You are reading a single comment by @Velocio and its replies.
Click here to read the full conversation.
-
any idea how these are modelled for seats where the boundary has changed quite a bit?
With a lot of complexity.
In essence by taking historical voting patterns, other boundary information of different granularity like council elections, etc, and attempting to project from those the likely demographic, and then mapping the existing polling to the demographic.
Most tactical voting websites will just show historical seat voting, as it's unambiguous and easy to communicate, but their recommendations are made in a more complex way to try and make the most likely to win recommendation factoring in the changes.
There are a lot of spreadsheets involved.
Velocio
aggi
Out of curiosity, and as you seem ideally placed on this given that you are in a seat that has massively changed boundaries and run a tactical voting website, any idea how these are modelled for seats where the boundary has changed quite a bit?