US test scores in Reading in 2026?
After a blind assessment of the evidence I place the probability of a meaningful 2026 decline in U.S. reading scores at about 40%; the most likely single outcome is a negligible net change rather than a large drop, and markets at ~47% are somewhat tilted toward expecting a pronounced decline.
Analysis
**Stage 1 — Blind analysis (ignore market prices):**
The most relevant empirical signal is mixed by age cohort. Fourth-grade reading remains well below 2019 levels (215 in 2024 vs 220 in 2019) and only about 31% of 4th graders are at or above proficiency. Older cohorts — notably 13- and 17-year-olds — show persistent weakness: 12th-grade reading is historically weak and 13-year-olds are still below pre-pandemic levels. At the same time, the NAEP long-term trend reported a 4-point gain for 9-year-olds versus 2022, but that gain only returns those cohorts to roughly 2013 levels. District-level analyses show about one-third of districts still below 2015 reading levels.
Taken strictly on these facts, the short-term trajectory into 2026 is ambiguous: there are pockets of modest recovery at younger ages, but stagnation or further weakness at middle/high grades. A few structural points push against a large, sudden national decline in 2026: recent improvements among younger cohorts reduce the universal downward pressure; major federal COVID-era remediation funds (and focused literacy programs) have been deployed and can show effects within a 1–3 year window in some places; and measurement noise and sampling variability in NAEP and long-term trend cycles mean small year-to-year swings are common. Offsetting that, the national baseline is low, teacher shortages/turnover continue in many districts, and the pandemic likely intensified preexisting negative trends that may re-emerge as students age into higher grades.
Weighing magnitude and timing, I judge a substantial, statistically significant nationwide decline in 2026 less likely than either no meaningful change or a modest mixed picture (small gains in younger grades offset by flat/weak older-grade performance). My blind (data-only) probability for a significant decrease is ~40%, for no significant difference ~45%, and for a significant increase ~15%.
**Stage 2 — Market calibration (compare independent view to market prices):**
Current market pricing places _Significant decrease_ at 47% and _No significant difference_ at 46% (binary Yes = 47%). My independent estimate (40%) is lower than the market's 47% by 7 percentage points. Possible reasons the market is leaning more bearish than my reading:
- Traders are overweighting the visible headline declines in 4th- and 12th-grade aggregates (large backward step from 2019 to 2024) and extrapolating that momentum forward into 2026. - The market may be focusing on vulnerability in upper grades where remediation is slower and where cohort-level deficits are already entrenched; those structural risks are salient and easy to price. - Ambiguity in the event wording and measurement thresholds (what counts as a 'decline' vs 'no significant difference') can lead traders to favor the downside as a simpler forecast. - Participation bias: active traders in these markets may skew to pessimistic interpretations of education data or are responding to concentrated news cycles emphasizing declines.
Conversely, the market might be underpricing the countervailing evidence I consider important: modest gains among younger cohorts, localized but impactful remediation efforts, and the high probability that any 2026 national change will be heterogeneous by grade and district — increasing the chance that net national change is not 'significant' on the metric used. If you prefer to trade relative value to the market, the 7-point gap suggests a moderate edge for positions that take the view of a smaller chance of a significant decline.
Summary conclusion: blind evidence and structural judgment point to a moderate likelihood of continued weakness but not overwhelmingly likely to produce a large, statistically significant nationwide drop in 2026; the market is modestly more bearish than my independent readout.
Arguments
For
- Older-grade stagnation: middle- and high-school reading remains weak and could drive a national down-tick when cohort effects surface.
- Low baseline vulnerability: with many districts below pre-2015 levels, small additional setbacks can register as measurable declines.
- Pandemic-amplified slide: the pandemic intensified preexisting negative trends which can continue to depress scores absent rapid remediation.
- District-level concentration: a substantial share of districts are still underperforming, increasing the chance of a noticeable aggregate decline.
Against
- Younger-cohort gains: 9-year-olds showed a 4-point gain versus 2022, signaling early recovery that cushions against broad declines.
- Partial rebound only: gains have restored some cohorts to roughly 2013 levels, which reduces the likelihood of an additional big drop in 2026.
- Policy response and investment: remediation programs and targeted literacy initiatives could blunt further decline within the time window.
- Measurement and sampling noise: year-to-year NAEP variation and the particular test cycle in 2026 could produce non-significant net changes despite local volatility.
Key drivers
- Grade-specific trends: modest gains among younger cohorts (9-year-olds) vs stagnation/weakness at middle and high school grades
- Baseline and momentum: already-low national reading levels that can amplify small negative shocks into measurable declines
- Policy and remediation: scale/timing of district- and state-level literacy interventions and use of federal recovery funds
- Measurement timing and noise: NAEP cycle timing, sampling variance, and definition of 'significant' change
Risk factors
- Event definition ambiguity — unclear thresholds for 'significant' decline increase outcome uncertainty
- Heterogeneity across districts — concentrated severe declines in many districts could still produce a national signal
- Lagged cohort effects — improvements at younger grades may not translate into upper-grade recovery within a single year
- Unobserved shocks (economic downturns, major policy reversals, or large staffing reductions) could worsen outcomes quickly
Scenarios
Best case
Targeted remediation and learning recovery accelerate through 2025–26, younger cohorts continue improving and interventions begin to lift middle-school performance; net national reading scores register a statistically significant increase in 2026.
Most likely
Mixed outcome: modest gains for younger grades offset by continued weakness in middle/high school, resulting in no statistically significant national difference (or only a small net decline) for 2026 when measured on the standard national metric.
Worst case
Worsening staffing shortages, funding contractions, or unaddressed cohort deficits cause reading performance to drop across multiple grades and regions, producing a clear, statistically significant national decline in 2026.
Outcomes
| Outcome | AI | MKT |
|---|---|---|
| Significant decrease | 40% | 47% |
| No significant difference | 45% | 46% |
| Significant increase | 15% | 14% |
More from this day
- FinancialsKalshi13y
Will OpenAI or Anthropic IPO first?
AI25%MKT87%Edge-62HypedI assess a 25% chance that OpenAI will IPO before Anthropic. Anthropic's earlier confidential filing and high probability of a late‑2026 announcement make it the clear favorite to go public first.
- PoliticsKalshi2y
Who will Trump pardon?
AI3%MKT46%Edge-43HypedExtremely unlikely — I assess a low single-digit chance (~3%) that Barron Trump will receive a presidential pardon before Jan 21, 2029; pardons are legally possible absent conviction, but there is no factual or political path that makes this plausible at scale.
- EconomicsKalshi2y
Will there be a Trump economic boom?
AI8%MKT51%Edge-43HypedI assess an 8% chance that U.S. real quarterly GDP (annualized) will exceed 5% in any quarter between Q1 2025 and Q4 2028 — a low-probability tail event driven only by extreme fiscal/monetary shocks, big inventory swings, or large data revisions.