How the predictor works
Last updated 8 October 2026.
You type your NEET-PG All India Rank (AIR) and your category. The predictor lists every seat that was allotted to someone at or after your rank in earlier counselling, and tells you how comfortably you clear each one. This page explains each step, so you know how far to trust it.
Where the data comes from
The main data is the seat-allotment result PDFs that the Medical Counselling Committee (MCC) publishes at mcc.nic.in for every round of 2023, 2024 and 2025: Round 1, Round 2, Round 3 and the stray vacancy rounds. We parse each PDF row by row and keep the AIR, the college, the course, the quota and the category of the seat. The ranks come from the NEET-PG result published by NBEMS.
State quota is a beta feature for eight states: Andhra Pradesh (Dr NTR University of Health Sciences), Bihar (BCECEB), Gujarat (ACPPGMEC), Karnataka (KEA), Kerala (CEE Kerala), Madhya Pradesh (DME MP), Maharashtra (State CET Cell) and Odisha (DMET Odisha). Their allotment lists are public. Only lists that print the NEET AIR are used as ranked seats. See All India vs state quota for the background.
What a closing rank is
For each seat group (one college, course, quota and seat category) and each year, the closing rank is the highest AIR number allotted in a round. A closing rank of 12,400 in Round 1 means the last person given that seat in Round 1 had AIR 12,400. If your rank is better (a smaller number) than that, someone with a worse rank than yours got the seat, so you would likely have got it too.
What Safe, Likely and Possible mean
The page looks at the closing ranks of the years that have data and takes the lowest (the hardest year) and the highest (the easiest year).
- Safe: your rank is at least 20% better than the hardest year's closing rank. That is, your rank is no more than 0.8 times that number.
- Likely: your rank is within the hardest year's closing rank, but not by a wide margin.
- Possible: the seat closed before your rank in at least one year, but went past it in another. Your rank sits between the hardest and the easiest year.
- Just out of reach: the seat closed before your rank in every year, but within 5% of it in some year.
- Opens in later rounds: the seat did not reach your rank in the round you are viewing (Round 1 by default), but a later round went past it. You can switch the view to Round 2, the final round, or any round.
A seat with only one year of data is never called Safe, because one year is thin. It is capped at Likely and carries a "1 year of data" or "New" tag.
What "Tight +N" means
On a Safe or Likely seat, the tag shows how many ranks of room you have in the hardest year. "Tight: +180" means that in the year the seat closed earliest, it went only 180 ranks past yours. We show the tag when the room is under 500 ranks.
Category logic
You see Open seats plus seats of your own category (OBC, EWS, SC or ST). Reserved candidates can also take Open seats on merit, so both lists apply. The comparison is always your AIR against the AIR of the people who got the seat. We do not use category rank. If you have a PwD certificate you also see PwD seats; otherwise those seats stay hidden. Seats with eligibility limits, such as NRI, minority, Armed Forces and central university institutional seats, show only if you say you qualify.
For state quota you pick a home state and its own categories. State reservation codes differ from MCC's, so each state's categories are listed as printed by the state.
What the predictor cannot do
- History is not a guarantee. Seat numbers, candidate numbers and cut-offs change every year, and the seat matrix changes too: new seats never appear in the history.
- In-service seats get no label. They are filled on service marks, not AIR, so the page lists where and when they went but never calls them Safe or Likely.
- Later rounds are only seats allotted in that round. A Round 2 closing rank is the worst rank that received a seat in Round 2, not the worst rank holding the seat afterwards.
Fees, bonds and the resident survey
Annual tuition on private seats comes from published college fee statements gathered by hand and shown with the year they apply to. We only use a figure when it clearly states one annual amount; ranges are left out. Stipends are the amounts colleges declare to MCC. Service bonds follow the seat, so check the college's notice and the current MCC bulletin.
The resident survey is a beta. It is The Nachiket Bhatia Show's anonymous survey of residents, rating departments on things like faculty, hands-on work, hours and stipend regularity. The responses are not verified, and the PDF prints no response count, so we show a minimum count inferred from the averages. Treat a rating built on a handful of replies as a rumour, not a ranking.
Official sources
For schedules, rules and seat matrices, use the official sites: mcc.nic.in, natboard.edu.in, and the state authority for your state, which is linked from each state's page in the app.