Prolepsia forecastYes / no

Will the Nasdaq-100 index close at least 20% below its previous all-time closing high on any trading day between 1 January and 31 December 2027?

As of 17 November 2026, Prolepsia puts the chance of yes at 64%.

64%chance of yes

Yes 64%No 36%
Prolepsia forecast made

How Prolepsia reasons

The report Prolepsia wrote with this forecast.

How forecasts are made · Track record

TL;DR

The forecast is 63.7% Yes and 36.3% No. A qualifying 2027 close is more likely than not because the Nasdaq-100 is already well below its record and volatility is elevated; this is not a prediction that it will finish 2027 in a bear market.

Key evidence

  • The existing threshold is closer than a fresh 20% decline. The corroborated November 16, 2026 close is 27,197.15, rather than the description’s 27,276.26. The previous record close was 31,224.69 on October 6, making the current qualifying threshold 24,979.752 index points. That requires a further 8.1531% decline from the latest completed close, absent a new record (Nasdaq; FRED).
  • Volatility increases the chance of reaching that threshold. VXN closed at 32.90 on November 16. It measures near-term option-implied volatility, not the probability of a bear market or volatility throughout 2027 (FRED VXN).
  • History supports substantial risk, but needs adjustment. Historical calculations identify qualifying closes in 25 of 40 calendar years from 1986–2025. Every year from 2000–2014 qualified, so the headline frequency repeatedly counts the aftermath of the same unrecovered peak. Among 24 years starting above the threshold, nine qualified (historical index data).
  • Financial stress is not uniform. Large AI-related financing commitments create shared exposures, but completed financing remains a counterweight. Oracle raised $10 billion in equity and issued $10 billion principal of notes on November 4 (Oracle filing).

Uncertainties

The largest unknown is whether volatility persists or fades. A strong rebound before January would create more distance from the existing threshold. Continued weakness would leave the index exposed to an early qualifying close.

Historical comparisons are small and dependent, and today’s position is not the unknown starting position for 2027. A breach confined to 2026 does not count. Conversely, one qualifying official close during 2027 is sufficient even if the index then recovers. New closing records update the reference high; intraday lows do not count.

Read the full report

TL;DR

The forecast is 63.7% Yes and 36.3% No for at least one qualifying Nasdaq-100 closing drawdown during 2027. The index already sits well below its record, and elevated volatility puts the existing threshold within reach. Read this as a forecast of one closing threshold crossing, not a forecast of the year-end return or a lasting financial crisis.

Context

The question concerns the standard Nasdaq-100 price index, NDX. A trading day qualifies if its official close is at or below 80% of the highest official close recorded strictly before that day. The historical peak carries into 2027 and updates after new closing records. Only a qualifying close during 2027 counts; an intraday breach or a decline confined to 2026 does not. Nasdaq’s official closing history is the primary resolution source (Nasdaq index history).

The corroborated November 16, 2026 close is 27,197.15, replacing the 27,276.26 figure in the question’s description. The relevant record is 31,224.69 on October 6, not the later October 22 close. Those values put the index 12.8986% below its record, with a current qualifying threshold of 24,979.752 index points and a further 8.1531% decline needed to reach it (Nasdaq; Nasdaq-sourced FRED history).

Evidence

The historical backbone is substantial but dependent. Calculations from the closing-price history identify qualifying days in 25 of 40 calendar years from 1986–2025: 1987, 1988, 1990, 1991, 1998, every year from 2000–2014, 2018, 2020, 2022, 2023 and 2025. The resulting 62.5% annual frequency describes the past. It is not, by itself, an appropriate forecast for the current situation (historical index series).

The reason is persistence. The closing peak established in 2000 was not exceeded until November 3, 2015. Every year from 2000–2014 therefore qualified without requiring a fresh bear-market episode each year. Among 24 historical years that began above the qualifying threshold, nine subsequently recorded a qualifying close. That is a more useful counterweight to the headline annual frequency, though it still does not condition closely on today’s drawdown and volatility (historical closing data).

The historical analysis identified eight distinct peak-to-recovery bear episodes. Their full set of reported maximum drawdowns is shown below. These are episode descriptions, not independent annual probabilities (historical price data).

Closing peak date Maximum episode drawdown
October 5, 1987 39.9%
July 16, 1990 32.9%
July 20, 1998 23.0%
March 27, 2000 82.9%
August 29, 2018 23.0%
February 19, 2020 28.0%
November 19, 2021 35.6%
February 19, 2025 22.9%

Closer historical comparisons are mixed. Four mid-November observations starting 10%–20% below the historical closing high split evenly between qualifying and non-qualifying following calendar years. The starting years were 1986 and 1997, followed by qualifying years, and 2014 and 2018, followed by non-qualifying years. That is a useful restraint on a strongly bearish reading, but four observations cannot settle the question (historical index series).

