Intro to Aperiodic Derivatives Metrics

This notebook introduces a simple derivatives regime dashboard using Aperiodic's funding, open_interest, and basis metrics for Binance BTC perpetuals over September 1, 2025 → February 28, 2026 at 5-minute frequency.

Why derivatives metrics matter

Derivatives data helps answer a different question than trade flow: how crowded is positioning, and how expensive is leverage?

• Funding tracks the periodic transfer that keeps perpetuals aligned with spot.
• Open interest helps show whether leverage is building or being unwound.
• Basis measures the futures/perpetual premium or discount relative to a reference price.

Together, these metrics are useful for spotting crowded bullish or bearish regimes, squeezes, and leverage resets.

Background reading

• Aperiodic derivatives market data: https://aperiodic.io/metrics/derivatives_market_data
• Binance Academy, funding rates: https://www.binance.com/en/academy/articles/what-are-funding-rates-in-crypto-markets
• CME Group, futures basis overview: https://www.cmegroup.com/education/courses/introduction-to-basis.html

from __future__ import annotations

import os
from datetime import date

import matplotlib.dates as mdates
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns
from aperiodic import get_derivative_metrics, get_ohlcv

sns.set_theme(style="whitegrid", context="talk", palette="deep")
pd.options.display.float_format = "{:,.4f}".format
plt.rcParams["figure.figsize"] = (14, 6)
plt.rcParams["figure.dpi"] = 140
plt.rcParams["savefig.dpi"] = 140
plt.rcParams["axes.spines.top"] = False
plt.rcParams["axes.spines.right"] = False

EXCHANGE = "binance-futures"
SYMBOL = "perpetual-BTC-USDT:USDT"
INTERVAL = "5m"
START_DATE = date(2025, 9, 1)
END_DATE = date(2026, 2, 28)
START_TS = pd.Timestamp(START_DATE)
END_TS = pd.Timestamp(END_DATE) + pd.Timedelta(days=1)

API_KEY = "..."  # Set via APERIODIC_API_KEY env var or .env file
if API_KEY == "...":
    API_KEY = os.getenv("APERIODIC_API_KEY", "...")
if API_KEY == "...":
    raise RuntimeError("Set APERIODIC_API_KEY in the environment or in .env.")


def clip_window(df: pd.DataFrame) -> pd.DataFrame:
    out = df.copy()
    out["time"] = pd.to_datetime(out["time"])
    out = out.loc[(out["time"] >= START_TS) & (out["time"] < END_TS)]
    return out.sort_values("time").reset_index(drop=True)

def format_time_axis(ax):
    locator = mdates.AutoDateLocator(minticks=6, maxticks=10)
    formatter = mdates.ConciseDateFormatter(locator)
    ax.xaxis.set_major_locator(locator)
    ax.xaxis.set_major_formatter(formatter)



def to_bps(series: pd.Series) -> pd.Series:
    return series * 10_000

Fetch price, funding, open interest, and basis

price = clip_window(
    get_ohlcv(
        api_key=API_KEY,
        timestamp="true",
        interval=INTERVAL,
        exchange=EXCHANGE,
        symbol=SYMBOL,
        start_date=START_DATE,
        end_date=END_DATE,
        output="pandas",
        show_progress=False,
    )
)[["time", "close"]]

funding = clip_window(
    get_derivative_metrics(
        api_key=API_KEY,
        metric="funding",
        timestamp="true",
        interval=INTERVAL,
        exchange=EXCHANGE,
        symbol=SYMBOL,
        start_date=START_DATE,
        end_date=END_DATE,
        output="pandas",
        show_progress=False,
    )
)

oi = clip_window(
    get_derivative_metrics(
        api_key=API_KEY,
        metric="open_interest",
        timestamp="true",
        interval=INTERVAL,
        exchange=EXCHANGE,
        symbol=SYMBOL,
        start_date=START_DATE,
        end_date=END_DATE,
        output="pandas",
        show_progress=False,
    )
)

basis = clip_window(
    get_derivative_metrics(
        api_key=API_KEY,
        metric="basis",
        timestamp="true",
        interval=INTERVAL,
        exchange=EXCHANGE,
        symbol=SYMBOL,
        start_date=START_DATE,
        end_date=END_DATE,
        output="pandas",
        show_progress=False,
    )
)

derivs = (
    price.merge(funding, on="time", how="inner")
    .merge(oi, on="time", how="inner")
    .merge(basis, on="time", how="inner")
)

derivs["price_return_bps_1h"] = derivs["close"].pct_change(12) * 10_000

derivs["funding_rate_bps"] = derivs["funding_rate"] * 10_000
for col in ["funding_rate_bps", "basis_bps", "open_interest"]:
    derivs[f"{col}_z_1d"] = (derivs[col] - derivs[col].rolling(288).mean()) / derivs[col].rolling(288).std()

derivs["oi_change_1h_pct"] = derivs["open_interest"].pct_change(12) * 100

derivs["next_1h_return_bps"] = derivs["close"].pct_change(12).shift(-12) * 10_000

derivs["stress_regime"] = (
    (derivs["funding_rate_bps_z_1d"] > 1.5)
    & (derivs["basis_bps_z_1d"] > 1.5)
    & (derivs["open_interest_z_1d"] > 1.0)
)

derivs["crowding_score"] = derivs[["funding_rate_bps_z_1d", "basis_bps_z_1d", "open_interest_z_1d"]].mean(axis=1)

summary = pd.Series(
    {
        "Rows": len(derivs),
        "Start": derivs["time"].min(),
        "End": derivs["time"].max(),
        "Mean funding (bps)": derivs["funding_rate_bps"].mean(),
        "Mean basis (bps)": derivs["basis_bps"].mean(),
        "Median open interest": derivs["open_interest"].median(),
        "Stress regime share": derivs["stress_regime"].mean(),
    }
)
summary

