Data to 5 October 2026
Sections

Dashboards

How much speculation is there?

Fear and greed: the Indian market's mood in one number

The index reads 31 out of 100 on 5 Oct 2026: fear. Lower than 76% of readings since 2007. It sums up six parts of the tape. Our tests found it does not forecast volatility or falls beyond what recent volatility already says. Its one useful region is extreme fear, which has been followed by better three months than usual.

Written for investors who want the market's mood with its parts on show, and want to know what such readings have meant. Data to 5 Oct 2026.

Today's reading and its parts

Each part is placed against its own history: the percentile is the share of past readings it is above, using only data up to each day. The score is their average. Today the biggest drag is safe-haven demand and the biggest lift is risk appetite.

31of 100Fear

Its 24th percentile since 2007. All six parts are in the reading.

0100

A week ago 34A month ago 58A year ago 49

ComponentTodayIn its own historyAdds to the score
MomentumNifty 500 against its 125-session average −3.9% from its average 20th percentile, since 2001 −5.1
Price strength52-week highs minus lows, share of liquid stocks −1.5 pts net highs 16th percentile, since 2007 −5.7
ParticipationShare of liquid stocks above their 50-day average 30% above 22nd percentile, since 2007 −4.6
VolatilityLast month's volatility against the last year's; high reads as fear 0.84× the year's 51st percentile, since 2002 +0.2
Safe-haven demandStocks minus 5-year G-secs, last 20 sessions −4.5 pts 15th percentile, since 2004 −5.8
Risk appetiteSmall caps minus large caps, last 20 sessions +1.2 pts 60th percentile, since 2008 +1.6

"Adds to the score" is the part's percentile minus the midpoint, divided by six: the parts' additions sum to the score minus 50.

The reading since 2007

The lowest reading was 0.4 on 25 Mar 2020; the highest 94.6 on 7 Dec 2020.

Fear and greed, weekly

Friday readings, 0 to 100, with the 13-week average in bold. Lines at 25 and 75 mark the extreme bands. Shaded: spells when the Nifty 50 was 20% or more below its peak, with the year and the deepest fall.

How to read it. Compare where the line sat before each big fall with where it sat after. The tests below ask whether the reading before a fall says anything that recent volatility does not.

Loading chart…

Source: NSE total-return indices (Nifty 500, Nifty 100, Smallcap 250, 5-year G-sec); NSE bhavcopy breadth of the liquid universe.
Method JSON

The last two years, daily

Every session, 0 to 100.

How to read it. Day to day the reading is noisy. Read the level and its direction over weeks.

Loading chart…

Source: NSE total-return indices (Nifty 500, Nifty 100, Smallcap 250, 5-year G-sec); NSE bhavcopy breadth of the liquid universe.
Method JSON

What happened next

Nifty 500 total returns after every session in each band, with the number of separate spells (windows that do not overlap), because thousands of overlapping sessions can look like far more evidence than they are. Read the spells column before the returns.

After each band: the next 3 and 12 months

Nifty 500 total return after every session in the band, 2007 onward. Today's band, fear, is in bold.

How to read it. Each dot is the median Nifty 500 return after sessions in that band, and the bar is the middle half of those returns. The shaded column is all sessions. The number on the right counts separate spells that do not overlap: thousands of sessions are only this many independent pieces of evidence. A band whose bar sits over the shaded column has said little; hollow rows have under five spells.

After each band: the next 3 and 12 months. 2 horizons, 5 bands. Extreme fear: median +2.9%, 33 spells; Fear: median +2.2%, 62 spells; Neutral: median +2.4%, 57 spells; Greed: median +3.9%, 60 spells; Extreme greed: median +7.9%, 29 spells. All sessions: median +3.7%.

Method
Every column in a table: share up and the worst fall too
Nifty 500 total return after each session in the band, 2007 onward. "Share up" is the share of sessions followed by a positive three-month return.
ReadingSessionsSeparate 3-month spellsNext 3 months, medianNext 3 months, middle half3 months: share upWorst fall within 3 months, medianNext 12 months, medianMore
Extreme fear82533+2.9%−3% to +11%64%−5.0%+11.7%
Fear1,30562+2.2%−5% to +8%57%−5.2%+7.9%
Neutral70257+2.4%−4% to +8%57%−3.6%+6.2%
Greed1,51960+3.9%−2% to +9%68%−3.0%+11.2%
Extreme greed46729+7.9%+4% to +12%86%−2.4%+17.5%
All sessions4,81877+3.7%−3% to +9%65%−3.8%+10.1%

Does it forecast anything?

