Tell us how you trade and we'll point you to the right research segment.
Talk to Our Team →Start with our beginner-friendly guides on market basics, order types, and risk management before you place your first trade.
Start Learning → Browse All Articles →IV rank and IV percentile are two related but distinct ways of putting a raw implied volatility figure into context against its own recent history, rather than looking at that day’s implied volatility number in isolation. A raw implied volatility reading on its own says very little, since what counts as high or low for a given underlying varies enormously depending on that instrument’s own typical volatility behaviour — these two metrics exist specifically to answer the more useful question of whether today’s implied volatility is high or low relative to where it has actually been over a defined recent period. This piece works through how each metric is calculated, why they can diverge from each other even when using the same underlying data, and how they are actually used to think about option strategy selection.
Implied volatility varies enormously across different underlyings, and even for the same underlying it varies across different periods depending on prevailing market conditions. A given implied volatility reading that would be considered unusually high for one instrument might be entirely ordinary for another, simply because the two instruments carry structurally different typical volatility levels.
This makes a single day’s raw implied volatility figure a poor basis for deciding whether option premiums are currently expensive or cheap in a meaningful sense. What is actually needed is a way to compare today’s reading against that same instrument’s own recent history, which is exactly the gap that IV rank and IV percentile are designed to fill.
The underlying intuition behind both metrics is straightforward even before working through the calculations: volatility, unlike a price trend, tends to be mean-reverting over time rather than persistently trending in one direction. A period of unusually calm price action is often followed, sooner or later, by a period of higher volatility, and the reverse holds too. Because of this tendency, knowing where current volatility sits relative to its own recent range carries more useful information than the raw number alone, since it hints at how much room there is for volatility to revert toward its more typical level.
IV rank measures where today’s implied volatility sits between the highest and lowest implied volatility readings recorded over a defined lookback period, typically the trailing year. It is calculated as the difference between today’s reading and the period’s low, divided by the difference between the period’s high and low, expressed as a position within that range.
A reading toward the higher end of this range indicates implied volatility is currently close to its highest point over the lookback period, while a reading toward the lower end indicates it is currently close to its lowest point over that same period. The calculation only requires the highest and lowest values from the lookback period along with today’s reading — it does not use every individual daily reading in between.
This simplicity is both the strength and the weakness of IV rank as a metric. It is quick to calculate and easy to interpret at a glance, since it always expresses today’s reading as a position within a clearly bounded range. The trade-off is that it discards a considerable amount of information about how implied volatility actually behaved throughout the rest of the lookback period, retaining only the two extreme values as reference points against which today’s reading is measured.
IV percentile works differently. Rather than measuring where today’s reading sits between the period’s high and low, it measures what proportion of all the individual daily readings over the lookback period were below today’s current reading. This means every single day’s implied volatility figure within the lookback period actually contributes to the calculation, not just the extreme high and low values.
Because IV rank only anchors on the two extreme values from the lookback period, it can be disproportionately influenced by a single unusual spike or an unusually calm stretch that happened once during that period, even if implied volatility spent the vast majority of the period somewhere in the middle of that range. IV percentile is less sensitive to a single outlier reading, since it reflects the distribution of the entire period’s data rather than only its two extremes.
Consider a concrete illustration of this behaviour. Suppose implied volatility for a given underlying spent almost the entire lookback period trading within a fairly narrow, moderate range, but spiked briefly to an extreme level around a single unusual event before settling back down. IV rank, calculated against that period’s high and low, would show today’s moderate reading as sitting quite low within the full range, because the brief spike stretched the top of that range far above where volatility actually spent most of its time. IV percentile, by contrast, would show today’s reading as sitting roughly in the middle, because the vast majority of individual daily readings across the period were also clustered around that same moderate level. Both are technically accurate calculations of what they are designed to measure, but they tell a noticeably different story about the same underlying data.
It is entirely possible for IV rank and IV percentile to show quite different readings for the same underlying on the same day, calculated from the exact same underlying data set, purely because of the difference in how each metric is constructed. If implied volatility spiked briefly to an unusually high level once during the lookback period and has otherwise stayed low, IV rank can show a relatively low reading today even while IV percentile also shows a low reading — but the gap between the two can widen considerably in cases where the distribution of readings across the period is more skewed.
