Hyperliquid Correlation & Beta: Hedging Crypto Perps (2026)

The short version: most large Hyperliquid perps move together, so holding five of them is not five independent bets. In our snapshot of 29 October 2026 (08:09 UTC), using daily candles up to the 28 October close, the 30-day correlation of daily returns between BTC and ETH was 0.95, and BTC with XRP was 0.93. HYPE was the outlier at 0.54 with BTC. In practice, a $50,000 long book split evenly across BTC, ETH, SOL, XRP and DOGE had almost the same daily risk as the five positions added up, and a BTC short removed about two thirds of the risk of an ETH long but only about a sixth of the risk of a HYPE long.

Correlation is the number traders ignore until a bad day. When everything you hold drops at once, your "diversified" portfolio behaves like one big leveraged position, and on a cross-margin account that can pull every position closer to liquidation at the same time. This guide shows how to measure correlation and beta on Hyperliquid from the free public API, gives real, dated numbers for six of the most traded perps, works through three portfolio and hedge examples, and ends with a short Python script you can run yourself, with no account and no API key.

Correlation and beta in plain terms

• Correlation runs from −1 to +1. It tells you how consistently two coins move in the same direction on the same day. At +1 they always move together; at 0 there is no linear relationship; at −1 they move in opposite directions. It says nothing about how big the moves are.
• Beta adds size. Beta of a coin against BTC is the correlation multiplied by the coin's volatility divided by BTC's volatility. A beta of 1.3 means that, on average over the window, the coin moved about 1.3% for every 1% BTC moved.
• Hedge effectiveness comes straight from correlation. The best hedge with BTC can only remove the part of a coin's daily variance that is explained by BTC, which is the correlation squared. The risk that is left, measured as standard deviation, is the square root of (1 − correlation²). At 0.95 about 31% of the risk remains; at 0.54 about 84% remains.

We measure everything on daily log returns of the perp close prices from Hyperliquid's candleSnapshot endpoint, on the dates all six coins share, and we drop the current day because its candle is still open.

Real numbers: 30-day correlation matrix on 29 October 2026

Daily log returns, 30 days ending with the 28 October 2026 close (UTC).

	BTC	ETH	HYPE	SOL	XRP	DOGE
BTC	1.00	0.95	0.54	0.85	0.93	0.88
ETH	0.95	1.00	0.65	0.91	0.92	0.93
HYPE	0.54	0.65	1.00	0.66	0.70	0.73
SOL	0.85	0.91	0.66	1.00	0.85	0.88
XRP	0.93	0.92	0.70	0.85	1.00	0.90
DOGE	0.88	0.93	0.73	0.88	0.90	1.00

Four things stand out:

• The majors are almost one trade. Every pair among BTC, ETH, SOL, XRP and DOGE was at 0.85 or higher over 30 days.
• HYPE was the least tied to BTC. 0.54 is still positive, so HYPE usually fell on days BTC fell, but much less reliably. It was more correlated with DOGE (0.73) and XRP (0.70) than with BTC.
• ETH, not BTC, was the most central coin. ETH had a correlation of 0.91 or more with BTC, SOL, XRP and DOGE.
• Correlation is not stable. Over 90 days (below), most pairs were lower. Short windows react faster but are noisier.

Beta to BTC: how much each coin moved per 1% of BTC

Coin	Beta vs BTC (30d)	Volatility vs BTC (30d)	Beta vs BTC (90d)	Volatility vs BTC (90d)
ETH	1.28	1.36x	1.22	1.37x
HYPE	1.15	2.12x	1.20	1.85x
SOL	1.23	1.44x	1.23	1.48x
XRP	1.42	1.53x	1.59	1.80x
DOGE	1.56	1.77x	1.45	1.73x

HYPE shows why beta and volatility are different things. It moved more than twice as much as BTC on a typical day over 30 days, yet its beta was only 1.15, because a large part of its moves had nothing to do with BTC. The 30-day daily standard deviations behind these ratios were 1.30% for BTC, 1.76% for ETH, 2.75% for HYPE, 1.87% for SOL, 1.98% for XRP and 2.30% for DOGE. For how to turn those into stop distances, see our volatility and ATR guide.

The 90-day view

	BTC	ETH	HYPE	SOL	XRP	DOGE
BTC	1.00	0.89	0.65	0.83	0.88	0.83
ETH	0.89	1.00	0.72	0.83	0.79	0.76
HYPE	0.65	0.72	1.00	0.70	0.65	0.58
SOL	0.83	0.83	0.70	1.00	0.78	0.79
XRP	0.88	0.79	0.65	0.78	1.00	0.88
DOGE	0.83	0.76	0.58	0.79	0.88	1.00

Over 90 days the majors were still tightly linked (0.76 to 0.89), while HYPE's link with BTC was a bit higher (0.65) than in the last 30 days. The lesson is not a specific number but a habit: check correlation on more than one window before you rely on it, and expect it to rise in sharp sell-offs, when the market tends to trade as one.

Example 1: the "diversified" long book

Suppose you hold five longs of $10,000 each in BTC, ETH, SOL, XRP and DOGE, $50,000 of notional in total. Using the 30-day daily returns from the snapshot:

• Adding up each position's own one-day standard deviation gives about $922.
• The standard deviation of the whole book, which accounts for correlation, was about $884.

Diversification reduced daily risk by only about 4%. For risk purposes this book behaved almost like a single $50,000 long on a coin about 1.4 times as volatile as BTC. If your rule is "risk no more than X per day", size the book as one position, not five. Our position size calculator guide shows the math.

