Forecasting on FB A∪B
94.8BI (Mkt)BLF +crowd+emp+cal
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| BLF +crowd+emp+calcomponents=crowd+emp+cal2026.04 | 94.8 | 72.2 | 83.5 | — | — | — | — | — | — | — | |
| Foresight-32Bsource=FB leaderboard, overlap=partial2026.04 | 91.4 | 67.1 | 79.2 | — | — | — | — | — | — | — | |
| GPT-5 ZS+freezemode=zero-shot, freeze=true, source=FB leaderboard2026.04 | 90.9 | 69 | 79.9 | — | — | — | — | — | — | — | |
| Crowd+emp (no LLM)LLM=false2026.04 | 90.6 | 65.4 | 78 | — | — | — | — | — | — | — | |
| Cassisource=FB leaderboard2026.04 | 90.4 | 68.6 | 79.5 | — | — | — | — | — | — | — | |
| Grok 4.20source=FB leaderboard, overlap=partial2026.04 | 89.3 | 68.3 | 78.8 | — | — | — | — | — | — | — | |
| BLF +crowd+emp+calcomponents=crowd+emp+cal2026.04 | 88.3 | 39.9 | 64.1 | — | — | — | — | — | — | — | |
| ZS+crowd+empmode=zero-shot2026.04 | 87.7 | 61.8 | 74.7 | — | — | — | — | — | — | — | |
| CassiFB leaderboard=true2026.04 | 81.5 | 33.1 | 57.3 | — | — | — | — | — | — | — | |
| Foresight-32BFB leaderboard=true, Partial overlap=true2026.04 | 81.4 | 27.6 | 54.5 | — | — | — | — | — | — | — | |
| Crowd+emp (no LLM)LLM included=false2026.04 | 80.8 | 29.4 | 55.1 | — | — | — | — | — | — | — | |
| GPT-5 ZS+freezeFB leaderboard=true, Mode=Zero-shot, State=freeze2026.04 | 78.5 | 35.4 | 56.9 | — | — | — | — | — | — | — | |
| ZS+crowd+empMode=Zero-shot2026.04 | 75.5 | 9 | 42.2 | — | — | — | — | — | — | — | |
| Grok 4.20FB leaderboard=true, Partial overlap=true2026.04 | 72.2 | 37.2 | 54.7 | — | — | — | — | — | — | — | |
| BLFcrowd=true, emp=true, cal=true2026.04 | — | — | — | — | — | — | 71 | — | — | — | |
| BLFCompared against=Cassi, n=7902026.04 | — | — | — | — | — | — | — | 4.4 | 3.5 | 4 | |
| BLFCompared against=GPT-5, n=7902026.04 | — | — | — | — | — | — | — | 3.8 | 3.2 | 3.5 | |
| BLFCompared against=Grok, n=423, Scope=A-only2026.04 | — | — | — | — | — | — | — | 5.9 | 4.3 | 5.1 | |
| BLFCompared against=Foresight, n=417, Scope=B-only2026.04 | — | — | — | — | — | — | — | 4 | 4.6 | 4.3 | |
| BLFCompared against=Crowd+emp, n=7902026.04 | — | — | — | — | — | — | — | 4.2 | 6.7 | 5.4 | |
| BLF +crowd+emp+calComponents=Crowd + Empirical + Calibration, Trials=K=52026.04 | — | — | — | 2.2 | 14.1 | 8.2 | — | — | — | — | |
| CassiFB leaderboard=true2026.04 | — | — | — | — | — | — | 68.7 | — | — | — | |
| CassiCompared against=Crowd+emp, n=7912026.04 | — | — | — | — | — | — | — | 0.3 | 3.2 | 1.5 | |
| CassiCompared against=GPT-5, n=7912026.04 | — | — | — | — | — | — | — | 0.6 | 0.4 | 0.5 | |
| CassiCompared against=Grok, n=423, Scope=A-only2026.04 | — | — | — | — | — | — | — | 0.4 | 1.3 | 0.9 | |
| CassiCompared against=Foresight, n=418, Scope=B-only2026.04 | — | — | — | — | — | — | — | 0 | 0.4 | 0.2 | |
| Cassi*Source=FB leaderboard2026.04 | — | — | — | 3.2 | 16.3 | 9.8 | — | — | — | — | |
| Crowd+empno LLM=true2026.04 | — | — | — | — | — | — | 67.7 | — | — | — | |
| Crowd+emp (no LLM)LLM Use=None2026.04 | — | — | — | 3.4 | 17.4 | 10.4 | — | — | — | — | |
| ForesightCompared against=Crowd+emp, n=418, Scope=B-only2026.04 | — | — | — | — | — | — | — | 0.6 | 0.1 | 0.4 | |
| Foresight-32BFB leaderboard=true, partial overlap=true2026.04 | — | — | — | — | — | — | 67.5 | — | — | — | |
| Foresight-32B*†Source=FB leaderboard, Overlap=partial, Backbone=Foresight-32B2026.04 | — | — | — | 3.1 | 17.9 | 10.5 | — | — | — | — | |
| GPT-5ZS=true, freeze=true, FB leaderboard=true2026.04 | — | — | — | — | — | — | 68.5 | — | — | — | |
| GPT-5Compared against=Crowd+emp, n=7912026.04 | — | — | — | — | — | — | — | 0.3 | 3.5 | 1.9 | |
| GPT-5Compared against=Grok, n=423, Scope=A-only2026.04 | — | — | — | — | — | — | — | 2.1 | 0.9 | 1.5 | |
| GPT-5Compared against=Foresight, n=418, Scope=B-only2026.04 | — | — | — | — | — | — | — | 0 | 1.6 | 0.8 | |
| GPT-5 ZS+freeze*Model State=freeze, Mode=Zero-shot, Source=FB leaderboard, Backbone=GPT-52026.04 | — | — | — | 3.9 | 15.9 | 9.9 | — | — | — | — | |
| GrokCompared against=Crowd+emp, n=423, Scope=A-only2026.04 | — | — | — | — | — | — | — | 2 | 4.4 | 1.2 | |
| Grok 4.20FB leaderboard=true, partial overlap=true2026.04 | — | — | — | — | — | — | 68.4 | — | — | — | |
| Grok 4.20*†Source=FB leaderboard, Overlap=partial, Backbone=Grok 4.202026.04 | — | — | — | 5 | 14.9 | 10 | — | — | — | — | |
| ZS+crowd+empMode=Zero-shot, Components=crowd+emp2026.04 | — | — | — | 4.7 | 23.1 | 13.9 | — | — | — | — | |
| ZS+crowd+empZero-Shot=true2026.04 | — | — | — | — | — | — | 62.8 | — | — | — |