Odds of War — Prediction-Market Probability Curves for 141 Historical Wars

—

Final odds

—

Hover for the event behind each move · shaded band = reconstruction uncertainty · ▶ Replay relives it without hindsight

Methodology

—

Primary sources

Key uncertainties

—

Key events & odds shifts

Does history price in?

Calibration of all — probability points across — resolved wars: when the curves said X%, how often did YES actually happen?

Stated odds vs observed outcomes

Opening favorite won

—

Avg opening odds on the eventual outcome

—

50% would mean wars open as pure toss-ups

Decided by a single shock

—

resolved wars with one ≥30-pt event

Wire-to-wire winners

—

outcome led the entire war, never trailed

Honesty note: most curves were reconstructed knowing how the war ended, so calibration here mostly measures the reconstruction's internal coherence — not genuine foresight. The market-anchored series (high-confidence wars only) is the subset with real contemporaneous prices behind it.

Across all wars

Computed live from every curve on this site. Click any row to jump to that war.

Biggest upsets

Most volatile curves

Largest single shocks

Confidence levels: high means contemporaneous market prices directly anchor the curve (gold premium for the Civil War; consol yields and Reich-bond prices for WWI/WWII/Napoleonic); medium means partial market data plus strong historiographical consensus; low means pure historian reconstruction with no market anchor. For low-confidence wars, treat the curve as one defensible interpretation rather than a measurement. Probabilities for ongoing wars reflect Polymarket and Manifold data plus the author's calibration.