The strongest current evidence for Yes is the distance to the threshold. The question does not require another 20% loss from the latest close. The required further decline against the existing record is 8.1531%. The October record is corroborated by an issuer disclosure as well as the index history. The disclosure reports the price; it does not amount to regulatory verification of it (filed historical index disclosure; FRED).

Volatility makes that remaining distance more consequential. VXN measures annualized, approximately 30-day option-implied volatility. The dated readings reported in the research show both an earlier spike and renewed pressure into the forecast date (FRED VXN history).

Date in 2026 VXN close
September 18 19.29
October 1 22.51
November 5 38.26
November 10 28.57
November 11 26.74
November 12 28.55
November 13 29.38
November 16 32.90

These readings support elevated near-term risk. They do not establish that the same conditions will last through 2027. Realized volatility also differs sharply by measurement window: the reported annualized readings are 36.0% over 20 sessions, 24.6% over 60 sessions and 21.6% over 252 sessions. The short-window spike is real, but treating it as a full-year condition would overstate what the data establishes (underlying closing-price series).

AI-related financing adds a plausible channel for shared earnings and valuation shocks. The Federal Reserve’s November 13 Financial Stability Report described advanced-computing debt issuance exceeding $680 billion through October 2026, compared with $130 billion in 2023, and related public debt outstanding above $1.1 trillion. Its seven-company analysis also identified more than $1 trillion in uncommenced leases and more than $2 trillion in other contractual obligations. Those categories are not interchangeable: multiyear and contingent commitments are not immediate cash demands (Federal Reserve report).

Company disclosures illustrate the connections. NVIDIA disclosed a $105 billion contingent guarantee tied to infrastructure leased to an OpenAI affiliate, with service conditions expected to begin in 2028. Microsoft reported an OpenAI receivable of $7.2 billion at September 30, up from $6.0 billion at June 30. I read these as channels through which a funding or demand disappointment could affect several companies together—not as proof of delinquency or payments falling due in 2027 (NVIDIA agreement; Microsoft filing).

The offsets matter. Oracle completed $10 billion of equity issuance and $10 billion principal of notes on November 4, demonstrating continued access to funding. Between November 11 and 16, VXN rose 6.16 points while broad high-yield option-adjusted spreads widened just four basis points. Equity stress had not received matching confirmation from broad credit markets (Oracle financing disclosure; VXN; high-yield spreads).

The macroeconomic evidence is mixed rather than recessionary across the board. The Federal Reserve maintained a 3.75%–4.00% policy-rate range on October 28, described strong activity and elevated inflation, and recorded two dissents favoring an increase. October unemployment was 4.1%, with payroll employment increasing 165,000. This supports both continued earnings growth and meaningful interest-rate risk (Federal Reserve statement; BLS employment release).

What's non-obvious

The key distinction is between a new crash and an inherited drawdown. The index enters the remaining forecast horizon already below its record. It does not need a fresh 20% decline from the latest level. Nor does it need to stay below the threshold for long. A brief closing breach during 2027 counts fully, even if a strong recovery follows. A positive year-end return and a Yes resolution are therefore compatible.

The historical headline has the opposite trap. A high share of qualifying calendar years does not mean equally frequent independent crashes. Long recoveries cause the same old peak to generate repeated qualifying years. Meanwhile, elevated equity volatility does not by itself establish a funding crisis. The forecast reflects the combination of a nearby threshold and unsettled prices, while retaining the historical and financing evidence that points toward recovery.

Uncertainties

Volatility persistence is the largest uncertainty. Rapid normalization and a strong rebound before January would leave the index farther from the existing threshold. Continued weakness would do the reverse. The closing position at the start of 2027 remains unknown, and a threshold crossing that ends before then does not qualify.

Historical comparisons offer limited precision. The closest annual examples are few, many daily comparisons overlap, and long underwater episodes dominate parts of the record. Small cross-source price discrepancies also remain in older history, although the current close and record used here are corroborated. These limits matter more for fine calibration than for the basic conclusion that the existing threshold is within reach.

The financing evidence also leaves gaps. Multiyear commitments do not reveal the timing or size of future cash pressure, and no contemporaneous AI-issuer credit-spread series was verified. Longer-window market breadth, issuer-specific funding costs, and evidence on whether customer receivables convert into cash would help distinguish a temporary repricing from persistent stress. None of those unknowns changes the saved forecast.

