Home › How we work
How we work
Where our data comes from, how we check it, and how our next-round predictions are made and tested.
Last updated 28 Oct 2026
Where the data comes from
All closing ranks come from the official MCC NEET PG counselling results (round-wise allotment results), published on the MCC website. We compile them into one table: one row for each college, course, quota, category and state, with the closing rank for every round of every year we cover.
Fees, stipend, bond years, bond penalty and hospital beds are taken from the information published for each seat. They can change, so always confirm with the college before joining.
What a closing rank means here
A seat’s closing rank for a round is the All India Rank of the last candidate holding that seat at the end of that round. This includes candidates who kept the seat from an earlier round, not only new allotments.
- A later (higher) closing rank means the seat was easier to get.
- An earlier (lower) closing rank means it was tougher.
- A dash (–) means no seat was allotted in that round, or the round did not happen.
- Virtual seats shown in brackets in some results are not counted yet.
How we check the data
Every time new data is added, an automatic check runs before anything reaches the website. It stops the update if it finds duplicate rows, missing rounds, empty result columns or impossible ranks, and it flags unusual jumps for a person to review. If the check fails, the website keeps the last good data.
How the next-round prediction works
Our prediction has three layers:
- Overall movement. How much the whole round tends to move from one round to the next, learned from past years (newer years count more). Until MCC publishes the next seat matrix, this is uncertain, so ranges are wider.
- Seat-level movement. A statistical model learns which seats move more or less than average, using the branch, state, quota, category, rank, position within its branch, fee, stipend, bond and how the same seat moved last year.
- Honest ranges. Each seat gets a likely range, not a single number, and a confidence label (High, Medium, Low). Ranges are tuned so that about 8 in 10 real results fall inside them in past tests.
Predictions are labelled provisional until the next seat matrix is released.
How accurate it is
The tables below are updated automatically. “Past tests” predict a round we already know, using only the data available before it. “Real rounds” compare our published prediction with the actual MCC result.
| Past test: 2025 R1→R2 | Typical error | Within ±10% | Within ±20% | Inside our range |
|---|---|---|---|---|
| Assume no change | 12.2% | 43.3% | 71.3% | – |
| Last year’s overall movement | 7.8% | 56.6% | 80.5% | – |
| CutoffBridge model | 9.1% | 54.3% | 83.7% | 76% |
| Past test: 2025 R2→R3 | Typical error | Within ±10% | Within ±20% | Inside our range |
|---|---|---|---|---|
| Assume no change | 9.8% | 50.6% | 77% | – |
| Last year’s overall movement | 7% | 59% | 82.1% | – |
| CutoffBridge model | 8.1% | 57.8% | 84.7% | 75.7% |
Updated 2026-10-28.
Limitations
- Predictions describe likely ranges based on past patterns. They are not guarantees, and this year can differ.
- Big changes in the seat matrix, new colleges or rule changes can move cutoffs in ways past data cannot show.
- We currently cover MCC counselling only. State quota counselling is separate and not included yet.
- Mistakes can happen. If you find one, tell us and we will fix it in our corrections log.