

]

The short answer

A Hyperliquid trading strategy is a fixed rule for when to open and close a position, which a bot or AI agent can run around the clock. We backtested six common ones on the same six markets and 180 days. None won everywhere: rules that sat in cash lost less as BTC and gold fell, then lagged HYPE’s rally.

What is a Hyperliquid trading strategy, and what changes on perps?

A trading strategy is a rule that decides, with no judgment calls, when to open a position, how big it is and when to close it. A bot, or an AI agent, is the software that follows the rule for you. On Hyperliquid, most strategies trade perps, short for perpetual futures: contracts that follow a price and never expire. Four things set them apart from buying a coin outright:

  • You can go short. A rule can try to profit from a fall instead of waiting in cash. We test that below.
  • Leverage. A position can be bigger than the money behind it, its margin. The catch is liquidation: Hyperliquid closes a position by force when the account’s equity falls below the maintenance margin, which is half the initial margin at the market’s maximum leverage.
  • Funding. Longs and shorts pay each other every hour to keep the perp near the price it tracks. Each payment is small; over months they add up.
  • Markets that never close. Crypto trades around the clock, and trade.xyz, an exchange built on Hyperliquid, lists perps on stocks, indices and commodities that keep trading on its own internal pricing while the real market is shut, as its specification index sets out market by market.

Rules suit these markets because a bot never sleeps and never hesitates. That is also the risk: it repeats its rule every time, the bad trades included. Automation doesn’t make a rule good, so the useful question is how each rule has behaved, and that is what this guide measures. For how bots connect to Hyperliquid and how to choose one, see our guide to Hyperliquid trading bots. For what an AI agent adds to a plain bot, see what an AI trading agent is.

How did we test the six setups?

Every setup ran on the same data, dates, markets and costs, so the results compare like with like. Our guide on how to backtest a trading strategy explains why each of these choices matters; here is what we used.

  • Data: daily candles from Hyperliquid’s public API, taken on November 14, 2026, at 8:52 PM UTC. We dropped that day’s candle, which was still open. One candle is one UTC day, weekends included.
  • Markets: BTC, ETH, HYPE and SOL, the four most traded perps on Hyperliquid’s main exchange that day, plus two trade.xyz perps: xyz:SP500, which tracks the S&P 500 index, and xyz:GOLD, which tracks one troy ounce of gold.
  • Window: 180 days, May 18 to November 13, 2026. The S&P 500 perp’s price history on Hyperliquid starts on March 18, 2026, and the slowest rule needs 50 days of prices before its first signal, so about six months is what all six markets share. It is also the daily window that Ethy’s backtester runs on its free plan, so you can test rules like these there.
  • Trades: each rule reads the daily close and trades at the next day’s open. Long only, no leverage, $10,000 per market, all in or all out. The grid and the weekly buys split the money, as their sections explain. One section below adds shorts.
  • Costs: every buy and every sell is charged 0.1% of its value to stand in for trading fees and slippage. Funding is not included.
  • Counting: a position still open on November 13 is valued at that day’s close and counts as a trade. A trade is “won” if it ended with more money than it started with.
  • Benchmark: buying at the May 18 open and holding to the November 13 close.
  • Settings: textbook defaults, fixed before the test and never tuned: EMAs of 20 and 50 days, a 14-day RSI at 30 and 70, a 20-day high and a 10-day low, 3 times a 14-day ATR, and a grid over the past 30 days’ range. Each indicator uses the market’s full history in the data, so all were warmed up by May 18.

Which Hyperliquid strategy did best in the test?

First, what the markets did. Over the 180 days, four of the six fell: BTC -11.9%, ETH -9.5%, SOL -8.4% and gold -14.5%. HYPE rose 56.7% and the S&P 500 1.8%. That split explains most of what follows.

