Research | Forecast Alpha

The engine computes every number; the AI layer only explains them. Select a market and generate a structured brief grounded in the engine's persisted inputs snapshot. Without an LLM key, a deterministic template renderer produces the brief from the same snapshot. Briefs are persisted and timestamped.

Will Netherlands win the 2026 FIFA World Cup?

Template renderer (no LLM key)

Generated Mon, 22 Jun 2026 18:00:16 GMTData quality

69Moderate

simModel confidence 0.89sim

Summary

Market prices Will Netherlands win the 2026 FIFA World Cup at 4.0%; the model estimates 20.5% (confidence 0.89). Best-side EV after modeled costs is 14.0% on YES. Engine signal: LONG YES.

What the market implies

The market-implied probability is 4.0%. At a liquidity score of 100/100 the modeled cost of taking a position is 2.5% (2.0% fees + 0.5% slippage), which any edge must clear before it is tradeable.

What the model estimates

Starting from a 22.6% base rate (-1.232 log-odds), the factor stack moves the raw model estimate to 22.5%: pm.momentum_7d -0.01 (-0.002 log-odds), pm.cross_market_divergence +0.00 (+0.000 log-odds), crypto.btc_eth_momentum_7d +0.00 (+0.000 log-odds), macro.rate_surprise +0.00 (+0.000 log-odds), macro.yield_curve_shift +0.00 (+0.000 log-odds), news.source_weighted_signal +0.00 (+0.000 log-odds), crowd.calibrated_aggregate +0.00 (+0.000 log-odds). Confidence of 0.89 then shrinks the final estimate toward the market price, landing at 20.5%.

Why they differ

The model sits 16.5% above the market. The largest single driver is the pm.momentum_7d factor at -0.01 (-0.002 log-odds): 7-day drift of the market's own implied probability — sustained moves carry information. Note that the sentiment and source factors are simulated inputs in this MVP — the disagreement structure is real, the inputs are not.

Base-rate analysis

The engine's prior for this market is 22.6%, intended to reflect how often events of this class resolve YES historically. The prior contributes -1.232 log-odds before any market-specific evidence; with factor adjustments totalling -0.002 log-odds, the raw model lands at 22.5%.

Comparable events

Event	Outcome	Relevance
Recession-within-a-year markets since 2008	Persistently overpriced vs realized frequency	Macro doom trades carry a structural premium.

Recent information

Trend and momentum factors are computed from the stored 30-day price path (trend undefined, momentum undefined). Sentiment and source-quality inputs are simulated in this MVP and are labeled as such wherever displayed.

Bull case

For YES: the model's 20.5% estimate against an entry near 4.0% leaves 14.0% of EV after costs. If the factor evidence is genuine information the market has not priced, expected value accrues as the market converges toward the model.

Bear case

Against the position: the market aggregates more information than any four-factor model. If the pm.momentum_7d signal is stale or spurious, the true probability is closer to the market's 4.0% and the position's EV is roughly the negative of its costs (2.5%).

Key uncertainty

Whether the pm.momentum_7d factor reflects real, unpriced information — and whether resolution follows the thesis or turns on a definitional edge in the criteria.

What could make this wrong?

The model's edge depends on its inputs being right. Concretely: the base rate of 22.6% may not apply if this event differs structurally from its reference class; the pm.momentum_7d factor could be noise rather than information at this horizon; and with confidence at 0.89, the model itself concedes meaningful estimation error. Resolution risk remains: the contract pays on the precise criteria — "This market will resolve according to the national team that wins the 2026 FIFA World Cup. If at any point it becomes impossible for this t…" — not on the thesis.

Turn model inputs into evidence-backed market briefs.

Brief context loaded

Research Engine