Guy in response to your items, I acknowledge that this is very much just a starting point for discussion and therefore the comments are welcomed.

I allocated the points purely as an instinctual/gut feeling based on my experience of the documents. The reasoning behind the varying points for the censuses was that each has a varying number of data points, and therefore more items that can be cross referenced against other evidence. As such if points were subtracted for inconsistencies then logically the later census documents have more chance of having those deductions than the earlier censuses.

I think this is why percentages are dangerous because you will have the potential to achieve 90% plus and in reality this could never be the case.

However if an individual has more documentation to support their existence and this data correlates with other sources then it would just visually display as a higher score. This would highlight potential weaker links in the ancestral line at a glance.

Your item number 7 is a similar approach but awarding points or accuracy rather than deducting them for inaccuracies