Volatility stop$11,012EMA trend$10,656Breakout$10,444Buy and hold$10,235RSI dips$10,035Grid$9,730Weekly DCA$9,525Volatility stop$11,012EMA trend$10,656Breakout$10,444Buy and hold$10,235RSI dips$10,035Grid$9,730Weekly DCA$9,525
What $10,000 split equally across BTC, ETH, SOL, HYPE, the S&P 500 and gold became under each setup, and under buying and holding the same six, May 18 to November 13, 2026. Every buy and sell charged 0.1%.Daily candles from Hyperliquid’s public API, taken November 14, 2026; Ethy’s calculation.
The six setups against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
SetupSix markets, $10,000 splitAverage worst dropTrades wonBeat holding in
EMA trend+6.6%-15.1%7 of 15 (47%)5 of 6 markets
Breakout+4.4%-13.7%8 of 17 (47%)4 of 6 markets
Volatility stop+10.1%-15.0%6 of 17 (35%)4 of 6 markets
RSI dips+0.4%-9.2%4 of 8 (50%)4 of 6 markets
Grid-2.7%-15.3%197 of 233 (85%)5 of 6 markets
Weekly DCA-4.7%-20.9%55 of 156 (35%)3 of 6 markets
Buy and hold+2.3%-24.1%Not applicableNot applicable

The average worst drop is the mean of each market’s deepest fall from a peak. RSI dips never traded two of the six markets, which flatters its average. What the numbers say, and what they don’t:

  • Beating holding mostly meant sitting in cash. In the four markets that fell, a setup beat holding in 23 of 24 cases, because each one spent part of the time out of the market. In HYPE, the market that rose most, every setup trailed holding: the best, the trend rule, made +14.0% against +56.7%.
  • The volatility stop came first and the trend rule second, on 17 and 15 trades across six markets. That is too few trades to call either one an edge.
  • A high win rate can hide a loss. The grid won 85% of its 233 trades and still lost money, because its losers were positions it still held, bought on the way down.
  • Weekly buys came last, at -4.7%: buying a little each week softened the falls but caught little of HYPE’s rally, which holding owned from the first day.

The ranking also depends on the dates. Over a longer window, further down, the order changes.

How do trend-following setups work on Hyperliquid?

Trend rules try to be in the market while it rises and out of it while it falls. They are slow on purpose: they buy after a move has started and sell after it has turned, so they give up the first and last part of every trend in exchange for staying out of the long falls.

1. Trend following: EMA 20 over EMA 50

The rule: be long while the 20-day exponential moving average (EMA) of the close is above the 50-day EMA, and in cash while it is below. An EMA is an average that gives more weight to recent days; StockCharts shows the formula.

As an automation: a check once a day, after the daily close. When the 20-day average crosses above the 50-day one, the bot opens a long at market; when it crosses back below, it closes. It is the classic moving-average cross. The rule has no stop of its own, so traders often add one for the days between signals.

When it works, and when it fails: it works in long, steady moves. It fails in choppy markets, where the averages cross back and forth and each false signal costs a little; StockCharts warns that crossovers are prone to these whipsaws. It is also late by design, since moving averages always trail the price.

The EMA 20/50 trend rule against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC-0.3%-11.9%-11.5%-24.5%1 of 2
ETH+21.4%-9.5%-15.2%-30.6%1 of 2
SOL+3.8%-8.4%-17.4%-36.0%2 of 5
HYPE+14.0%+56.7%-32.7%-30.1%2 of 3
S&P 500+2.1%+1.8%-4.8%-6.4%1 of 2
Gold-1.8%-14.5%-9.0%-17.1%0 of 1

What the test showed: it beat holding in 5 of 6 markets, mostly by stepping aside. On BTC it sold on June 1 and bought back on August 21, so it sat out BTC’s fall to a close of $58,603 on June 30. Its best trade held HYPE from May 18 to July 26 for +26.9%. SOL shows the whipsaw: three losing trades between July 11 and July 25, then +18.4% from August 20 to October 28.

$7k$8k$9k$10k$11k$12kMay 2026Aug 2026Nov 2026BuySellBuySellEMA 20/50 ruleBuy and hold$6k$8k$10k$12kMay 2026Nov 2026EMA 20/50 ruleBuy and hold
Value of $10,000 on BTC following the EMA 20/50 rule (solid line) and buying and holding (dashed), May 18 to November 13, 2026. Dots mark each buy and sell.Daily candles of BTC from Hyperliquid’s public API, taken November 14, 2026; Ethy’s calculation.

2. Breakout: a 20-day high in, a 10-day low out

The rule: buy when the daily close rises above the highest high of the previous 20 days; sell when it closes below the lowest low of the previous 10. Those highs and lows are the edges of a price channel, which StockCharts credits to Richard Donchian; it is also called a Donchian channel.