Forecast history

How Prolepsia’s forecast has movedOne forecast so far: 64% on 17 Nov 2026.17 Nov 2026: 64%
DateForecast
64% chance of yes

Sources Prolepsia read

  1. History for NDXindexes.nasdaq.com
  2. NASDAQ-100 (NASDAQ100) | FRED | St. Louis Fedfred.stlouisfed.org
  3. NASDAQ-100 (^NDX) Historical Data - Yahoo Financefinance.yahoo.com
  4. filed historical index disclosurewww.sec.gov
  5. Financial Stability Reportwww.federalreserve.gov
  6. Employment Situation Summary - 2026 M10 Resultswww.bls.gov
  7. Holidays & Trading Hourswww.nyse.com
  8. Nasdaq‑100 Index® Methodology Update: Why Now, and What It Means | Nasdaqwww.nasdaq.com
  9. # INFORMATION ON THE INDEXwww1.hkexnews.hk
  10. History for NDXindexes.nasdaqomx.com

The question

This question asks whether the Nasdaq-100 Index (Nasdaq ticker: NDX) will record an official daily closing value at least 20% below its previous all-time closing high on any trading day from January 1, 2027, through December 31, 2027, inclusive. The Nasdaq-100 is a modified market-capitalization-weighted index tracking 100 of the largest non-financial companies listed on the Nasdaq Stock Market. The question concerns the standard Nasdaq-100 price index (NDX), not the Nasdaq Composite, Nasdaq-100 Total Return Index, or an exchange-traded fund such as Invesco QQQ. A decline of 20% or more from a previous closing high is commonly associated with a bear market. As of November 17, 2026, the Nasdaq-100 has experienced significant volatility, with Nasdaq's official historical data reporting a closing value of 27,276.26 on November 16, 2026, representing a daily decline of 2.39%. The index's closing value on October 22, 2026, was 28,958.69, illustrating the substantial market movements observed in recent weeks. These figures are available from Nasdaq's official index history (https://indexes.nasdaq.com/Index/History/NDX) and SEC-filed historical index disclosures. The forecast will assess whether the index experiences a qualifying drawdown during calendar year 2027, regardless of whether the index subsequently recovers. The reference high will be the highest official daily closing value recorded strictly before the trading day being evaluated, including historical closing values from before January 1, 2027. The reference high will update whenever the index achieves a new all-time closing high. Consequently, a decline from a peak established in 2026 or earlier can qualify if the index closes at least 20% below that peak on a trading day in 2027.

How it resolves

The question resolves YES if, on at least one official Nasdaq-100 trading day between January 1, 2027, and December 31, 2027, inclusive, the index's official closing value is less than or equal to 80% of its previous all-time closing high. For each eligible trading day t, define C(t) as the official Nasdaq-100 Index closing value on that day and H(t) as the maximum official closing value recorded on any trading day strictly before t, using the complete historical series of the Nasdaq-100 price index. The question resolves YES if C(t) ≤ 0.80 × H(t) for at least one eligible trading day. Equality qualifies. The question resolves NO if no eligible trading day during 2027 meets this condition. The primary resolution source will be Nasdaq's official Nasdaq-100 Index historical data, available at https://indexes.nasdaq.com/Index/History/NDX. Nasdaq's published official closing values will take precedence over third-party data. If Nasdaq's historical records are unavailable, Bloomberg's historical daily closing values for NDX will serve as the first fallback, followed by LSEG/Refinitiv or another established financial data provider reporting the official NDX closing series. The determination will use closing values, not intraday lows, intraday highs, futures prices, or ETF prices. The reference high must precede the evaluated trading day and is not restricted to the 2027 calendar year. The question can resolve YES as soon as an official qualifying closing value is confirmed. Otherwise, it will resolve NO after the final trading session of 2027, once the official closing values for all eligible trading days are available. If the index is discontinued or permanently replaced during 2027, the question will be evaluated using official NDX closing values published before discontinuation. If a qualifying close occurred, it resolves YES; otherwise, the question will be annulled if discontinuation prevents evaluation of the full observation period. Temporary trading suspensions, holidays, and market closures do not themselves cause annulment. If the historical data needed to determine whether a qualifying drawdown occurred remain unavailable from both the primary and fallback sources, the question will be annulled.

Fine print

The observation window is based on trading dates in the America/New_York time zone. January 1, 2027, is a U.S. market holiday, so the observation period begins with the first actual Nasdaq-100 trading session of 2027 and ends with the final trading session on or before December 31, 2027. Only official end-of-day index closing values count. A 20% or greater intraday decline that reverses before the official close does not qualify. The reference high is a rolling historical maximum of official closing values, not the highest intraday level or the highest value within a fixed lookback period. A drawdown that begins before 2027 qualifies if the index closes at least 20% below its previous all-time closing high on an eligible trading day during 2027. A drawdown that reaches the threshold only after December 31, 2027, does not qualify. Similarly, a 20% decline that occurs entirely during 2026 does not qualify unless the threshold is also met on at least one eligible trading day in 2027. Standard index constituent changes, rebalancing, and methodological adjustments do not reset the reference high. Official historical index values will be used as published by Nasdaq, including any corrections or retrospective adjustments Nasdaq explicitly applies to its historical series. The percentage decline will be calculated using the highest available precision of the official published closing values, without rounding the resulting percentage before comparison with the 20% threshold. If Nasdaq revises an official closing value, the corrected value will take precedence, provided the correction is published before final resolution. A qualifying drawdown requires only one eligible closing value, regardless of how long the decline lasts or whether it is subsequently reversed.