Cracking the Match: Everything a current (or future) residency applicant needs to know about the Match algorithm

Few buzzwords are more unnerving to a medical student than mentioning “The Match.” Other items on the dreadful shortlist might be “board exams,” “night shift,” or “student loans.” These fearsome notions, while menacing, are far less nebulous than the omnipotent algorithm that dictates the fate of student-physicians with a single keystroke.

But fear not! This article includes all the necessary backstory, quips, bridges-over-pitfalls, and empowering strategies that a current or soon-to-be residency applicant might need to be well-informed when it comes to the mechanics of The Match.

First, why does The Match algorithm exist?

The National Resident Matching Program (NRMP) Match algorithm is not a new implementation. The first iteration of a Match algorithm was put in place in 1952. Prior to 1952, the void between student doctorhood and residency was somewhat of a jungle. Hospitals wanted to fill spots early, while applicants benefited by delaying.

This resulted in a combatant back-and-forth between the two, which produced policies and behaviors that were counterproductive. This included medical schools holding transcripts and letters of recommendation until the fourth year of medical school, residency programs offering residency positions as far as two years before residency began, and hospitals calling and issuing “exploding” offers, where student doctors had to accept or decline a residency offer before hanging up the phone.

Enter: the algorithm. The NRMP Match algorithm is a step up from that chaos. The algorithm and its appendages have continued to evolve over the past 70 years to what is used today.

One of the most notable evolutions in the process occurred in the late 1990s. This was when the algorithm was changed from a “program-proposing” system to an “applicant-proposing” system. In short, this change put the applicant’s desires over the program’s desires in the inner workings of the algorithm (more on this in just a moment). The economists who developed this algorithm received the Nobel Prize award for Economics in 2012.

But how does the algorithm actually work?

Well, the NRMP has a wonderful explanation here, but here are the basics: the algorithm being applicant-proposing means that the applicant’s desires are prioritized over the program’s desires. The algorithm will attempt to match an applicant into their top program if at all possible. Here’s a scenario to illustrate:

Backstory: Imagine applicants are eager theatergoers. Each residency program is a specific movie theater screen showing the movie of their residency training experience. The algorithm is a theater usher hoping to please all patrons.

Step 1: The algorithm starts off by putting an applicant in line at the program of their first choice.

Step 2: The algorithm continues down the row, placing all applicants in line at their first choice. The order of the applicants in line is dictated by the order in which programs placed the applicants – first choices in front.

Step 3: Of course, programs do not have an infinite number of seats in each theater. Once the line at a particular program exceeds the number of vacancies, the applicants past that point will be swept away and the algorithm will now place those applicants in line at the program they chose second. Obviously, the applicants before that point have a confirmed match and will be ushered into the screen for which they are in line.

Step 4: This continues until all options are exhausted for every applicant.

What does this mean for applicants?

The simplicity of the system is commonly lost to applicants who are accustomed to studying complex physiological systems with balances, counterbalances, fail-safes, and counter fail-safes. Furthermore, deep in the bowels of forums, applicants find whispers and secrets on how to trick the machine into doing their bidding. Naturally, the crushing pressure placed on applicants can squeeze their after-hours searching into places where warped truths and anecdotal examples seem like feasible ideas.

The truth is that the algorithm has been studied extensively. Investigators have even tried to slither through the perceived cracks in the NRMP armor. The salient fact that rings true every time is simply this: When ranking programs for the NRMP Match, applicants should rank programs according to their true and honest desires. Even if you think you might be at the very back of the line for your dream program, the algorithm does not penalize you for giving it a shot.

Additionally, the following are some important guidelines to consider when applicants are thinking through their rank list:

Guideline 1: It is worth repeating: Applicants should put together their rank list in the order of their true preference.

Guideline 2: Participating in The Match is a binding contract. If an applicant is matched to a program, they are obligated to start residency at that program. Consequently, applicants should only rank programs where they would be comfortable training for residency.

Guideline 3: Applicants should be realistic about their competitiveness. The NRMP has objective resources that can be used as a barometer for this insight. Applicants should also use the resources available to them at their respective medical schools.

Guideline 4: Generally, applicants who have longer rank lists have a greater likelihood of matching. Certainly consider guideline 2, but applicants should know that shorter lists mean fewer lines to “stand” in at the end of the day.

If you need additional information, The DO has tips on preparing your rank list during a pandemic here. The NRMP also has a wealth of articles, videos, explanations, and resources available here. This includes information on what might occur if an applicant is not placed by The Match algorithm. The DO has previously covered what to do if you don’t match as well. This is a rare but possible occurrence that all applicants should be informed of and prepared for.

The aforementioned investigators trying to shed light on potential vulnerabilities in the NRMP algorithm said it best: “If programs and students can accept a degree of uncertainty as the price of optimum performance and simply behave in their own best interests, they will soon realize that the ‘game’ they are playing is more solitaire than poker.”