Skip to content

NSF MSPRF Guide

Updated 6 official sources

Quick Answer: NSF Mathematical Sciences Postdoctoral Research Fellowships eligibility turns on 7 checks: candidate status, math or statistics fit, host scientist, fellowship option, one-proposal rule, prior PI boundary, and the next deadline, October 20, 2027.

Full answer and how we checked it

NSF Mathematical Sciences Postdoctoral Research Fellowships eligibility turns on 7 checks: candidate status, math or statistics fit, host scientist, fellowship option, one-proposal rule, prior PI boundary, and the next deadline, October 20, 2027.

Who this is for: Recent mathematics and statistics PhDs, sponsoring scientists, host departments, and sponsored-research offices deciding whether MSPRF is the right postdoctoral fellowship route.

Data note: This guide uses the official NSF MSPRF page (last updated by NSF October 5, 2026), the NSF 23-603 solicitation (opens in a new tab), Grants.gov listing 349280, the NSF PAPPG (NSF 24-1 with Supplement 1, NSF 26-200, and Supplement 2, NSF 26-202), and NSF Award Search records, rechecked October 9, 2026, with eligibility rechecked October 13, 2026 and the deadline October 24, 2026.

In this guide

According to NSF Mathematical Sciences Postdoctoral Research Fellowships (opens in a new tab), MSPRF supports postdoctoral fellows pursuing research and training in mathematics and statistics, including applications to other disciplines. NSF lists NSF 23-603, the next deadline as October 20, 2027, and an estimated 35 to 45 awards.

The real question is whether the applicant is a fit for a fellowship submitted by the individual, with a host scientist and a research-training plan that belongs in mathematics or statistics. A strong research idea is not enough if candidate or host rules fail.

Use this guide with the live Grantsights record. Then compare NSF Geometric Analysis guide, NSF Probability guide, NSF grants complete guide, what is a NOFO, and the Grantsights grant database.

From Manusights, our sister product

Writing an NSF proposal? Manusights offers an AI critique of your proposal against the review criteria in your solicitation: $149 for a research proposal, $49 for a GRFP or other fellowship application. Back within 3 business days of your upload.

Order a grant review

7 NSF Mathematical Sciences Postdoctoral Research Fellowships Eligibility Stop Checks

Stop checkStrong MSPRF signalRed flag
Candidate statusIndividual candidate fits the rulesCandidate is outside eligibility
Research areaMathematics or statistics is centralProject belongs outside DMS
Host scientistMentor adds clear research trainingHost role is vague
Option choiceResearch Fellowship or Instructorship fitsOption is chosen late
Prior PI boundaryPrior award history is cleanPrior PI status creates risk
Submission limitOne proposal per individual per year is respectedDuplicate plan exists
TimingOctober 20, 2027 deadline is feasibleDocuments are late

Postdoc applicants sometimes focus on the theorem and miss the fellowship mechanics. MSPRF needs both.

Get research funding calls by email

About once a week, we email open research funding calls with a real deadline, from federal agencies such as NIH and NSF, picked for research institutions. One click to unsubscribe.

No spam. Unsubscribe anytime.

5 Source Facts Applicants Should Know

According to NSF, MSPRF supports research and training in mathematics and statistics, including applications to other disciplines. NSF says there are 2 options: Research Fellowship and Research Instructorship.

According to NSF, the 2026 cycle closed October 21, 2026, the next deadline is October 20, 2027 at 5 p.m. submitter's local time, and the annual deadline is the third Wednesday in October. NSF lists an estimated 35 to 45 awards, subject to availability of funds. The solicitation says applicants can expect an email decision in late January or early February, and the requested start date must fall between June 1 and October 1 of the award year.

According to USAspending.gov (opens in a new tab) data, assistance listing 47.049 had about $1.60B in FY2024 obligations across 5,546 awards active that year.

