- Uncertainty due to drift between now and the election
- Uncertainty due to difference between last election poll and final vote share
- Better prior by incorporating more past elections
- Think about how to:
  - Inject error
  - Inject correlated error

---

Not as good as gallup.
State uncertainty higher than national.
Should calculate republican error instead. 

Final error:
National drift + state more uncertain than national + not as good as gallup error + idosyncratic state error

later: error correlated across states