We tested 49 causal chains. One worked.

Most causal claims in finance don't survive contact with data.

Yes No Capital was built to test relationships between prediction market events. Academic literature offers dozens of plausible mechanisms — inflation depresses incumbent vote share, oil spikes precede recessions, tariffs raise consumer prices, war depresses markets, rate cuts move equities, credit spreads forecast corporate distress. Forty-nine such chains, each sourced to a peer-reviewed paper.

Each chain becomes a keyword rule: match inflation events with election events; when the cause resolves YES, measure whether the effect resolves YES more often than the category baseline. A chain earns a PREDICTIVE verdict only when the effect is materially lifted at meaningful sample size. Everything else is NO EDGE, INVERTED, or INSUFFICIENT.

After more than ten thousand resolved pairs, one chain survives: nuclear policy events correlate with nuclear-related markets above baseline, stable across time-split validation. That is weak. It is also the only one that passes all the gates.

The majority of chains show no edge despite 500+ pairs each. Inflation → elections. GDP → elections. Oil → inflation. Tariffs → GDP. At scale, the "obvious" macro relationships don't discriminate prediction market outcomes. Keyword pairing is not causation.

This is why backtests are easy to publish and hard to trust. Sample-cap an hour earlier, filter by volume, restrict to a favorable window, and any chain can look predictive. The honest version of a scoreboard is rarely what publication incentivizes.

What the validation catches

• Walk-forward validation catches chains that had past edge but lost it. Overall lift can look fine when stability across time splits is not.
• Uncapped sampling catches chains that only look predictive under curation.
• Directional spread catches chains with overall lift but no real discriminant power between cause-true and cause-false states.

The scoreboard at yesnocapital.com/track-record is updated hourly as new events resolve. Chains auto-deprecate when the evidence is in — the library is self-cleaning.

One surviving chain is not a product. It is a start. The point isn't any single chain; it is that most things don't work, and making that visible is the only honest version of this kind of tool.