IV Rank Explained: When Selling Options Pays Best
Every quote on ETFap carries an IV rank badge. It answers one question: are options expensive or cheap right now compared with the past year? Option sellers are told to sell when IV rank is high. I tested that with 35 years of VIX data and an SPX put credit spread backtest, and the answer is more interesting than “high is good.”
See today’s IV rank on a live SPX trade. The SPX Desk shows today’s put credit spread read with IV rank, POP, POT and P50, free.
Open the SPX Desk → Open Lines expected moveCovered Call & PMCC TrackerWhat IV rank is
Implied volatility (IV) is the market’s forecast of how much a stock or index will move, backed out of option prices. Higher IV means pricier options and bigger premiums for sellers. But IV levels differ a lot by underlying: 20 is calm for bitcoin and jumpy for the S&P 500. IV rank fixes that by comparing IV with its own 1-year range:
IV rank = (current IV − 1-year low) ÷ (1-year high − 1-year low) × 100
0 means IV is at its lowest of the year; 100 means its highest. ETFap’s badges use 252 trading days, with VIX for SPY and SPX, VXN for QQQ and Deribit’s bitcoin DVOL for IBIT.
IV rank vs IV percentile
IV percentile asks a different question: on what share of the past year’s days was IV lower than today? IV rank only looks at the two extremes, so one spike can squash it for a whole year. Here are the site’s badges from the Nov 2, 2026 build:
| Fund (volatility index) | IV | 1-year range | IV rank | IV percentile |
|---|---|---|---|---|
| SPY (VIX) | 21.15 | 13.47–31.05 | 44 | 87 |
| QQQ (VXN) | 30.89 | 17.20–33.54 | 84 | 97 |
| IBIT (bitcoin DVOL) | 39.55 | 33.59–82.62 | 12 | 40 |
SPY’s IV rank was a middling 44, yet VIX was higher than on 87% of the past year’s days. The 31.05 high (VIX’s 1-year peak came on Mar 27, 2026) stretched the range. Across all 9,019 days since Dec 31, 1990, VIX’s IV rank was below 50 while its IV percentile was 50 or more on 27.5% of days. That’s why each ETFap badge shows both.
How to calculate IV rank
Using Cboe’s VIX closes: on Oct 29, 2026 VIX closed at 17.20. Over the prior 252 trading days its low was 13.47 and its high 31.05. So IV rank = (17.20 − 13.47) ÷ (31.05 − 13.47) × 100 = 21, while its IV percentile was 51.
35 years of VIX: what each IV rank level meant
I computed VIX’s IV rank for every day from Dec 31, 1990 to Oct 29, 2026 and compared it with what the S&P 500 did over the next 21 trading days (Cboe daily data):
| VIX IV rank | Share of days | Avg VIX | Avg realized vol, next 21 days | VIX above realized | SPX fell past the expected move |
|---|---|---|---|---|---|
| 0–25 | 54.9% | 16.0 | 12.1 | 87.8% | 6.8% |
| 25–50 | 27.1% | 21.0 | 16.9 | 83.4% | 9.2% |
| 50–75 | 12.4% | 25.3 | 21.3 | 77.2% | 7.3% |
| 75–100 | 5.7% | 30.1 | 26.7 | 82.6% | 4.9% |
Three takeaways. First, IV rank is low most of the time: below 25 on 55% of days, above 75 on only 6%. Second, VIX ran above the next month’s realized volatility at every level, by about 3.4 to 4.1 points on average. That volatility risk premium is the seller’s edge (more in selling outside the expected move). Third, the edge didn’t grow with IV rank: high IV rank buys bigger premium, but also bigger moves.
SPX put credit spreads by IV rank: the backtest
Next I grouped all 726 trades of the SPX Desk’s model backtest (Jun 2, 2009 to Oct 23, 2026; a 7-day, 25-point SPX put credit spread near 9 delta, entered only above the 200-day average, with a stop) by VIX’s IV rank on the entry day:
| VIX IV rank at entry | Trades | Win rate | Avg credit | Avg result | Avg % of risk |
|---|---|---|---|---|---|
| 0–25 | 411 | 96.8% | $123 | $65 | +2.7% |
| 25–50 | 226 | 97.3% | $124 | $80 | +3.4% |
| 50–75 | 62 | 88.7% | $125 | −$107 | -4.5% |
| 75–100 | 27 | 88.9% | $129 | −$25 | -1.1% |
In this model, the best results came at low to middle IV rank. Trades at IV rank 50–75 won 89% of the time and averaged −$107 a spread, against 97% and $65 below 25. The credit barely changed ($123 vs $125), because the strategy picks strikes by delta: higher IV pushes the short strike further away rather than paying more. Grouped by IV percentile, the pattern is similar: the top half still made money, but much less than the bottom half.
| VIX IV percentile at entry | Trades | Win rate | Avg credit | Avg result | Avg % of risk |
|---|---|---|---|---|---|
| 0–25 | 173 | 98.8% | $123 | $97 | +4.1% |
| 25–50 | 137 | 99.3% | $123 | $109 | +4.6% |
| 50–75 | 185 | 93.0% | $123 | $2 | +0.1% |
| 75–100 | 231 | 94.4% | $125 | $22 | +0.9% |
Be careful with the high buckets: only 89 trades, so a few losers swing the average. This is a model priced from VIX9D, not real fills. Still, it doesn’t support “the higher the IV rank, the better.” High IV rank tends to show up right after a shock, when the market can keep moving.