As an automation: a daily check of the close against the two channel lines. Some traders place a buy order just above the 20-day high instead, so it fills during the day rather than at the next open.

When it works, and when it fails: it catches the big moves that start with a new high, and it pays for them with false breakouts: new highs that fade, each one a small loss.

The 20-day breakout with a 10-day exit against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC+9.2%-11.9%-12.5%-24.5%1 of 2
ETH-4.5%-9.5%-12.3%-30.6%1 of 3
SOL+12.5%-8.4%-12.8%-36.0%2 of 3
HYPE+13.7%+56.7%-30.9%-30.1%2 of 4
S&P 500-5.3%+1.8%-5.7%-6.4%1 of 4
Gold+1.0%-14.5%-7.7%-17.1%1 of 1

What the test showed: its biggest win was HYPE from August 18 to September 11, +31.9%. The S&P 500 shows how it fails: four breakouts, one small win, -5.3% in all while the index rose +1.8%. Each new high faded before it could run.

3. Volatility stop: three ATRs from the best close

The rule: be long while the close stays above a stop set three ATRs under the highest close since the trend turned up, and sell when a close falls below it. Then wait in cash until a close rises above a line three ATRs over the lowest close since the sale. ATR, the average true range, is J. Welles Wilder’s measure of how far a market moves in a typical day, gaps included, usually over 14 days (StockCharts).

As an automation: a trailing stop whose distance follows volatility and is recomputed every day. Hyperliquid’s own trailing stop trails by a fixed distance or percentage, so a rule like this needs a bot or agent that moves the stop as ATR changes, or a percentage you work out from the ATR when you set it.

When it works, and when it fails: the wide stop lets a trend run through normal pullbacks. The price of that room is giving back a lot before the stop is hit, and getting stopped out of slow, grinding markets.

The 3 ATR volatility stop against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC+19.7%-11.9%-7.7%-24.5%1 of 2
ETH+36.7%-9.5%-11.0%-30.6%1 of 1
SOL+21.0%-8.4%-13.6%-36.0%1 of 2
HYPE-1.7%+56.7%-39.7%-30.1%2 of 4
S&P 500-5.4%+1.8%-8.0%-6.4%0 of 4
Gold-9.6%-14.5%-10.4%-17.1%1 of 4

What the test showed: the room paid off on BTC and ETH: ETH’s single trade, July 13 to October 9, made +36.7%. The same room cost it on HYPE, where two trades gave back 17.7% and 16.5% before the stop was hit, and on the S&P 500, where all four trades lost a little.

How do dip-buying and grid setups work, and why do they struggle in trends?

These two bet the other way: that a move too far in one direction comes back. They do well while a market swings inside a range and badly when it trends away, the opposite of the three setups above, so it pays to know which kind of market you are in.

4. Buying dips: RSI under 30, out over 70

The rule: buy when the 14-day RSI closes below 30; sell when it closes above 70. The RSI, or Relative Strength Index, is Wilder’s 0 to 100 gauge of recent gains against recent losses, and StockCharts explains why readings under 30 are called oversold and over 70 overbought.

As an automation: an indicator condition checked at each daily close that opens a long when it fires. Because the rule has no stop, a fall that keeps going is held all the way down, so traders often add one.

When it works, and when it fails: it works when sharp drops bounce. It fails when oversold gets more oversold in a downtrend, and it can wait months for a signal.

The RSI(14) 30/70 dip rule against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC-1.0%-11.9%-12.3%-24.5%1 of 2
ETH+13.5%-9.5%-15.8%-30.6%1 of 2
SOL-6.2%-8.4%-18.9%-36.0%1 of 2
HYPE0.0%+56.7%No trades-30.1%No trades
S&P 5000.0%+1.8%No trades-6.4%No trades
Gold-4.2%-14.5%-8.2%-17.1%1 of 2

What the test showed: on BTC, ETH and SOL it bought on June 3 and sold on August 20, for +3.7%, +20.9% and +14.9%. Then it bought into the autumn slide, and those trades were still under water on November 13: SOL’s, bought on October 30, was down 18.4%. It never traded HYPE or the S&P 500, whose RSI never closed under 30 in the window.

