Those registration numbers aren't reflected in the super dooper "MRB" poll either....which was also done a week or so ago and doesn't reflect where we are now as much.
Yougov weightings are one of the harshest against 2017 as well. Easily 6 points differential, potentially more if 2017 levels of turnout from all groups are kept the same.
Factor in as well more than one Poll has always moved to 7% difference (from the 11% they are working from) and there have been swings shown from 3-7% in the week this MRP has been conducted, which again puts us around the 7% phase).
So essentially you have 2 options. One is that the polls are say upwards of 6% out, the other is that the polls have moved say 4% in the last week. Both could also be true, but what would be the impact if one where the case?
On this model, a 2.5% swing (so a 5% closing from Con to Lab) would move 67 seats from Con-Lab on this prediction. Not only is that enough to prevent a Tory Majority government, probably wipe out any chance of a Tory minority government, we would be looking at a situation whereby the Tories were potentially not even the biggest party.
There are still two weeks to run as well. Where Labour really started piling swings up last time. It's easy to see why the tories and Cummings are getting increasingly agitated.
If Labour get within 6% they are winning seats back net on this model.