Rows                                  52128
Start                   2025-09-01 00:00:00
End                     2026-02-28 23:55:00
Mean funding (bps)                   0.3911
Mean basis (bps)                    -4.2997
Median open interest            89,319.8745
Stress regime share                  0.0053
dtype: object

Charts 1-2 — BTC price and funding rate

fig, (ax_price, ax_funding) = plt.subplots(2, 1, figsize=(14, 10), sharex=True)
ax_price.plot(derivs["time"], derivs["close"], color="#2563eb", linewidth=1.2)
ax_price.set_title("BTC perpetual close price")
ax_price.set_ylabel("Price (USDT)")
format_time_axis(ax_price)

ax_funding.plot(derivs["time"], derivs["funding_rate_bps"], color="#dc2626", linewidth=1)
ax_funding.axhline(0, color="black", linestyle="--", linewidth=0.8)
ax_funding.set_title("Funding rate (bps)")
ax_funding.set_ylabel("bps")
format_time_axis(ax_funding)
plt.tight_layout()

Chart 3 — Funding distribution

fig, ax = plt.subplots()
ax.hist(derivs["funding_rate_bps"].dropna(), bins=80, color="#f97316", alpha=0.8, edgecolor="white")
ax.axvline(0, color="black", linestyle="--", linewidth=1)
ax.set_title("Distribution of 5-minute funding readings")
ax.set_xlabel("Funding rate (bps)")
plt.tight_layout()

Chart 4 — Price vs open interest

A rising market with rising open interest often suggests leverage is building alongside the move.

fig, ax1 = plt.subplots(figsize=(14, 6))
ax1.plot(derivs["time"], derivs["close"], color="#2563eb", linewidth=1.1, label="Close")
ax1.set_ylabel("Price (USDT)", color="#2563eb")
ax2 = ax1.twinx()
ax2.plot(derivs["time"], derivs["open_interest"], color="#059669", linewidth=1, alpha=0.8, label="Open interest")
ax2.set_ylabel("Open interest", color="#059669")
ax1.set_title("Price and open interest")
format_time_axis(ax)
plt.tight_layout()

Chart 5 — Basis in basis points

fig, ax = plt.subplots()
ax.plot(derivs["time"], derivs["basis_bps"], color="#7c3aed", linewidth=1)
ax.axhline(0, color="black", linestyle="--", linewidth=0.8)
ax.set_title("Basis (bps)")
ax.set_ylabel("bps")
format_time_axis(ax)
plt.tight_layout()

Chart 6 — Crowding score and highlighted stress regimes

fig, ax = plt.subplots(figsize=(14, 6))
ax.plot(derivs["time"], derivs["crowding_score"], color="#111827", linewidth=1.2)
ax.fill_between(
    derivs["time"],
    derivs["crowding_score"],
    where=derivs["stress_regime"],
    color="#ef4444",
    alpha=0.2,
    label="Stress regime",
)
ax.axhline(0, color="black", linestyle="--", linewidth=0.8)
ax.set_title("Composite derivatives crowding score")
ax.set_ylabel("Average z-score")
ax.legend(frameon=True)
format_time_axis(ax)
plt.tight_layout()

Chart 7 — Funding by weekday and hour

seasonality = derivs.assign(
    weekday=derivs["time"].dt.day_name().str[:3],
    hour=derivs["time"].dt.hour,
).pivot_table(index="weekday", columns="hour", values="funding_rate_bps", aggfunc="mean")
seasonality = seasonality.reindex(["Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"])

fig, ax = plt.subplots(figsize=(16, 5))
sns.heatmap(seasonality, cmap="coolwarm", center=0, ax=ax)
ax.set_title("Average funding rate by weekday and hour (bps)")
ax.set_xlabel("Hour of day")
ax.set_ylabel("")
plt.tight_layout()

Takeaways

• Funding, basis, and open interest are complementary measures of leverage, crowding, and carry.
• Persistent positive funding plus rich basis often signals an increasingly expensive long trade.
• Rising open interest during a price move can indicate fresh positioning rather than mere price drift.
• The most useful regime read usually comes from the combination of these metrics, not any one series in isolation.

Further reading

• Aperiodic derivatives market data: https://aperiodic.io/metrics/derivatives_market_data
• Funding rates: https://www.binance.com/en/academy/articles/what-are-funding-rates-in-crypto-markets
• Basis overview: https://www.cmegroup.com/education/courses/introduction-to-basis.html