Example 2: hedging an ETH long with a BTC short

You are long $10,000 of ETH and want to keep the ETH-specific view but cut market risk. With a 30-day beta of 1.28, the hedge is a BTC short of about $12,800. In the snapshot window:

• Unhedged daily standard deviation: about $176.
• Hedged daily standard deviation: about $56, roughly a 68% reduction, in line with the 0.95 correlation.

A simple dollar-neutral pair (long $10,000 ETH, short $10,000 BTC) had a daily standard deviation of about $67, also far below the unhedged position. Keep in mind that a hedge has costs: you pay trading fees on both legs and funding on the short leg can be positive or negative. Our funding rate guide explains how that adds up.

Example 3: hedging a HYPE long with a BTC short

Same idea with $10,000 of HYPE. The 30-day beta was 1.15, so the hedge is a BTC short of about $11,500. The result:

• Unhedged daily standard deviation: about $275.
• Hedged daily standard deviation: about $232, only about 16% less.

With a correlation of 0.54, most of HYPE's daily risk is HYPE-specific and a BTC hedge barely touches it. If you want to reduce HYPE exposure, the honest options are a smaller HYPE position or a tighter stop, not a hedge in another coin. This is the most common hedging mistake: assuming a short in BTC protects any crypto long.

Why correlation matters more on Hyperliquid's cross margin

On a cross-margin account, all positions share the same collateral. When correlated longs fall together, unrealized losses pile up against one margin pool and every liquidation price moves closer at once. Isolated margin limits the damage to each position but does not change the fact that they lose money on the same days. Read our isolated vs cross margin guide and the liquidation guide before running several correlated positions at high leverage.

Measure it yourself: Python script

The script below uses only the public candleSnapshot endpoint (one request per coin, well within the rate limits). It aligns the coins on shared dates, drops the unfinished daily candle and prints 30- and 90-day correlation matrices plus beta and relative volatility versus BTC. Change COINS to any perp listed on the main DEX.

import requests, time, math, statistics as st
from datetime import datetime, timezone

API = "https://api.hyperliquid.xyz/info"
COINS = ["BTC", "ETH", "HYPE", "SOL", "XRP", "DOGE"]
WINDOWS = (30, 90)  # look-back windows in daily candles

def candles(coin, days):
    now = int(time.time() * 1000)
    c = requests.post(API, json={"type": "candleSnapshot", "req": {
        "coin": coin, "interval": "1d",
        "startTime": now - (days + 5) * 86_400_000, "endTime": now}}, timeout=20).json()
    return {x["t"]: float(x["c"]) for x in c[:-1]}  # drop today's open candle

def corr(a, b):
    ma, mb = st.mean(a), st.mean(b)
    cov = sum((x - ma) * (y - mb) for x, y in zip(a, b))
    return cov / math.sqrt(sum((x - ma) ** 2 for x in a) * sum((y - mb) ** 2 for y in b))

print("snapshot", datetime.now(timezone.utc).strftime("%Y-%m-%d %H:%M UTC"))
px = {c: candles(c, max(WINDOWS)) for c in COINS}
days = sorted(set.intersection(*(set(p) for p in px.values())))  # common dates only
rets = {c: [math.log(px[c][b] / px[c][a]) for a, b in zip(days, days[1:])] for c in COINS}
print("last close:", datetime.fromtimestamp(days[-1] / 1000, timezone.utc).date())

for w in WINDOWS:
    print(f"\n{w}-day correlation of daily log returns")
    print("      " + "".join(f"{c:>7}" for c in COINS))
    for a in COINS:
        print(f"{a:6}" + "".join(f"{corr(rets[a][-w:], rets[b][-w:]):>7.2f}" for b in COINS))
    print(f"{'coin':6} {'beta vs BTC':>12} {'vol ratio':>10}")
    btc = rets["BTC"][-w:]
    for c in COINS[1:]:
        r = rets[c][-w:]
        beta = corr(r, btc) * st.stdev(r) / st.stdev(btc)
        print(f"{c:6} {beta:>12.2f} {st.stdev(r) / st.stdev(btc):>10.2f}")

Ideas to extend it: use 4-hour candles for a faster read, add a rolling 30-day correlation to see how it changes, or compare your wallet's daily PnL with BTC returns to estimate the beta of your own trading. For longer history and other data sources, see our historical data guide.

See your real exposure, not just your positions

Correlation tells you how your positions will behave together; your account shows whether that is happening. Hyperfolio is a free, read-only tracker for Hyperliquid: paste one or several wallet addresses (no keys, no wallet connection) and see open positions, leverage, liquidation prices, PnL and funding across them in one place. If you split trading over several wallets, our guide to tracking multiple Hyperliquid wallets shows how to see the combined exposure.

FAQ

How correlated are BTC and ETH on Hyperliquid?

Very. In our 29 October 2026 snapshot, the correlation of daily returns was 0.95 over 30 days and 0.89 over 90 days. The relationship changes over time, so recalculate it with the script.

Is HYPE correlated with Bitcoin?

Positively, but loosely: 0.54 over 30 days and 0.65 over 90 days in our snapshot, the lowest of the six coins we measured. That makes BTC a weak hedge for a HYPE position.

Does Hyperliquid show correlation in the app?

No. The trading interface and the info API give prices and candles but no correlation or beta figures. You calculate them from candles, as in the script above.

Which window should I use, 30 or 90 days?

Use both. Thirty days reacts faster to the current market; ninety days is more stable. If the two disagree a lot, treat the relationship as unreliable.

Does a BTC short fully hedge an altcoin long?

No. A beta-sized BTC hedge can only remove the share of risk explained by correlation. In our snapshot it cut an ETH long's daily risk by about 68% but a HYPE long's by only about 16%.

Trade for less. New to Hyperliquid? Use referral code HYPERFOLIO for 4% off trading fees (how the referral works), then track your positions, leverage and PnL for free in Hyperfolio.