5. Grid: a ladder across last month’s range

The rule: draw ten price lines spread evenly from the lowest low to the highest high of the 30 days before the start, and split the money into nine slots of $1,111, one between each pair of lines. A slot buys when the price falls to its lower line and sells when it reaches the next line up, over and over. At the start, the slots above the price are bought at the open, so the grid has something to sell if the price rises. Our grid trading bot guide runs this design on BTC through a range, a fall and a rally.

As an automation: a ladder of resting limit orders, each re-placed on the other side after it fills. Hyperliquid’s scale order places several limit orders across a price range at once; once they fill they are done, so a grid that keeps re-arming needs a bot.

When it works, and when it fails: it collects many small gains while the price swings inside the range. When the price leaves the range, it ends up either holding every slot, bought on the way down, or holding cash while the market rallies without it.

The 10-line grid against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC-3.4%-11.9%-23.3%-24.5%65 of 74
ETH-3.6%-9.5%-26.1%-30.6%38 of 47
SOL-0.3%-8.4%-27.2%-36.0%36 of 45
HYPE+0.6%+56.7%0.0%-30.1%2 of 2
S&P 500+2.4%+1.8%-0.7%-6.4%37 of 37
Gold-11.8%-14.5%-14.5%-17.1%19 of 28

What the test showed: the price left the range in every market. The close sat below the grid’s range on 86 of 180 days on BTC, 90 on ETH, 76 on SOL and 129 on gold. There the grid ended with all nine slots filled, its lots about 13% under water on average. HYPE closed above its range on all 180 days, so the grid made two trades and kept most of its money in cash (+0.6%) while HYPE rose +56.7%. One limit of this test: daily candles don’t show whether the high or the low came first, so we assumed the low came first on days that closed up and the high first on days that closed down. A real grid fills at its lines during the day and may trade more often.

Is DCA a good Hyperliquid strategy for beginners?

6. Weekly buys: the same amount every Monday

The rule: buy the same dollar amount every Monday at the open, $384.62 a week for 26 weeks here, so the $10,000 is fully invested by the end, and never sell. This is dollar-cost averaging, or DCA: investing equal amounts at set intervals, whatever the price does. Our DCA backtest runs this same idea over three years of BTC and ETH prices against a lump sum.

As an automation: a schedule, not a signal: one market order a week. On a perp you also pick the leverage (none here) and pay or receive funding for as long as you hold. Hyperliquid’s TWAP order is a different tool: it splits one order into small pieces over at most seven days, to enter one position gently.

When it works, and when it fails: it takes timing off your hands and lowers the average price in a decline. It is not a way to beat buying at once: Vanguard found that investing a lump sum beat cost averaging about two-thirds of the time in historical and simulated data, because money waiting in cash misses what the market gains. Its case is for someone who would otherwise not invest at all, or who can’t stomach putting everything in on one day.

Weekly buys against buy and hold on six Hyperliquid markets, May 18 to November 13, 2026
MarketSetupHoldingSetup’s worst dropHolding’s worst dropTrades won
BTC-5.1%-11.9%-20.0%-24.5%11 of 26 buys in profit
ETH-7.6%-9.5%-28.5%-30.6%10 of 26 buys in profit
SOL-10.0%-8.4%-33.9%-36.0%10 of 26 buys in profit
HYPE+3.6%+56.7%-23.9%-30.1%13 of 26 buys in profit
S&P 500-0.8%+1.8%-5.9%-6.4%11 of 26 buys in profit
Gold-8.6%-14.5%-13.3%-17.1%0 of 26 buys in profit

What the test showed: on BTC the weekly buys averaged $71,907, 7.1% below the first day’s price, so DCA lost 5.1% where holding lost 11.9%. On HYPE they averaged $69.18, 51.4% above the first day’s $45.69, and made +3.6% against +56.7%. Gold fell for most of the window, and none of its 26 buys was in profit at the end.

For a beginner, DCA and the trend rule are the easiest of the six to reason about: one fixed schedule, or one slow signal, with no leverage. Easy to follow is not the same as profitable; on these dates DCA finished last.

Should an automated perps strategy go short too?

On a perp, the trend rule doesn’t have to wait in cash. Its long and short version is long while the 20-day EMA is above the 50-day EMA and short while it is below, at 1x, flipping at the next open. Same dates, markets and charges as before.

