NSF GRFP

I am not sure why you would listen to a horse ( That is, of course, unless the horse is the famous Mr. Ed.) but here is the NSF explanation of Intellectual Merit. The Project Description must contain, as a separate section within the narrative, a section labeled "Intellectual Merit."  The Project Description should provide a clear statement of the work to be undertaken and must include the objectives for the period of the proposed work and expected significance; the relationship of this work to the present state of knowledge in the field, as well as to work in progress by the PI under other support. The Project Description should outline the general plan of work, including the broad design of activities to be undertaken, and, where appropriate, provide a clear description of experimental methods and procedures. Proposers should address what they want to do, why they want to do it, how they plan to do it, how they will know if they succeed, and what benefits could accrue

The selection of a reference(s) to stand up for you in your application is a challenging prospect.  You want someone that knows you (well) but also someone with some gravitas.  That isn't an easy thing to determine, but title and reputation count especially to reviewers who will mostly be faculty at other colleges and universities.  The temptation is to pick someone you know will say something nice but they might be relatively inexperienced in writing letters or not have a well-known reputation.  On the other hand you could go after a Nobel Laureate but it might be someone who doesn't really know you.  Most important is the confidence you have in this person and their willingness to get the job done.  To paraphrase Wayne Gretzky, you miss 100% of the references that don't get submitted.  Make sure you get started early and have your references lined up in advance.  Don't play games and ask for the letters to be done one month in advance of the deadline.  On the other ha