IV/HV Ratio Screener Data for Volatility Research

Use IV/HV ratio, spread, cross-sectional rank, and z-score together, then require settled values and event context before interpreting a volatility premium as rich or cheap.

Alphanume's IV/HV Premium dataset pairs approximately 30-day at-the-money implied volatility with 30-trading-day close-to-close realized volatility for liquid optionable equities. It derives an implied-minus-realized spread, implied-divided-by-realized ratio, daily ranks, z-scores, and a notional-volume reference.

The ratio normalizes the gap by realized volatility, while the spread preserves the absolute volatility-point difference. They can rank the same ticker differently. A high ratio can come from high IV, unusually low HV, or both, and the reason matters before any trade hypothesis is written.

Know the four comparison lenses

iv_hv_ratio > 1 means implied volatility exceeds the realized measure, while a value below one reverses that relationship. It does not prove options are mispriced because realized volatility looks backward and implied volatility can contain information about future event risk.

Inspect both numerators and denominators

Two stocks can share a ratio of two with very different economics. IV of 20% over HV of 10% has a 10-point spread, while IV of 80% over HV of 40% has a 40-point spread. Add the raw levels, spread, and days to expiration to every ratio screen.

Cross-sectional rank controls for the daily universe but not for each ticker's own history. If the question is whether a name is elevated relative to itself, add IV/HV Rank or a separate historical distribution. Do not treat daily ratio rank as a 52-week percentile.

Query the richest settled decile

The endpoint supports exact dates, date ranges, ratio bounds, and min_ratio_rank. Set only_final=true to exclude provisional intraday rows from an end-of-day screen.

import os
import requests

response = requests.get(
    "https://api.alphanume.com/v1/iv-hv-premium",
    headers={"X-API-Key": os.environ["ALPHANUME_API_KEY"]},
    params={
        "date": "2026-06-12",
        "min_ratio_rank": 0.90,
        "only_final": "true",
    },
    timeout=30,
)
response.raise_for_status()
rows = response.json()["data"]

assert all(row["is_final"] == 1 for row in rows)

A 0.90 floor returns the daily top decile under the ratio rank. Save the full candidate response rather than only ticker and ratio so days to expiration, spot, ATM strike, notional volume, spread, and z-score remain auditable.

Handle intraday and settlement timing

Today updates provisionally around every 30 minutes during the session with is_final=0 and settles around 4:30 PM New York time with is_final=1. Past rows are final. A settled same-day screen therefore supports next-session execution, not a trade assumed before the regular close.

If the strategy acts intraday, archive the exact provisional response and use last_updated as part of the decision record. Historical final rows cannot recreate which provisional value was visible at an earlier timestamp unless those snapshots were separately saved.

For historical evaluation, form the candidate list from each settled date and delay any assumed entry until the next eligible quote. Select the ratio threshold, spread threshold, and liquidity floor on an earlier development window. Measure turnover as well as average outcomes, since a screen that changes names daily can lose an apparent gross edge through spreads and commissions. Report days with few or no qualifying rows instead of loosening the rule after the fact.

Add event and liquidity gates

• Earnings. A rich ratio can be rational ahead of a scheduled announcement.
• Corporate events. FDA, financing, merger, and legal catalysts can create jump risk.
• Liquidity. Underlying notional volume is context and does not replace option spreads or depth.
• Term structure. One approximately 30-day ATM point does not describe every expiration.
• Skew. A single ATM measure cannot price wing risk or directional asymmetry.

Free access provides a trailing 20-trading-session window ending one session behind the latest observation. The proof page provides product evidence and does not establish that selling the richest decile is profitable after costs and tail losses.

Run one multi-lens screen

Read the IV/HV Premium field reference, pull one settled date, and export the richest ratio decile with raw IV, HV, spread, ratio, rank, z-score, notional volume, expiration, and event flags. Compare rankings by ratio and spread and inspect the largest disagreements.

Then define one contract-level hypothesis and test it with historical bid-ask data, fixed entry and exit times, commissions, and a later holdout. Preserve candidates removed for events or liquidity so the exclusions remain part of the result rather than disappearing after the screen.