The EMA 20/50 trend rule long only and long and short, May 18 to November 13, 2026
MarketLong onlyLong and shortHoldingLong and short worst drop
BTC-0.3%+11.4%-11.9%-16.4%
ETH+21.4%+56.3%-9.5%-17.8%
SOL+3.8%+7.6%-8.4%-42.0%
HYPE+14.0%-25.8%+56.7%-58.3%
S&P 500+2.1%+2.4%+1.8%-5.2%
Gold-1.8%+11.6%-14.5%-9.5%

Shorting helped where the trend fell for long stretches: ETH went from +21.4% long only to +56.3%, and gold from -1.8% to +11.6%. It hurt where the trend kept turning: HYPE’s two closed shorts lost 27.2% and 17.0%, and that account fell 58.3% from its peak. Across the six markets, the long and short version ended at +10.6% against +6.6% long only, with deeper drops on the way.

Three things to weigh before you automate shorts:

  • Rises have no ceiling. A 1x long is wiped out only if the price falls almost to zero. A 1x short is liquidated before the price doubles, and prices do double.
  • Leverage shrinks the room. On a market whose maximum is 10x, a 10x isolated position is liquidated after a move of about 5% against it, by Hyperliquid’s liquidation rules. A daily rule like this one routinely sits through moves bigger than that.
  • Funding changes sides. When the perp trades above its reference price, longs pay shorts; when it trades below, shorts pay longs. Our tests leave funding out, long or short.

Does the test window change which strategy wins?

A lot. We ran the same rules, unchanged, over the longest window the data allows for the four crypto markets with the same 50 days of history before it: 350 days, November 29, 2025 to November 13, 2026. The S&P 500 and gold perps don’t have that much history. Over those dates BTC fell 24.9%, ETH 36.4% and SOL 43.1%, while HYPE more than doubled, +104.7%.

$10,000 split across BTC, ETH, SOL and HYPE under each setup, over two windows that end on November 13, 2026
Setup180 days, from May 18350 days, from November 29, 2025
EMA trend+9.7%+12.5%
Breakout+7.7%-11.9%
Volatility stop+18.9%-9.2%
RSI dips+1.6%-1.0%
Grid-1.7%-12.1%
Weekly DCA-4.8%+6.6%
Trend, long and short+12.4%+26.1%
Buy and hold+6.7%0.0%
  • The leader changed. Over 180 days the best setup on these four markets was the volatility stop; over 350 days it was the trend rule, and the volatility stop went to -9.2%.
  • Only the trend rule, long only or long and short, beat holding in both windows. Even that rests on a handful of trades per market.
  • The grid lost in both. A range set once, from one month of prices, didn’t hold in any of these markets.

A single test window can crown almost any of these setups. Before you automate one, test it over more than one stretch of the market, on the markets you plan to trade, and then on paper, with live prices and no money at stake.

How do you run these setups with Ethy?

Ethy is an AI trading agent you ask before you trade. You can describe rules like these in plain English, or start from one of its templates. Ethy shows the rules it built so you can check them, backtests the exact rules it would run, and runs them as an automation, live or in paper mode, which uses live prices and moves no money.

Where each of the six setups starts in Ethy
SetupWhere to start in Ethy
1. Trend followingThe MA Cross template, or ask for “long while EMA 20 is above EMA 50 on the daily”
2. BreakoutThe Donchian Breakout template
3. Volatility stopA trailing stop on the position or in the automation; it can tighten in tiers as the trade gains
4. Buying dipsThe RSI Reversion template
5. GridThe closest is the Accumulation Grid template, for spot tokens; it is not the same grid as the one tested here, so read its rules in the preview
6. Weekly buysThe DCA or Smart DCA template, for spot tokens, on a schedule such as every Monday
  • Backtests: on Hyperliquid’s main exchange and trade.xyz perps, over up to 365 days: 180 or 365 on the daily, with the free plan running 180. Up to 10 markets in one run, with the return, max drawdown, trades, win rate and the comparison with buy and hold, and every entry and exit on the chart.
  • The same reading live: indicator conditions are checked when a bar closes, on the last closed candle, the same way the backtest reads them.
  • Exits and limits: take profit and stop loss as a percent or a price, and take profit in dollars too; stops behind market structure; trailing stops; a max daily loss and a max drawdown in dollars; trading-hours windows; and a pause after 10 losing trades in a row, on by default.
  • Running: you press Activate once, and the automation trades by itself on Ethy’s servers, even while you’re offline. One automation can run the same rule on up to 10 markets.
  • Limits to know: one automation per Hyperliquid market per wallet, because Hyperliquid keeps one position per market. No conditions across markets, such as “buy ETH when BTC drops”. No resting limit orders: an entry at a level fills at market when the level prints.