FactSource-backed read
SolicitationNSF 23-603
DirectorateMathematical and Physical Sciences
DivisionMathematical Sciences
Next deadlineOctober 20, 2027 (2026 cycle closed October 21, 2026); third Wednesday in October in later years
Live listingGrants.gov listing 349280

Eligibility Rules That Return Proposals Without Review

The NSF 23-603 solicitation (opens in a new tab) lists seven conditions, and NSF returns a proposal without review if any one fails. The ones that trip people up:

  • Citizenship: you must be a U.S. citizen, U.S. national, or permanent resident at the time of submission, not at the start date.
  • Degree clock: you can't have held the doctorate more than 2 years as of January 1 of the award year. For an award made in 2027, that means a degree conferred on or after January 1, 2025.
  • Prior NSF role: you can't have been PI or co-PI on an NSF award, except a Graduate Research Fellowship or a conference or workshop award.
  • One shot per offer: you can't apply if MSPRF has previously offered you an award, and you can't duplicate the research plan in another NSF proposal.
  • One proposal per deadline: you submit it yourself in Research.gov as an individual, not through a sponsored projects office. Hosts have no limit on how many fellows they take.

What the Fellowship Pays and How It Runs

According to NSF 23-603, every award runs 48 months and pays 24 months of stipend: 18 academic-year months plus 6 summer months. The full-time stipend is $5,833 a month (up to $70,000 a year), and half-time is $2,917. A separate $50,000 allowance arrives as $30,000 in year one and $20,000 in year two for research costs or fringe benefits such as health insurance, retirement, dependent care, and moving.

OptionSupport patternWho it suits
Research FellowshipAny 18 academic-year months of full-time support within 3 years, in blocks of at least 3 consecutive monthsFellows who want maximum research time and flexibility to move between sites
Research InstructorshipUsually 1 academic year full-time, then 2 years half-time; NSF says the half-time years give the Fellow teaching experienceFellows who want teaching on the CV

Up to 2 months of paid parental or family leave fits inside the 48 months. The doctorate has to be conferred before the appointment starts. You can switch to the Instructorship option before the appointment starts, with NSF approval.

Every fellowship is the same size. Grantsights analysis of program element 060Y00 in NSF Award Search (opens in a new tab), pulled October 9, 2026, shows 42 MSPRF awards made in 2025 and 30 so far in 2026, every one for $190,000. The 2026 titles run from algebraic geometry and number theory to spin glasses, scientific computing, and statistical theory for machine learning, so applied and statistics projects do get funded alongside pure mathematics.

How Panels Judge MSPRF Proposals

A panel of mathematical scientists applies NSF's two standard criteria plus solicitation-specific ones: your ability and potential as shown by past research and reference letters, the suitability and availability of the sponsoring scientist and colleagues, the host's space and services, the likely effect of the sponsor and host on your development, the quality of research likely to come out, and your potential contribution to NSF's education goals. The narrative is capped at 5 pages and must justify the choice of sponsor and host. The sponsor's statement is capped at 3 pages. You need 3 reference letters (up to 4 allowed), and one must come from your doctoral adviser. No written reviews come back, only a panel summary with ratings and your relative rank.

What NSF Doesn't Tell You

Within 21 days, answer 4 practical questions. First, fellowship fit is not the same as project quality. The candidate, host, prior award history, option choice, and training plan all matter.

Second, the host scientist is not a formality. A candidate moving to a host with a clear new method and seminar environment can show training gain. A candidate staying with the dissertation group with no new research setting needs a stronger case, because the panel scores the host's effect on your development.

Third, the Research Fellowship and Research Instructorship options should be chosen for strategic reasons, not because one sounds better.

Fourth, NSF's reference-letter writers need their own Research.gov accounts to submit through the Reference Letter Submission module. Send invitations early; a missing letter is your problem, not the writer's.

6 Evidence Assets That Make This Proposal Stronger

Within 21 days, build 6 evidence assets. Asset 1 is the candidate-fit memo. State degree timing, citizenship or residency fit if applicable, prior submission history, and prior PI status.

Asset 2 is the research-area memo. Explain why the project belongs in mathematics or statistics, including applications if relevant.

Asset 3 is the host-scientist memo. Identify the sponsoring scientist, mentoring plan, research environment, seminars, collaborators, and training gain.

