If I was a fraud this is exactly what I would do. Take an existing paper and just make up the results that confirm the work with a slight twist to make it publishable. Rinse and repeat and nobody will ever catch you.