Risk & VolatilityUpdated daily after close · as of 2026-07-01
CBOE SKEW Index: Tail-Risk Pricing, Live
What the options market charges for crash protection — the CBOE SKEW index from deep out-of-the-money S&P options, daily since 1990, scored against its own trailing year. And the part most SKEW charts skip: the per-era test of whether it predicts returns. (Spoiler: the sign flips by decade — it describes hedging demand, not the future.)
Today's reading
As of the July 1, 2026 close, the CBOE SKEW index sits at 154.8 — extreme tail-risk pricing, +1.6σ versus its trailing year (94th percentile of the past year, 98th of all readings since 1990). Deep out-of-the-money S&P puts are expensive relative to the rest of the surface. The all-time record is 183.1 (2025-02); the long-run average is 123. Historically, SKEW extremes have NOT been a reliable directional signal — the per-era study on this page shows the relationship flipping sign by decade.
Source
CBOE SKEW index ($SKEW.X) daily closes from TradeStation (1990–present); SPY closes for the forward-return studies
Methodology
State = rolling 252-session z-score of the SKEW close (Complacent ≤ −1.5σ … Extreme ≥ +1.5σ); per-era forward SPY return tables at 5/21/63 sessions
Updates
Daily after US market close (~1pm PT)Last: 2026-07-01
Maintained & reviewed by Yuriy Matso — methodology shown on the page.
SKEW index$SKEW · 2026-07-01 · close
EXTREME
154.8
z +1.6 vs the trailing year · 94th pctile (1y) · 98th all-time
All-time avg
123
Record high
183
2025-02
Record low
101
1991-03
Deep out-of-the-money S&P puts are expensive relative to the surface. A pricing fact, not a forecast — the per-era study below shows why.
01
SKEW vs its own era
The daily SKEW close (purple) against its rolling one-year mean and ±1.5σ band (gray). The band IS the era adjustment — SKEW averaged ~115 in the 1990s and sits in the 140s–150s today as tail hedging became institutional, so raw levels across decades can't be compared directly.
Window:loading…
Loading…
SPY price (top, log)CBOE SKEW (daily close)Rolling 252-session mean±1.5σ band (state edges)
02
Does SKEW predict returns? The per-era answer
Forward SPY returns after extreme-skew and complacent sessions, split by era — because the full-history average hides a sign flip. Extreme tail-risk pricing beat its era baseline in 1994–2004, lagged it sharply in 2005–2013, and lagged mildly since 2014. A relationship that flips sign by decade is a description of hedging demand, not a signal — anyone selling a SKEW-based system should show you this table.
1994-2004
| State | Next 5 sessions | Next 21 sessions | Next 63 sessions | N |
|---|---|---|---|---|
| Extreme tail-risk pricing (z ≥ +1.5) | +0.26% · 57%↑ | +0.95% · 61%↑ | +4.06% · 72%↑ | 305 |
| Complacent (z ≤ −1.5) | +0.74% · 61%↑ | +2.59% · 74%↑ | +5.99% · 85%↑ | 286 |
| All sessions in era (baseline) | +0.20% · 56%↑ | +0.81% · 61%↑ | +2.43% · 66%↑ | 2,760 |
2005-2013
| State | Next 5 sessions | Next 21 sessions | Next 63 sessions | N |
|---|---|---|---|---|
| Extreme tail-risk pricing (z ≥ +1.5) | -0.42% · 52%↑ | -1.43% · 48%↑ | +0.22% · 61%↑ | 245 |
| Complacent (z ≤ −1.5) | +0.34% · 56%↑ | +0.33% · 60%↑ | +1.33% · 60%↑ | 115 |
| All sessions in era (baseline) | +0.13% · 57%↑ | +0.53% · 64%↑ | +1.60% · 66%↑ | 2,260 |
2014-present
| State | Next 5 sessions | Next 21 sessions | Next 63 sessions | N |
|---|---|---|---|---|
| Extreme tail-risk pricing (z ≥ +1.5) | +0.14% · 61%↑ | +0.97% · 67%↑ | +2.37% · 72%↑ | 427 |
| Complacent (z ≤ −1.5) | +0.17% · 58%↑ | +0.88% · 57%↑ | +1.84% · 59%↑ | 214 |
| All sessions in era (baseline) | +0.27% · 61%↑ | +1.11% · 67%↑ | +3.25% · 75%↑ | 2,944 |
States use the rolling one-year z-score, so each era's “extreme” is measured against its own regime. Forward returns on SPY closes; overlapping windows. We publish this table so the conclusion can be checked, not believed.
How SKEW Index Works
- 1What SKEW measuresCBOE computes SKEW from deep out-of-the-money S&P 500 options — essentially the price of crash protection relative to everything else. At 100, option prices imply a textbook lognormal world with no tail premium; the higher SKEW goes, the more investors are paying up for far-below-market puts.