How I use IV rank
- Context, not a trigger. IV rank tells me whether premium is rich or thin. The trade still has to work on the expected move, trend and size.
- Read rank and percentile together. Low rank with high percentile usually means an old spike is distorting the range.
- High IV rank, smaller size. Fat premium comes with bigger moves, so I size down and stay further out, the same thinking as my bankroll rules.
- Low IV rank, think about who’s paying. Thin premium makes covered calls and spreads less worth it; long-dated buyers like the Slingshot Trade and PMCC long calls get cheaper.
- Mind the calendar. IV often rises into scheduled events like FOMC days and drops after, which IV rank alone won’t tell you.
A trade example with the badge
From the Oct 30, 2026 SPX chain, with SPX near 7,639 and a 149-point expected move to Nov 5, 2026. This is an example, not a recommendation:
| SPX put credit spread (example) | Credit | Max loss | POP | POT | P50 |
|---|---|---|---|---|---|
| Sell 7525 / buy 7500 put, Nov 5, 2026 expiry | $4.25 | $20.75 | 85% | 29% | 89% |
POP is the model chance of keeping the full credit, POT the chance SPX touches the short strike first, and P50 the chance of closing at half the credit (POP vs POT vs P50 explained). IV rank sets the backdrop; these numbers decide the trade.
Sources and method. Cboe daily closes for _VIX and _SPX (cdn.cboe.com delayed quotes history), Dec 31, 1990 to Oct 29, 2026. IV rank and IV percentile use the trailing 252 trading days including the current day. Realized volatility = standard deviation of the next 21 daily log returns × √252; “fell past the expected move” = 21-day SPX return below −VIX × √(21/252). Backtest: the SPX Desk’s model (prices modeled from VIX9D, fees and slippage included, no historical option quotes), grouped by VIX IV rank on the entry date; results per one spread. Badge values: ETFap’s IV rank feed (Cboe VIX/VXN and Deribit DVOL closes). Option example: Cboe delayed quotes, bid/ask mids; POP, POT and P50 are model estimates.
FAQ
What is IV rank?
IV rank shows where implied volatility sits in its range over the past year: (current IV − 1-year low) ÷ (1-year high − 1-year low) × 100. An IV rank of 0 means IV is at its 1-year low; 100 means it is at its 1-year high.
What is the difference between IV rank and IV percentile?
IV rank uses only the high and low of the past year. IV percentile is the share of the past year’s trading days when IV closed below today’s level. One spike can keep IV rank low for months while IV percentile is high.
Is a high IV rank good for selling options?
High IV rank means richer premiums, but not automatically better results. In our SPX put credit spread model backtest, trades entered when VIX’s IV rank was 0–50 won 97–97% of the time, while trades above 50 won about 89% and averaged a loss. High IV rank often means the market is still moving.
What is a good IV rank to sell options?
There is no magic number. Many traders like IV rank above 30 or 50 for selling premium, but the data here shows that price trend, strike distance and size matter more. I use IV rank as context next to the expected move, not as a trigger.
How do you calculate IV rank?
Take today’s implied volatility, subtract the lowest close of the past 252 trading days, and divide by the 1-year high minus the low. Example: VIX closed at 17.20 on Oct 29, 2026, with a 1-year range of 13.47 to 31.05, so IV rank was 21.
Why does ETFap use VIX for SPY’s IV rank?
VIX is Cboe’s 30-day implied volatility of S&P 500 options, and SPY tracks the S&P 500. Free daily history of SPY’s own implied volatility isn’t available, so VIX is the stand-in; QQQ uses VXN and IBIT uses Deribit’s bitcoin DVOL.
Does implied volatility usually overstate the real move?
Usually. From Dec 31, 1990 to Oct 29, 2026, VIX was above the S&P 500’s realized volatility over the next month on most days at every IV rank level. That gap, the volatility risk premium, is the option seller’s edge.
Is this financial advice?
No. It is education from my own process. Options involve substantial risk, and backtests are hypothetical.
Check IV rank and the odds before you sell. The SPX Desk shows today’s put credit spread read with IV rank, POP, POT and P50, free.
Open the SPX Desk → Open Lines expected moveCovered Call & PMCC Tracker