There is an understanding that if a party is number 1 on the senate ticket they will pick up extra votes from people who aren't serious and just number from the left. Has this been studied and if so what is the estimated benefit to the party.
The Wikipedia article suggests that it's a percentage point or two, without citing evidence that I can see.
This ABC article offers evidence that it increases with the length of the ballot paper, with electronic voting and further with internet voting, and cites a few examples where it has exceeded 4%.
This ELJ article cites a 1963 study that it's 3% and that the authors later reduced their estimate, and that candidates believe the effect is ten times this size.
All of the above had a theme that of the many impacts of ballot order on election outcomes, the most significant is which of the leading candidates appears first.