In research policy it is difficult to know which exploratory path will be more fruitful or which researchers will make a breakthrough.
Before 2004, for example, the publication and citation records of Andre Geim and Konstantin Novoselov gave no hint that they were about to discover graphene, resulting in thousands of citations a year and a Nobel prize. Many breakthroughs cannot be predicted and come from researchers who do not have an exceptional record.
Handling risk is an important task for funding agencies. They must decide whether it is more effective to give large grants to a few elite researchers or small grants to many researchers. The evidence, including a study based on data from the national research council of Canada, shows no correlation between grant size and citation impact, suggesting that larger grants do not lead to larger discoveries and that funders might therefore do better to target diversity rather than narrowly defined excellence.