Implied Volatility vs Vega: The Difference

Implied volatility vs vega comes down to one distinction that an option chain never spells out. Implied volatility, or IV, is an input: the volatility number you solve a pricing model for once the market has already set the option's price. Vega is a rate of change: the dollars the option's price moves per one point change in that input. IV tells you what volatility is being charged. Vega tells you how much of that charge a position is carrying.

Implied volatility vs vega: an input and a rate

An option pricing model takes a handful of facts anyone can look up: the spot price, the strike, the time left, and a financing rate. It also takes one fact nobody can look up, which is how much the underlying will move between now and expiry. Hand the model a volatility figure and it returns a price.

Implied volatility runs that backwards. You start from the price the market is quoting and search for the volatility figure that makes the model print exactly that price. That solved number is the IV on the chain, and how implied volatility is calculated walks through the search itself. It is not a forecast. It is the volatility assumption sitting inside today's price, and implied volatility moves whenever option prices move, even on a day the stock itself barely budges.

Vega comes out of the same model one step later. Hold everything else fixed, raise the volatility input by one point, and record how far the model price travels. That distance is vega. Take a hypothetical contract with a vega of 0.20 per share: a one point rise in IV lifts its model price by roughly 20 cents per share, about $20 on a standard 100 share contract, and two points is about $40. IV is the price of volatility. Vega is the size of the ticket. Option vega covers the mechanics in full.

Why does a longer dated option have more vega?

Volatility is uncertainty per unit of time, and a longer contract collects more of it. Two contracts on the same underlying at the identical IV can carry very different vega, and the gap widens the further out the expiry sits.

The panel below groups AAPL contracts traded through July 2026 by days to expiry, keeps the strikes within 5% of the spot price, and indexes each bucket's average vega to the nearest-dated bucket. Per-contract daily IV and greeks come from global_markets.options_greeks, filtered to rows where the volatility solve converged and the contract actually traded that day.

WITH ntm AS
(
    SELECT
        intDiv(days_to_expiry, 30) * 30 AS dte_floor,
        avg(toFloat64(vega))            AS vega_avg,
        avg(implied_volatility)         AS iv_avg,
        count()                         AS contract_days
    FROM global_markets.options_greeks
    WHERE underlying_symbol = 'AAPL'
      AND date >= '2026-07-01'
      AND date <  '2026-08-01'
      AND iv_converged = 1
      AND volume > 0
      AND days_to_expiry BETWEEN 1 AND 545
      AND abs(toFloat64(strike_price) / toFloat64(underlying_close) - 1) < 0.05
    GROUP BY dte_floor
    HAVING count() >= 50
)
SELECT
    concat(toString(dte_floor), ' to ', toString(dte_floor + 29), ' days') AS dte_bucket,
    round(iv_avg * 100, 1)                                                 AS avg_iv_pct,
    round(vega_avg / first_value(vega_avg) OVER (ORDER BY dte_floor ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING), 2) AS vega_vs_front,
    toUInt32(contract_days)                                                AS contract_day_count
FROM ntm
ORDER BY dte_floor

Averaged across the month, the 0 to 29 days bucket priced at an IV of 33.2%, against 29.5% for the 510 to 539 days bucket. Close enough to call it the same volatility being charged. The vega column across those same rows is nowhere near as flat: the longest bucket carried 8.14 times the vega of the shortest, on 14 buckets of the curve.

That multiple is the practical point. One point of IV is one point on both contracts. The money that point moves is not the same on both. Someone holding the long-dated contract holds several times the volatility exposure of someone holding the front-month contract at the same quoted IV, and that exposure changes shape as the contract ages, which option greeks over time traces.

Does vega change with the strike too?

It does, and the shape is a hump. Vega peaks near the money and falls away in both directions. A far out-of-the-money contract holds little premium to revalue, so moving the volatility assumption moves its price by less in absolute terms.

The panel below holds expiry roughly fixed, 20 to 45 days out, and walks across strikes instead, indexing each strike bucket's average vega to the biggest bucket on the curve.

WITH mny AS
(
    SELECT
        floor((toFloat64(strike_price) / toFloat64(underlying_close) - 1) * 50) / 50 AS strike_offset,
        avg(toFloat64(vega))    AS vega_avg,
        avg(implied_volatility) AS iv_avg,
        count()                 AS contract_days
    FROM global_markets.options_greeks
    WHERE underlying_symbol = 'AAPL'
      AND date >= '2026-07-01'
      AND date <  '2026-08-01'
      AND iv_converged = 1
      AND volume > 0
      AND days_to_expiry BETWEEN 20 AND 45
      AND abs(toFloat64(strike_price) / toFloat64(underlying_close) - 1) <= 0.12
    GROUP BY strike_offset
    HAVING count() >= 30
)
SELECT
    concat(toString(toInt32(round(strike_offset * 100))), '% from spot') AS strike_vs_spot,
    round(iv_avg * 100, 1)                                              AS avg_iv_pct,
    round(vega_avg / max(vega_avg) OVER (), 2)                          AS vega_vs_peak,
    toUInt32(contract_days)                                             AS contract_day_count
FROM mny
ORDER BY strike_offset

The -12% from spot bucket carried 0.46 of the peak bucket's vega, and the 10% from spot bucket 0.52. The IV column shifts across those same strikes as well, 35.5% at one end of the curve and 29.3% at the other. That tilt is volatility skew. Skew is a property of the input. The hump is a property of the sensitivity.

