Last updated: 8/31/2026· Written by the MatchPal Advising Team
Quick answer: For years, applicants believed competitive specialties required 20 to 37 “research items,” a number that came from NRMP reporting the mean of combined abstracts, presentations, and publications. The 2026 Charting Outcomes reports changed two things: they report the median instead of the mean, and they split research into separate categories. The result: the median number of publications for matched applicants is far lower than the old figures suggested (neurosurgery about 12, dermatology about 9, orthopaedics about 7), and the overall median for matching into any specialty is just 1 to 2 publications. Research still matters, especially for competitive fields, but the quantity you actually need is well below what the old averages implied, and for most specialties research output barely differed between matched and unmatched applicants.
The value of research in the Match is one of the most debated topics in medical education. For years, the conversation was driven by a set of intimidating numbers, and in 2026, those numbers quietly changed. If you have ever felt behind because you weren’t sitting on a stack of publications, this is for you.
Historically, NRMP reported the mean of a single combined figure: abstracts, presentations, and publications lumped together. Those means looked enormous. Roughly 37 for neurosurgery, 27.7 for dermatology, 23.8 for orthopaedic surgery. The takeaway most applicants absorbed was simple and wrong: if you’re not at 15-plus research items, you’re behind.
That framing did real damage, especially to students without a home research department. They either spent egregious amounts of time hunting for projects or took an entire research year, all to chase a number that was never what it appeared to be.
A mean is easily distorted by a handful of outliers. One applicant with 60 publications drags the average up for everyone, so the reported number describes almost nobody. It tells you what the extremes look like, not what a typical successful applicant looks like.
In 2026, NRMP changed two things about the way they report research experience: it reported the median, and it separated the old combined figure into distinct categories (research experiences, abstracts, presentations, publications). When you look at the median number of publications reported in the NRMP Charting Outcomes report for matched applicants, the picture is dramatically different:
- Neurosurgery: about 12
- Dermatology: about 9
- Orthopaedic surgery: about 7
And the median for a student matching into any specialty at all is just 1 to 2 publications (2 for MD seniors, 1 for DO seniors). The bar was never as high as the averages made it look.
![A data table titled "Mean vs Median: Research Requirements Shift" comparing historical mean research items to 2026 median publication counts for matched residency applicants. The table includes the following data points: Neurosurgery: Previous Mean of 37; New Median of 12. Dermatology: Previous Mean of 27.7; New Median of 9. Orthopaedic Surgery: Previous Mean of 23.8; New Median of 7. Overall (Any): Previous Mean of [N/A]; New Median of 1 to 2. A subtitle notes the data reflects the research output of matched applicants only. A footnote explains that historical means included all research items (abstracts, presentations, publications) and were skewed by high outliers, whereas the new medians are a stable measure counting only publications. The source is listed as the NRMP 2026 Charting Outcomes Report.](https://matchpalmedical.com/wp-content/uploads/2026/08/2026-NRMP-mean-to-median-shift-1024x683.jpg)
- Research still matters for NRMP. It remains a real way to distinguish yourself, likely valued above work and volunteer experience and below academic performance, and it matters most in competitive specialties. Keep pursuing it.
- You need far less than the old numbers implied. The quantity of research required to match is a lot lower across the board than the historic means suggested.
- For most specialties, research output barely separated matched from unmatched applicants. Read that again. It is unlikely that one or two extra publications are the reason one applicant was ranked over another, all else equal.
If you’re targeting a competitive specialty in 2026, this is not permission to ignore research. It absolutely still counts. But if you already have a decent foundation, the next hour is often better invested elsewhere: a personal statement that earns interviews, away rotations that turn into strong letters, and a program rank list and signal strategy built for your specialty. Beyond a point, another publication yields less than a sharper application.
The applicants who match competitive specialties are rarely the ones with the single highest publication count. They are the ones who built a complete, coherent application and knew where their time was best spent.
(A note on the numbers: 2026 figures are medians, the pre-2026 combined “research items” figure was a mean, so old and new numbers are not directly comparable, and a few specialties do still show a meaningful matched-vs-unmatched research gap.)
Knowing how much research you actually need is the easy part. Building the application that turns your research, your scores, and your story into interviews is the hard part, and it’s exactly what our team does: 80+ resident physicians who recently matched your specialty and know where your time is best spent this cycle.
Built by residents. For the students who are next.
How many publications do you need to match residency?
The 2026 median for matching into any specialty is 1 to 2 publications. For competitive specialties the median is higher (neurosurgery about 12, dermatology about 9, orthopaedics about 7), but far below the 20-plus figures older averages implied.
Why did the NRMP research numbers drop so much in 2026?
Two reasons. NRMP switched from reporting the mean (easily inflated by outliers) to the median, and it split the old combined “abstracts, presentations, and publications” figure into separate categories.
Does research still matter for the Match?
Yes, especially for competitive specialties. It is likely valued above work and volunteer experience and below academic performance. It just takes less than the old numbers suggested.
Is more research always better when it comes to residency applications?
Not necessarily. For most specialties, research output did not differ dramatically between matched and unmatched applicants. Once you have a solid foundation, additional time is often better spent on your personal statement, aways, letters, and program strategy.
