| cff-version: 1.2.0 | |
| message: "If you use this dataset, please cite it as below." | |
| title: "AFOS · Germany 2025 Electoral Divergence Dataset" | |
| abstract: >- | |
| Open dataset cross-referencing opinion polls and prediction markets for Germany's | |
| 2025 federal election (Bundestag, 23 February 2025), with explicit poll-versus-market | |
| divergence rather than a blended average. Polls measure party vote share; the market | |
| prices the probability of winning the most seats. The AfD held ~21% of the vote (second | |
| party) yet only ~3% to win the plurality — vote share is not the same as winning, and | |
| the spread is the signal. Poll figures compiled deterministically from public pollster | |
| releases (via the Wikipedia aggregation); market odds from Polymarket. | |
| type: dataset | |
| authors: | |
| - name: "AFOS Analytics" | |
| website: "https://afos-analytics.com" | |
| keywords: | |
| - elections | |
| - Germany | |
| - opinion polls | |
| - prediction markets | |
| - divergence | |
| - Europe | |
| - open data | |
| license: CC-BY-4.0 | |
| repository: "https://huggingface.co/datasets/AFOS-Analytics1/germany-2025-electoral-divergence" | |
| url: "https://afos-analytics.com" | |
| version: "2026.06" | |
| date-released: "2026-06-11" | |