Is high implied volatility the same as high vega?

No. IV is a level, and a level only means something against its own history. That comparison is the job of IV rank or IV percentile, which places today's IV inside the range the same name has traded over the past year: 30% IV can sit near the bottom of one ticker's range and near the top of another's. Whether high implied volatility is good depends on which side of the contract you are holding, and vega is how the size of that side gets measured.

The panel below tracks weekly average IV for three household names at a comparable point on the curve, 20 to 45 days to expiry with strikes within 5% of spot, over the trailing year.

SELECT
    toString(toMonday(date))                                              AS week,
    formatDateTime(toMonday(date), '%b %e, %Y')                           AS week_label,
    round(avgIf(implied_volatility, underlying_symbol = 'AAPL') * 100, 1) AS aapl_iv_pct,
    round(avgIf(implied_volatility, underlying_symbol = 'NVDA') * 100, 1) AS nvda_iv_pct,
    round(avgIf(implied_volatility, underlying_symbol = 'KO') * 100, 1)   AS ko_iv_pct
FROM global_markets.options_greeks
WHERE underlying_symbol IN ('AAPL', 'NVDA', 'KO')
  AND date >= '2025-08-01'
  AND date <  '2026-08-01'
  AND iv_converged = 1
  AND volume > 0
  AND days_to_expiry BETWEEN 20 AND 45
  AND abs(toFloat64(strike_price) / toFloat64(underlying_close) - 1) < 0.05
GROUP BY week, week_label
HAVING countIf(underlying_symbol = 'AAPL') > 0
   AND countIf(underlying_symbol = 'NVDA') > 0
   AND countIf(underlying_symbol = 'KO') > 0
ORDER BY week

The series holds 53 weekly points, from the week of Jul 28, 2025 through the week of Jul 27, 2026. In that final week AAPL averaged 32%, NVDA 43.6% and KO 21.2%. Three names, three volatility levels, and not one of those numbers says how much a one point move would be worth to any particular position. That second question belongs to vega.

It is also why a position can be short volatility and close to vega neutral at once. Vega nets across legs. Sell a 30 day option, buy a smaller quantity of a 90 day option, and size the two until their vegas cancel: a parallel one point lift in IV across the whole surface then barely moves the book, while the position stays short the front expiry and long the back one. That structure is a calendar spread. Vega neutral describes the sensitivity, not the view.

IV rank answers whether volatility is expensive. Vega answers how much of that bet is on the book. The third number the same input produces, the distance the market is pricing the stock to travel by expiry, is the expected move from implied volatility.

FAQ

Is implied volatility the same as vega?

No. Implied volatility is an input to an option pricing model, solved backwards from the price the market is already quoting. Vega is that model's sensitivity to the input: the price change per one point move in IV. One is a level, the other is a rate.

Does higher implied volatility mean higher vega?

Not on its own. Vega is set mostly by time to expiry and by how far the strike sits from the spot price. A long-dated at-the-money contract can carry a multiple of the vega of a front-week contract quoted at the identical IV, which is what the first panel above measures.

What does a vega of 0.20 mean?

It means the model price of that contract moves about $0.20 per share for a one point change in implied volatility, roughly $20 on a 100 share contract. Vega is itself an estimate that shifts as spot, time and volatility shift, so the $20 holds for small moves rather than large ones.

Can a position be short volatility and vega neutral at the same time?

Yes. Vega adds and cancels across legs, so a book that is short a near expiry and long a further one can net to roughly zero vega while still holding a view on the front contract's volatility. A parallel IV move across all expiries barely touches it. A change in the shape of the term structure does.

Where do the IV and vega numbers on this page come from?

Every panel reads per-contract daily greeks from global_markets.options_greeks, which carries implied volatility, delta, gamma, vega, theta and rho for each listed contract since August 2021. The panels keep only rows where the volatility solve converged and the contract traded that session.

SELECT
    toString(toStartOfMonth(date))                AS month,
    formatDateTime(toStartOfMonth(date), '%b %Y') AS month_label,
    countIf(underlying_symbol = 'AAPL')           AS aapl_contract_day_count,
    countIf(underlying_symbol = 'NVDA')           AS nvda_contract_day_count,
    countIf(underlying_symbol = 'KO')             AS ko_contract_day_count
FROM global_markets.options_greeks
WHERE underlying_symbol IN ('AAPL', 'NVDA', 'KO')
  AND date >= '2025-08-01'
  AND date <  '2026-08-01'
  AND iv_converged = 1
  AND volume > 0
  AND days_to_expiry BETWEEN 20 AND 45
  AND abs(toFloat64(strike_price) / toFloat64(underlying_close) - 1) < 0.05
GROUP BY month, month_label
ORDER BY month

Every panel here carries the exact SQL underneath it, expand one to see how each average was counted. To run the same IV and vega comparison on a ticker you follow, ask for it in plain English on the Strasmore terminal.