Five questions, written down before any result was computed. Risk is where real predictability exists, so the risk tests use hard baselines: recent volatility over a week, a month and a quarter, the US VIX, and whether the index is below its 200-day average.

  1. FG-H1 Does the score say anything about next month's volatility that recent volatility, the US VIX and the 200-day rule do not? Failed No. p = 0.74, and adding the score made out-of-sample forecasts slightly worse (2016 onward).
    p-value, score added to the volatility forecast bar 0.01 10.10.010.0010.0001 0.74

    p-values on a log scale; smaller is stronger evidence, to the right. Shaded: the side that passes the bar set in advance.

    • Same sign in both halves: met
    • Improves out-of-sample forecasts: not met
  2. FG-H2 Does it add to the odds of a 5% fall within a month? Failed Close, but no. Lower scores did come before more 5% falls, in both halves of the sample and out of sample, but p = 0.03 against a pre-set bar of 0.01.
    p-value, score added to the odds of a 5% fall bar 0.01 10.10.010.0010.0001 0.029

    p-values on a log scale; smaller is stronger evidence, to the right. Shaded: the side that passes the bar set in advance.

    • Same sign in both halves: met
    • Improves out-of-sample forecasts: met
  3. FG-H3 Does it add to the odds of a 10% fall within three months? Failed No. p = 0.29; out of sample it made forecasts worse.
    p-value, score added to the odds of a 10% fall bar 0.01 10.10.010.0010.0001 0.29

    p-values on a log scale; smaller is stronger evidence, to the right. Shaded: the side that passes the bar set in advance.

    • Same sign in both halves: met
    • Improves out-of-sample forecasts: not met
  4. FG-H4a Has extreme fear (the score's lowest tenth) been followed by better three months than usual? Passed Yes. The Nifty 500 returned +8.1% on average over the next three months, against +3.5% for all sessions, across 14 separate episodes since 2011; p = 0.006, and the gap was positive in both halves.
    p-value, zone against all sessions bar 0.01 10.10.010.0010.0001 0.0062
    Next 3 months, average return all sessions +3.5% +8.1%

    p-values on a log scale; smaller is stronger evidence, to the right. Shaded: the side that passes the bar set in advance.

    • Same sign in both halves: met
  5. FG-H4b Has extreme greed (the highest tenth) been followed by worse three months? Failed No. It was followed by better than usual returns too (+7.6% against +3.5%), so the warning reading of greed is not supported.
    p-value, zone against all sessions bar 0.01 10.10.010.0010.0001 0.99
    Next 3 months, average return all sessions +3.5% +7.6%

    p-values on a log scale; smaller is stronger evidence, to the right. Shaded: the side that passes the bar set in advance.

    • Same sign in both halves: not met

Tests, samples and pass bars were fixed in the pre-registered specification before any result was computed. A pass needs every condition written there, not only a small p-value.

What this means. Use the score as a description of the day's tape, with its parts in view. Do not read greed as a warning. Extreme fear has been the one region with a record, and it has pointed towards better returns, not worse. It is one finding on 14 episodes, and the gap between halves of the sample was +3.8 and +5.2 points.

How it is built

Six parts of the tape, each turned into a percentile of its own history so that a 5% move in one and a 5-point move in another can be averaged. Equal weights, fixed in advance; nothing is fitted.

The parts. Momentum (Nifty 500 against its 125-session average), price strength (52-week highs minus lows), participation (share of liquid stocks above their 50-day average), volatility (the last month's against the last year's, read inversely), safe-haven demand (stocks against 5-year G-secs over 20 sessions) and risk appetite (small caps against large caps over 20 sessions). Stock counts use the liquid, survivorship-free NSE universe from breadth.

Against CNN's index. CNN's Fear & Greed Index uses seven parts for the US market. We keep the idea and use what India's data can support honestly. There is no put/call ratio: NSE option data here starts in 2020, and in India the ratio's sign is disputed because option writers dominate open interest. Junk-bond spreads are left out because corporate bond yields are licensed. The US VIX is left out so that it can serve as a baseline in the tests. India VIX is not in our data yet.

Against the earlier attempt. A previous fear and greed index for India was tested on 2017–2024 and failed as a forecast of returns and of risk. This one is built afresh on longer data. From 2008 it covers several full cycles and about 4.5 thousand sessions, more than twice the earlier sample. Its risk tests use harder baselines, and it tests the tails separately from the whole range.

No look-ahead. Every percentile uses only data up to that day, and each part needs three years of its own history before it counts. The reading is published once four of the six parts exist (from 2007); the tests use only days with all six (from 2008).

The workings

Method

Data

Every chart has a table view, a CSV download and a link to its published JSON. Bundles are listed in data status.

Read with