This is the starting point: the swing to the Conservative Party before the IHT-focused voters start switching.
The model counts the people in each seat whose estates would face a bill, takes off the ones who are voting Conservative already and the ones who aren't voting. The slider then sets the % of those in this group who will vote Conservative if they are impacted by IHT. Their old votes come off the other parties in proportion to how well each did in that seat.
About 27% of over-65s did not vote in 2024. This slider sets the % of those who will now vote Conservative if impacted by IHT. The slider stops at 20%, which is twice the largest net shift ever seen: the best any single election has managed is to pull in about a tenth of previous non-voters.
Someone is less likely to switch their vote over a small theoretical future tax, so this is how big the potential tax hit has to be before the other two sliders apply to them. It's as a share of what the household owns. At 1% almost every exposed household counts; at 20% you are down to estates over about £2m (because of the effect of the nil rate band).
When selected, children's votes move with their parents (if they are 16+). That makes sense logically, but data limitations mean we have to make the assumption that the children live in the same constituency as their parents. That is clearly a poor-quality assumption, which is why we give you the option of switching this on and off.