Asset 4 is the option memo. Compare Research Fellowship and Research Instructorship, then explain the chosen route.

Asset 5 is the adjacent-route map. Compare MSPRF with AAPF, GRFP, CAREER, DMS research programs, AAG, institutional postdocs, and non-NSF routes.

Asset 6 is the submission checklist. Include statements, forms, host documents, references, research plan, training plan, and PAPPG checks.

5 Route Choices Before Drafting

Within 21 days, compare 5 routes.

RouteBetter first read when...Why
MSPRFIndividual math or statistics postdoc route fitsFellowship route fits
AAPFAstronomy or astrophysics dominatesDifferent fellowship fits
DMS research programPI-led project is strongerGrant route may fit
CAREERCandidate is faculty and eligibleEarly-career faculty route fits
Institutional postdocMSPRF rules do not fitLower rule risk

The strongest MSPRF applications make the candidate, host, research area, and option choice align.

6 Bad-Fit Signals

First, the candidate is not eligible. Second, the research is not in mathematics or statistics. Third, prior PI status creates a rule problem.

Fourth, the host scientist role is vague. Fifth, the option choice is unexplained. Sixth, the October 20, 2027 deadline will be missed because documents are late.

These signals point to route risk. They should be addressed before drafting.

21-Day MSPRF Readiness Plan

Days 1 through 3 should confirm candidate eligibility, host scientist, prior submission history, prior PI status, and deadline. Days 4 through 7 should write the candidate-fit and research-area memos.

Days 8 through 12 should build the host-scientist and option memos. Days 13 through 16 should compare MSPRF with AAPF, DMS research programs, CAREER, and institutional postdocs. Days 17 through 21 should finish PAPPG review, forms, host materials, references, and final route proof.

6 Final Research Checks Before Submission

Before the October 20, 2027 deadline, finish 6 final checks. Check 1 is the MSPRF program page, which NSF last updated October 5, 2026. Check 2 is NSF PAPPG (opens in a new tab) compliance. Check 3 is NSF 23-603 fit. Check 4 is the October 20, 2027 deadline for the next cycle. Check 5 is host-scientist proof. Check 6 is option-choice proof.

You're ready when the candidate can state eligibility, research area, host reason, option choice, and deadline readiness in 5 direct sentences.

4 Decision Questions Before You Start

Within 21 days, ask 4 decision questions. Question 1: "Is the candidate eligible?" Question 2: "Does the project belong in mathematics or statistics?" Question 3: "Does the host scientist add training value?" Question 4: "Which fellowship option fits?"

If question 1 fails, stop. A strong project doesn't fix a rule problem, since NSF returns ineligible proposals without review.

3 Last-Mile Checks For MSPRF Applicants

Within 21 days, run 3 last-mile checks. Check 1 is candidate proof. Check 2 is host proof. Check 3 is option proof.

If those checks are weak, a host rewrite or route review may save the application. A final note for applicants: write the fellowship fit sentence before the abstract. In 2 sentences, state the research area, host reason, option choice, and reason MSPRF is the best NSF home.

Sources

Award data for this program

Frequently Asked Questions

What does NSF MSPRF support?

It supports postdoctoral fellows pursuing research and training in mathematics and statistics, including applications to other disciplines.

What solicitation applies?

NSF lists NSF 23-603 for Mathematical Sciences Postdoctoral Research Fellowships.

What is the next deadline?

The 2026 cycle closed October 21, 2026. NSF now lists October 20, 2027, by 5 p.m. submitter's local time, and the third Wednesday in October each year after that.

How many awards are expected?

NSF lists an estimated 35 to 45 awards, subject to availability of funds.

Who should skip this route?

Skip it if you aren't a U.S. citizen, national, or permanent resident, if you'll have held the doctorate more than 2 years by January 1 of the award year, if you've been PI or co-PI on an NSF research award, or if MSPRF has already offered you an award.

Last updated October 24, 2026. Corrected October 24, 2026: what changed. Published by Grantsights from public sources. See our editorial policy, how we build and check guides, corrections and how we make money.