*maybe* for the postdocs, but if you're a grad student just learning the techniques and research methods, there's decent odds that you don't have the skills to spot that yet. And if something fails to replicate, as discussed in the article...

...most of the time it *is* going to be something with the method or experiment not working. Fraud happens, but not so much that it's going to be someone's first assumption, especially a grad student. And honestly I don't think it *should* be anyone's first assumption.