Have some portion of academic funding distributed by assigning a pool of money to grantors and having them bet portions of that money on papers they consider impactful AND likely to replicate. Allow the authors to use the money on any research they want.
This lets authors get some research money without writing costly and wasteful grant proposals. It takes advantage of the fact that experts in the field can generally tell which studies are likely to replicate (I'm assuming granting agencies can find experts in the field to do this).
Have some portion of academic funding distributed by assigning a pool of money to grantors and having them bet portions of that money on papers they consider impactful AND likely to replicate. Allow the authors to use the money on any research they want.
This lets authors get some research money without writing costly and wasteful grant proposals. It takes advantage of the fact that experts in the field can generally tell which studies are likely to replicate (I'm assuming granting agencies can find experts in the field to do this).