You are correct in saying that voter turnout on advance polling days is not a great predictor of overall voter turnout. That is something which has been noted on a number of occasions.
Secondly, it is also difficult to say whether advance polling days allow voters who would not otherwise have voted to vote, or whether they simply result in those voters who would have voted anyway voting on a different day. It is very difficult to say with any certainty.
One of the problems is that we cannot really hold dry runs to try out different approaches. We simply have to implement them, either at a by-election or at the general election, and see whether they do actually bring about the desired effect.
Perhaps the committee would like to invite Mr. Blais to appear before you to explain his study and his methodology.