Summary

  • A Hyperliquid trading strategy is a fixed rule, and a bot or AI agent runs it. On perps the rule can also short, use leverage and pay or receive funding.
  • We tested six setups on the same 180 days (May 18 to November 13, 2026), six markets and a 0.1% charge per trade. Holding the six made +2.3%; the setups ranged from -4.7% to +10.1%.
  • Most setups beat holding in the markets that fell, mostly by sitting in cash, and every one trailed HYPE’s +56.7%.
  • A high win rate can hide a loss: the grid won 85% of its trades and still lost money.
  • Over a second window of 350 days on the crypto markets the ranking changed; only the trend rule, long only or long and short, beat holding in both.
  • Ethy can backtest rules like these on Hyperliquid perps over up to 365 days and run them as automations with stops, targets and trailing stops, live or on paper.

Questions people ask

What is the best Hyperliquid trading strategy?

None of the six won everywhere. In our test, the volatility stop led over 180 days and lost money over 350 days on the four crypto markets, and only the trend rule, long only or long and short, beat holding in both windows, on few trades. Pick a rule whose way of losing you can live with, and test it on your own markets and dates.

What is the best Hyperliquid strategy for beginners?

Start with rules you can explain in one sentence and run without leverage, such as weekly buys or a slow moving-average trend rule, and run them on paper first. Learn how each one loses money before you add size, leverage or a second rule.

Can you automate a Hyperliquid strategy without coding?

Yes. Hosted bot platforms, Telegram bots and AI agents run rules without code, and with an agent such as Ethy you can describe the rule in plain English. If you code, you can write the bot yourself against Hyperliquid’s public API.

How much money do you need to run these setups?

Hyperliquid rejects orders worth less than $10, so each weekly buy or grid slot needs at least that. A grid with nine slots needs at least $90. Start with an amount you can afford to lose while you learn how the rule behaves.

Do these backtests include funding?

No. Funding is paid every hour on Hyperliquid perps: longs pay shorts when the perp trades above its reference price, and shorts pay longs when it trades below. Positions held for months, such as weekly buys on a perp, are the most exposed to it.

Should you use leverage with these strategies?

Our tests used none. Leverage multiplies results in both directions, drawdowns included, and adds liquidation: a big enough move closes the position by force before the rule’s own exit. The worst drops in our long-only tables reached about 40%; at 3x or 5x they would have been several times deeper, and a deep enough one ends in liquidation.

Can AI find a profitable Hyperliquid strategy for you?

AI can help you write a rule, test it and run it, and it reads charts and data faster than you. It can’t make a rule that worked on past prices keep working. Judge an AI-made strategy as you would your own: by its trades, its worst drop, and how it did on dates it wasn’t built on.

Sources

  1. Hyperliquid Docs: Info endpoint (candle snapshot)hyperliquid.gitbook.io
  2. Hyperliquid Docs: Fundinghyperliquid.gitbook.io
  3. Hyperliquid Docs: Order types (scale, TWAP, trailing stop)hyperliquid.gitbook.io
  4. Hyperliquid Docs: Liquidationshyperliquid.gitbook.io
  5. trade.xyz Docs: Specification Index (SP500 and GOLD contracts and sessions)docs.trade.xyz
  6. StockCharts ChartSchool: Moving Averages, Simple and Exponentialchartschool.stockcharts.com
  7. StockCharts ChartSchool: Relative Strength Index (RSI)chartschool.stockcharts.com
  8. StockCharts ChartSchool: Price Channels (Donchian Channels)chartschool.stockcharts.com
  9. StockCharts ChartSchool: Average True Range (ATR)chartschool.stockcharts.com
  10. Vanguard Research, Cost averaging: Invest now or temporarily hold your cash? (February 2023)corporate.vanguard.com

Trading involves risk, including the loss of your money. Ethy AI is a research and trading tool and does not give investment advice or recommendations. Backtests are simulations on historical data and do not predict future results.

Ethy blog

Read more

The $ETHY ecosystem

Stop watching charts.
Start living.

Deploy your AI trading agent in one click.