Compound Fork — Robotaxi × AI pause
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
| Robotaxi ↓ × AI pause → | AI_PAUSE_2026 prior 5% | AI_PAUSE_2027 prior 10% | AI_PAUSE_2028 prior 10% | NO_AI_PAUSE_5Y prior 75% |
|---|---|---|---|---|
ROBOTAXI_TESLA_2026 prior 40% | 122 claims · Σ|Δ| 15.98 | 122 claims · Σ|Δ| 15.98 | 123 claims · Σ|Δ| 16.04 | 121 claims · Σ|Δ| 15.20 |
ROBOTAXI_NATIONWIDE_2028 prior 45% | 124 claims · Σ|Δ| 16.14 | 124 claims · Σ|Δ| 16.14 | 125 claims · Σ|Δ| 16.20 | 122 claims · Σ|Δ| 15.31 |
ROBOTAXI_MASS_2030 prior 30% | 121 claims · Σ|Δ| 15.76 | 121 claims · Σ|Δ| 15.77 | 122 claims · Σ|Δ| 15.83 | 118 claims · Σ|Δ| 14.92 |
ROBOTAXI_DELAYED prior 20% | 124 claims · Σ|Δ| 16.15 | 124 claims · Σ|Δ| 16.16 | 125 claims · Σ|Δ| 16.22 | 122 claims · Σ|Δ| 15.32 |
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.