Understanding why the two can diverge is more useful than memorising a rule for which one to prefer. A trader who only looks at IV rank without understanding its sensitivity to extreme values can be misled by a single historical spike into believing current implied volatility is lower, relative to its own history, than the fuller picture from IV percentile would suggest.
A commonly cited principle among option traders is that strategies benefiting from a decline in implied volatility, such as selling premium through credit spreads or similar structures, are generally considered more attractive when IV rank or IV percentile is elevated, since implied volatility sitting near the high end of its recent range has more room to fall back toward its typical level than to keep rising indefinitely.
Conversely, strategies that benefit from a rise in implied volatility, such as buying options outright, are generally considered more attractively priced when IV rank or IV percentile is low, since option premiums embed relatively little volatility expectation at that point, and a wider range of subsequent volatility outcomes work in the buyer’s favour compared with buying options when implied volatility is already elevated.
These are general tendencies worth understanding rather than mechanical rules to be applied without further thought. A low IV rank reading does not guarantee implied volatility will rise from current levels, and a high IV rank reading does not guarantee it will fall — the metrics describe where volatility sits relative to its own recent history, not a forecast of where it is headed next.
Both IV rank and IV percentile depend entirely on the lookback period chosen for the calculation, and a different lookback period can produce a meaningfully different reading for the exact same underlying on the exact same day. A trailing year is a commonly used default, but a shorter lookback period will be more sensitive to recent conditions, while a longer lookback period will smooth out shorter-term fluctuations in favour of a broader historical context.
Neither choice of lookback period is inherently correct — the appropriate period depends on the trading timeframe and strategy being considered. A trader focused on short-dated option strategies may find a shorter lookback period more relevant to their actual decision, while a trader thinking in terms of longer holding periods may prefer the fuller context a longer lookback period provides.
Whatever period is chosen, it is worth applying it consistently across the underlyings being compared, since comparing an IV rank calculated over one lookback period for one underlying against an IV rank calculated over a different lookback period for another underlying produces a comparison that looks consistent on the surface but is not actually measuring the same thing.
Both metrics are backward-looking by construction, describing where implied volatility has been relative to today’s reading rather than predicting where it is going next. Treating either metric as a forecasting tool on its own, rather than as one input alongside a broader view of the underlying and prevailing market conditions, risks placing more weight on the reading than it can actually support.
It is also worth remembering that these metrics are calculated from an underlying’s own historical implied volatility data, which means a relatively new listing or an underlying that has recently gone through a structural change in its typical volatility behaviour may not have a long enough or representative enough history for either metric to be particularly meaningful yet.
Finally, both metrics look only at implied volatility’s own historical range and say nothing directly about how implied volatility compares with actual realised volatility over the same period. A trader deciding whether option premiums genuinely look rich or cheap in an absolute sense will usually want to bring realised volatility into the picture as well, rather than relying on IV rank or IV percentile in isolation to make that judgment.
No. IV rank measures where today’s implied volatility sits between the highest and lowest readings over a lookback period, while IV percentile measures what proportion of all daily readings over that period fall below today’s reading. The two can diverge meaningfully depending on the distribution of historical readings.
Neither is inherently more reliable — they measure slightly different things. IV percentile is generally less sensitive to a single outlier reading in the lookback period, since it uses the full distribution of daily data rather than only the two extreme values IV rank relies on.
No. A low IV rank or IV percentile reading indicates implied volatility is currently near the low end of its recent range, but it does not guarantee a rise — it describes relative positioning within recent history, not a forecast of future direction.
A trailing year is a commonly used default, but there is no single correct period — a shorter lookback period reflects more recent conditions, while a longer one provides broader historical context, and the appropriate choice depends on the trading timeframe being considered.
They require a sufficiently long and representative history of implied volatility data to be meaningful. A recently listed underlying, or one that has recently undergone a structural shift in its typical volatility behaviour, may not yet have a history long enough for either metric to be particularly informative.
Explore our Options Tips Provider service or get in touch with our research team.