How do you make the "vaccines" look more effective?

And at the same time make it easier to blame the unvaccinated?

You lie, of course. You make sure to not just omit the two weeks after a jab from the “vaccinated” column, you move them into the “unvaccinated” column. Gateway Pundit has a nice writeup on this, quoting others who have pulled the data from Alberta:

https://www.thegatewaypundit.com/2022/01/alberta-canada-inadvertently-published-quickly-deleted-health-data-exposing-half-vaccinated-deaths-counted-unvaccinated/

Then of course, after someone publicizes your incriminating data, you must then hide that data from public view. Too late now.

BTW, this is not the only data set showing negative efficacy early on after the jabs.