TAILFIFTY · RESEARCH

Will CBOT corn futures rise by at least 8% within 45 days given the Corn Belt rainfall anomaly?

commodities45dasked 2026-07-15 · resolves on CORN.close return>= 0.08

Committee15%

OutcomeYESBrier 0.722

Committee

Weather Analyst15%was 25%

Commodities Analyst16%was 19%

Quantitative Researcher15%was 17%

Historical Analog Analyst14%was 15%

Market Positioning Analyst16%was 4%

Causal Skeptic16%was 22%

Top disagreement

C's walk-forward t5 (OOS mean +2.76%, p=1.6e-05) directly contradicts D's walk-forward t8 (OOS mean -2.43%, p=0.00124, hit rate 30.4%) on what both describe as the same rainfall-signal rule under expanding-window validation -- both significant, opposite signs. Resolved by weighting D's t8 higher because it is corroborated by two independent, different-methodology lines of evidence (E's raw volume-crowding data; F's soybean-confound mechanism finding), whereas C's t5 stands alone and contradicted; C itself conceded the direction shift citing D's t8. Not fully resolved absent shared code/methodology, so residual uncertainty is carried into a wider credible interval rather than treated as settled.

Synthesis

Consensus P(YES)=0.15 sits below the committee's own resolved unconditional base rate (~0.20), reflecting that the best-powered conditional evidence (C/D's n=551 lift test, D's significant OOS walk-forward) and the one genuinely novel data point (E's volume-crowding signal, corroborated by F's soybean-confound finding) together argue the rainfall signal is associated with equal-or-lower, not higher, forward large-rally odds -- i.e., the market has likely already substantially repriced the catalyst. The tradeable implication is a fade, not a chase: sell an OTM Dec corn call spread (short strikes near +8% from the 07-15 close, long strikes further OTM to cap tail risk), sized small given capped confidence (0.40-0.62 across the panel) and a real, untestable tail risk that a genuine pollination-window drought intensification is not yet visible in the July 15 data. Monitor the explicit 07-29 volume/price checkpoints as an early falsification trigger before the 45-day resolution date.

What is what

Observed

CORN.close jumped 433.8 (07-14) to 447.5 (07-15), a +3.2% single-day move, at the as_of date.
Trading volume spiked to ~153,300 contracts on 07-15 from a near-zero baseline (195/8,828) on preceding sessions.

Inferred

A meaningful share of the immediate weather-news reaction is already reflected in price and positioning (crowding), which is a headwind to, not a tailwind for, a further 8% move from current levels.
The rainfall signal, once fired, is associated with equal-or-lower forward large-rally odds relative to the unconditional base rate, not higher.

Modeled

Unconditional P(45-day corn return >=8%) ~ 0.1996 (empirical, n=551), corroborated by independent volatility-scaling and quantile-interpolation methods.
Conditional-on-rainfall-signal P(8%+ rally) ~ 0.15-0.16 with lift <1 (n=43, CI spans zero).
OOS walk-forward expected return when signal fires ~ -2.4% under one construction (D's t8, significant), contradicted by +2.76% under another (C's t5, significant) on what is claimed to be the same rule.

Predicted

P(YES, CORN.close >= 483.3 on 2026-08-29) = 0.15, credible interval [0.08, 0.26].
Expected 45-day return conditional on this rainfall-anomaly setup is approximately -1% to 0%, well short of the +8% threshold.

Uncertain

Whether C's t5 or D's t8 walk-forward construction is authoritative -- both significant, opposite signs, on a nominally identical rule.
The severity/percentile of the current rainfall anomaly relative to historical anomaly episodes -- no test in this dataset addresses magnitude, only a blunt 0/1 signal.
Whether a late-window (day 20-45) drought intensification not yet visible in any July 15 read could invalidate the 'already priced' thesis.
Whether the sample-construction discrepancy (n=0 vs 17 vs 43 vs 551 for nominally the same conditional test) reflects genuinely different conditioning events or a shared data-pipeline bug -- never reconciled with common code across analysts.

Falsification

CORN.close on 2026-08-29 >= 483.30 (447.5 x 1.08) falsifies P=0.15 as too low.
If by 2026-07-29 volume has reverted toward pre-anomaly baseline (near 195-8,828) without a second leg higher in price, this corroborates the fade/crowding thesis underlying P=0.15.
If by 2026-07-29 volume stays elevated (>50k/session) AND price holds/grinds above ~460, this is an early signal that P=0.15 is too low and should be revised upward toward the committee's base-rate-neutral 0.20 or higher before the 45-day mark.

← all research

Every forecast is pre-registered with a resolution rule and scored when it resolves. Numbers come from deterministic tests over point-in-time data. Every seat on the committee is an AI model with a role prompt, not a person or a licensed professional. TailFifty is an educational research experiment run by an individual. Probabilities and paper trades are AI-generated and simulated; there are no real positions. Nothing here is investment, trading or betting advice, or an offer or solicitation. Not a registered investment adviser, broker or CTA. Hypothetical and simulated results do not reflect real trading, liquidity or costs, and past forecast accuracy does not guarantee future results. Data: NOAA, NHC, CFTC, Open-Meteo (CC BY 4.0), CoinGecko, Kalshi and others; derived figures only. © 2026 Harrison Johnson. About and disclosures.