That would be the implication of this research study, which makes a fairly simple statistical point.  We already know that papers are more likely to be published if they include a positive result (they reject the null).  Less politically correct to admit is that when the data gets messy, researchers tend to make a lot of choices that lead to rejection of the null, like eliminating outliers, redefining variables, and the like.  Whether you call that ‘increasing power’, ‘motivated reasoning’ or something worse, it leads to a number of papers being published with results that are not likely to be replicated.  And papers that don’t replicate the positive results don’t get published, so we don’t read about them.

Now, what happens when an area (like fair value) gets hot?  There is a lot more work being done, so the bias toward positive results results in even more nonreplicable publications.  Even worse, researchers have greater incentives to publish in a hot area, making power-increasing analyses even more likely.

By the way, the study is in chemistry, not exactly a ‘soft’ science.