- 2Why raw levels misleadSKEW's baseline has drifted structurally: it averaged around 115 through the 1990s and sits in the 140s-150s in the modern era, as institutions permanently hedge tails. A reading of 140 was extreme in 1995 and unremarkable in 2025 — raw levels across decades are not comparable.
- 3The era-proof stateWe score each day against its own trailing year (a rolling 252-session z-score): Complacent (z ≤ −1.5), Subdued, Normal, Elevated, and Extreme (z ≥ +1.5). That is the read the hero card shows, alongside the raw level and its percentiles.
- 4The myth checkThe study section reports forward SPY returns after extreme-skew and complacent days separately for three eras — because the relationship flips sign across decades. We publish the per-era tables instead of a cherry-picked full-history average, and the conclusion they support: SKEW describes hedging demand; it has not been a reliable timing signal in either direction.
Who Uses SKEW Index
Risk Managers
A clean read on what the market charges for tail protection right now, normalized against its own recent regime — useful as context for hedging costs even though it isn't a timing signal.
Options Traders
Elevated SKEW means deep OTM puts are rich relative to the surface; complacent readings mean crash protection is historically cheap to own. That is a pricing fact, independent of any directional forecast.
Skeptics
Every few months a viral chart claims SKEW "predicted" something. The per-era tables here are the antidote: high skew beat its era baseline in 1994-2004, lagged sharply in 2005-2013, and lagged mildly since 2014.
Macro Watchers
Sustained regime shifts in SKEW (the rolling mean, not the daily print) track the institutionalization of tail hedging — a structural story worth knowing even without a signal.
Pro Tips
01
Treat it as a price, not a prophecy
SKEW tells you what crash insurance costs today. Like most insurance prices, it reflects demand more than clairvoyance — the record high (183, February 2025) did not mark a top, and record lows have not marked bottoms.
02
The z-score is the read
Because the baseline drifts, compare today only to the trailing year. Our state bands do that automatically; if you take one method from this page, take that one.
03
Beware single-era backtests
Any SKEW study that doesn't split by era is implicitly betting you won't check. The 2005-2013 window makes high skew look scary; 1994-2004 makes it look bullish. Both are true — which is the point.
04
Pair it with VIX, don't substitute
VIX prices at-the-money volatility (everyday turbulence); SKEW prices the far tail (crashes). They can disagree — a calm VIX with extreme SKEW, like mid-2026, means markets expect quiet trading but are paying up against a jump.
Common Issues & Solutions
Is a high SKEW reading bearish?▾
History says it is not a reliable signal. Relative to each era's own baseline, extreme-skew days led to better-than-average returns in 1994-2004, much worse in 2005-2013, and slightly worse since 2014. A sign that flips by decade is not a signal you can trade.
Why does SKEW sit so much higher than it used to?▾
Structural demand for tail hedges. After 2008 (and again after the 2010s vol events), systematic tail-hedging programs became permanent buyers of deep OTM puts, lifting SKEW's baseline from the ~115 of the 1990s to the 140s-150s today. That is exactly why we score it against a rolling window.
SKEW vs VIX — what's the difference?▾
VIX measures the implied volatility of near-the-money S&P options over the next 30 days — the expected size of ordinary moves. SKEW measures the asymmetry priced into deep out-of-the-money options — how much extra a crash costs to insure against. Calm VIX and elevated SKEW can coexist.
Where does the data come from?▾
The CBOE SKEW index ($SKEW.X), daily closes since January 1990 from our TradeStation feed, joined with SPY closes from our own price database for the forward-return studies.
Frequently Asked Questions
What is the CBOE SKEW index?▾
A CBOE index computed from deep out-of-the-money S&P 500 options that measures the perceived risk of extreme negative moves — tail risk. A value of 100 means option prices imply no tail-risk premium; higher values mean investors are paying progressively more for crash protection. It has ranged from about 101 (1991) to 183 (February 2025).
Is a high SKEW index bearish for stocks?▾
The data says it has not been a reliable signal in either direction. Measured against each era's own baseline, days with extreme skew (1.5σ above the trailing year) preceded above-average SPY returns in 1994-2004, sharply below-average in 2005-2013, and mildly below-average since 2014. A relationship that flips sign by decade describes hedging demand, not the future.
What does a SKEW reading above 150 mean?▾
Deep out-of-the-money puts are very expensive relative to at-the-money options — the market is paying up heavily for crash insurance. In the modern era that is roughly the top decile of readings. It tells you protection is dear; it does not tell you a crash is coming (or not coming).
What is the difference between SKEW and VIX?▾
VIX prices the expected size of ordinary 30-day moves from near-the-money options; SKEW prices the asymmetry of the far tail from deep out-of-the-money options. They answer different questions and frequently diverge — a calm VIX alongside an extreme SKEW means quiet expected trading but expensive jump insurance.
How often is this updated?▾
Daily after the US close, from CBOE's official SKEW index values. The z-score, state and percentiles recompute against the full 1990-present history.
Explore Other Tools
Last updated: 2026-07-01