How far down do programs go on their rank list?

[AlGamaty]

Hello,

I'm applying to IM and I'm really interested in matching with my #1 program (it's the only IV from the city my family are in). The past few days I've been trying to analyze my likelihood of matching with them.

Based on table 17 in this NRMP report, the average categorical IM program in 2022 needed to rank 7.3 applicants to fill 1 position. My #1 program has 20 PGY-1 positions, so 20*7.3 = 146. So they should technically fall to around the 146th ranked applicant in order to fill all 20 spots. And last year they only gave interviews to 200 applicants.

So am I right in assuming that I only need to be ranked in the top 146 out of 200 applicants in order to match with them? Or are there other considerations at play that I'm missing? Keep in mind that this is not a super competitive program by any means. If anything, I would say they are average or a bit below average in terms of desirability (they're highly desirable to me though because of location).

I would appreciate any input on this. Thanks!

---

Members don't see this ad.

 

[Dral]

That is not a correct assumption. They technically only need to rank 20 people to fill all their spots. We know this isn't how it will work out, but it helps demonstrate the point that things can change from year to year. They may only go down to rank 70 this year for instance. I can tell you things change from year to year for my program (not IM) in terms of how far down our list we go.

When communicating with programs, some will tell candidates that they are 'ranked to match'. Then the person finds out they didn't match there. It's because programs will use the term 'ranked to match' as 'you are ranked within the number of people we usually have to go through to fill our program'. But that year, the program does not get down their list as far as usual and the person doesn't match there. Again, the only way for you to be absolutely guaranteed to match at your target program would be if you ranked them 1 and they ranked you within their top 20. If they ranked you at 40, it's probably a good chance you'd match. 70? Probably still good chance. 130? I'm not typically a betting man, but I wouldn't want to put a bunch of money on it (though still possible to match at that rank based on your math).

 

[DOVinciRobot]

Incorrect assumption. You need to be ranked in the top 20 to be guaranteed to match with them. There is no way to remotely analyze your chances outside of that.

 

[NotAProgDirector]

Your math is "correct". But your conclusion is shaky.

Yes, the average ranks to fill a slot in IM was 7.3. So, on average, a program with 20 slots would need 146 ranks to fill.

The problem is that you have no idea whether this program is "average" or not. Even if you think they are not as competitive, maybe they specifically interview less competitive applicants such that they fill higher on their list.

There's also the interesting question of what an "average" means in this situation. I assume they are referring to an arithmetic mean. But the distribution of ranks to fill is unlikely to be normal. The lowest possible value is 1, where the highest possible value is unbounded (although practically can't be higher than 40-50). So the distribution is likely to have a longer tail to the right, so the arithmetic mean may be greater than the median -- hence not represent the 50% percentile (which would be less).

So the bottom line is that there's no way to answer your question. And knowing the answer won't really change anything.

All that said, there's this sense on SDN that if you're not in the top 20 ranks at a program you're not going to match. For large programs like IM, that's just ridiculous. Programs don't want to waste their time interviwing gobs of students for no gain. Programs with sizable numbers of spots will want to fill 75% of the way down their list -- this gives them plenty of cushion to stay out of SOAP, yet doesn't waste lots of time/resources interviewing people.