If an IPO's retail category is subscribed 14 times, your chance of getting an allotment is roughly 1 in 14 — about 7%. That reasoning is the basis of virtually every “IPO allotment probability calculator” online, and for the retail category it is close to correct. But applied to the large HNI (bNII) category, the same formula understates the true odds by roughly five times. The reason is not a modelling choice — it is a specific provision in SEBI's allotment rules that the simple formula ignores.
1. What SEBI actually requires
Two provisions do the work here. Both come from SEBI's own Frequently Asked Questions on the Issue of Capital and Disclosure Requirements (ICDR) Regulations, published May 2025:
“It is to be noted that there is no discretion in the allotment process. Further, the allotment of specified securities to applicants other than to the retail individual investors, non-institutional investors and anchor investors shall be on a proportionate basis within the respective investor categories.”
“The allotment to each retail individual investor and NIIs shall not be less than the minimum bid lot, subject to availability of shares in retail individual investor category, and the remaining available shares, if any, shall be allotted on a proportionate basis.”
That second sentence is the whole story. Allotment in the lottery categories is not proportional to how much you bid — the regulation caps what a single applicant receives at the minimum bid lot whenever there are more applicants than lots to go round. Bidding for ten lots does not give you ten times the chance; it gives you the same one lot as everybody else, if you win at all.
2. The category structure
For a book-built issue under the profitability route, SEBI sets these floors and ceilings on the net offer to the public:
| Category | Share of net offer | Application size | How allotment is decided |
|---|---|---|---|
| Retail (RII) | Not less than 35% | Up to ₹2 lakh | Minimum bid lot to as many as possible, then draw of lots |
| Small HNI (sNII) | One-third of the NII portion | Above ₹2 lakh, up to ₹10 lakh | Minimum NII lot, then draw of lots |
| Large HNI (bNII) | Two-thirds of the NII portion | Above ₹10 lakh | Minimum NII lot, then draw of lots |
| QIB | Not more than 50% | Institutional | Proportionate — not a lottery |
The NII category as a whole must be at least 15% of the net offer, and it is split one-third to small HNI and two-thirds to large HNI by application size — a sub-categorisation introduced by amendment to the ICDR Regulations in 2022. An unsubscribed portion in either sub-category may be allocated to the other.
Note the different structure for a compulsory book-built issue: at least 75% to QIB, not more than 15% to NII and not more than 10% to retail. Retail odds on those issues are far tighter for this reason alone.
3. Why large HNI breaks the simple formula
Put the bid floor and the allotment unit side by side and the gap becomes obvious:
- Retail — applies in lots, is allotted one lot. Floor and unit match.
- Small HNI — applies above ₹2 lakh, is allotted the minimum ₹2 lakh HNI lot. Floor and unit match.
- Large HNI — must apply above ₹10 lakh, but is allotted the same minimum ₹2 lakh HNI lot. Floor and unit are five times apart.
Subscription times is computed from money bid against shares offered. So when the large HNI category shows 50×, that 50 reflects bids averaging at least ₹10 lakh — but the shares are handed out in ₹2 lakh units. The pool stretches roughly five times further across applicants than the headline number suggests. Dividing 100 by 50 gives 2%; the real figure is nearer 10%.
4. Checking it against a real basis of allotment
This is testable. After an issue closes, the registrar publishes a basis of allotment stating, per category, how many applications were received and how many applicants were actually allotted shares. Dividing one by the other gives the true odds — no modelling involved.
Taking the basis of allotment for Laser Power & Infra (16 July 2026) and comparing the actual outcome against the simple estimate:
| Category | 100 ÷ times | Actual odds | Allottees / applications | Ratio |
|---|---|---|---|---|
| Retail | 14.39% | 16.37% | 1,73,540 / 10,60,040 | 1.14× |
| Small HNI | 2.74% | 2.79% | 1,769 / 63,450 | 1.02× |
| Large HNI | 1.99% | 9.59% | 3,538 / 36,882 | 4.83× |
Small HNI needs essentially no correction (1.02). Retail runs slightly better than the formula (1.14) because a good number of retail investors bid more than the minimum lot, so there are fewer distinct applicants than the money implies. And large HNI comes in at 4.83× — almost exactly the ₹10 lakh ÷ ₹2 lakh ratio the rule predicts.
The same document confirms the mechanism directly rather than by inference: it reports “Equity Shares Allotted per Bidder = 980” for both small HNI and large HNI — the identical allotment unit for two categories with five-fold different bid floors. The 3,538 large HNI allottees are exactly 34,67,290 shares ÷ 980.
5. QIB odds are not a probability at all
Because QIB allotment is proportionate rather than a draw of lots, applying a “chance” to it is a category error. In the same issue, 101 QIB applications received 101 allotments — a 100% allotment rate, in a category reported as many times oversubscribed. Every QIB got shares; they simply got fewer shares each than they bid for. For QIB, 100 ÷ times is a fill rate, never a probability — which is worth knowing if you see a site quoting QIB “allotment chances”.
6. What this means for your application
- Extra lots do not buy extra odds in retail. Above the minimum lot, additional lots raise what you might receive only after everyone eligible has had one lot. In a heavily oversubscribed issue that surplus rarely arrives.
- More applications do. Allotment is per valid application, and each PAN counts once. Separate family members with their own demat accounts and PANs are separate entries.
- Crossing ₹2 lakh moves you out of retail. An application above ₹2 lakh is treated as non-institutional, where the reserved portion is smaller. Applications above ₹2 lakh and up to ₹5 lakh can still be made via UPI, but they compete in the NII category.
- Check the category, not the headline. An issue at 40× overall can have retail at 8× and QIB at 90×. Only your own category's figure affects your odds.
7. Limits of this method
Being straight about what these numbers are and are not:
- The correction factors are empirical, measured on published allotment documents. The large HNI factor follows from the rule and is stable in structure, but the exact value shifts a little with lot rounding — 4.83 on the issue above, 4.93 on another.
- The retail 1.14 depends on how many investors bid more than one lot, which varies by issue. It is the least transferable of the three.
- Subscription figures during bidding are provisional. Final allotment is computed after invalid and duplicate applications are rejected, so live odds move.
- These are estimates of probability, not a prediction of your outcome. The draw of lots is random, and a 20% chance means four in five applicants get nothing.
Sources
- SEBI — Frequently Asked Questions on the ICDR Regulations (May 2025): minimum bid lot, proportionate allotment, category reservations.
- SEBI — ICDR (Third Amendment) Regulations, 2022: one-third / two-thirds sub-categorisation of the NII portion by application size.
- Registrar's basis of allotment for Laser Power & Infra, 16 July 2026: application, allottee and shares-per-bidder figures used in section 4.