Compound Fork — AGI × Recession
Each cell = a joint scenario "Both branches fire". Cell intensity = total |Δ| across the 200 highest-conviction tradeable predictions vs their unconditional posterior. Joint probabilities approximated via log-odds combination (assumes conditional independence given the prediction — coarse but bounded). Click a cell to drill into the dominant single-fork view.
Pick two fork families
| AGI ↓ × Recession → | RECESSION_2026 prior 20% | RECESSION_2027 prior 30% | RECESSION_2028 prior 30% | NO_RECESSION_5Y prior 20% |
|---|---|---|---|---|
AGI_FAST_2027 prior 30% | 127 claims · Σ|Δ| 19.30 | 128 claims · Σ|Δ| 19.37 | 128 claims · Σ|Δ| 19.37 | 129 claims · Σ|Δ| 19.42 |
AGI_MID_2029 prior 35% | 125 claims · Σ|Δ| 19.00 | 126 claims · Σ|Δ| 19.08 | 126 claims · Σ|Δ| 19.08 | 127 claims · Σ|Δ| 19.12 |
AGI_SLOW_2031 prior 25% | 133 claims · Σ|Δ| 19.87 | 134 claims · Σ|Δ| 19.94 | 134 claims · Σ|Δ| 19.94 | 135 claims · Σ|Δ| 19.99 |
AGI_WINTER_2036PLUS prior 10% | 129 claims · Σ|Δ| 19.68 | 130 claims · Σ|Δ| 19.76 | 130 claims · Σ|Δ| 19.76 | 132 claims · Σ|Δ| 19.86 |
Method note
Joint conditional probability is approximated via log-odds combination: logit(P(pred|A,B)) ≈ logit(P(pred|A)) + logit(P(pred|B)) − logit(P(pred)). This is the closed-form Bayesian update assuming A and B are conditionally independent given the prediction. It's correct when the two scenarios act on the prediction through different causal paths; it's pessimistic when they overlap. The exact joint requires running the Gibbs sampler with both scenarios clamped, which would be N×M=16 sampling runs (~12 minutes per refresh) instead of N+M=8 — a 2× cost for higher fidelity.