I’ve done some tweaking of the model, assuming EPMU vote is not as strong for Robertson as previously, and also that Cunliffe does better in the members vote. The final model (not a prediction – this is more a learning exercise) is:
|Modelled Labour Vote by Section|
|Final Weighted Vote||Cunliffe||Robertson|
|Total Final Vote||52.3%||47.7%|
What it doesn’t take account of is different turn-out by region for the members. I understand Camp Cunliffe had a 48 hour call bank running where they called almost every member likely to vote for him. An operation like that can make a big difference.
But to balance that is one MP is worth several hundred members.