Coming soon

Predicting IPOs that actually return.

We combine statistical models with our brain models to selectively pick IPOs worth investing in. Most IPOs underperform — we're building the system to find the ones that won't.

Most IPOs lose money for retail investors.

Hot IPOs go to institutions. Retail gets the leftovers. The majority of new listings trade below their offer price within twelve months — but a small minority deliver outsized returns. The only thing that matters is being able to tell them apart.

~60%

of IPOs trade below their offer price within 12 months.

Negative

Median first-year return for IPOs since 2000.

3x S&P

Average first-year return of the top-quartile IPOs over the same period.

Figures are illustrative industry estimates aggregated from academic studies on IPO underperformance and are not predictive of future results.

Our approach

Statistical models. Brain models. Selective conviction.

Statistical models

We ingest S-1 filings, financials, comparable IPO history, sector trends, lockup expiries, and aftermarket flow. ML models score each upcoming IPO across multiple return horizons.

Brain models

Models miss context. Our team layers human judgment on top — founder track record, narrative quality, market timing, and the question models can't answer: why this IPO, why now.

Selective conviction

We don't grade every IPO. We publish a prediction only when statistical signal and human conviction agree. Most IPOs get a pass — that's the point.

What goes into a prediction

Six categories of signal feed every prediction we make.

  1. 1

    Financials & growth quality

    Revenue durability, gross margin, burn, path to profitability.

  2. 2

    Comparable IPO performance

    How similar listings have performed in the same window.

  3. 3

    Underwriter track record

    Lead bank pricing accuracy and after-market support history.

  4. 4

    Lockup & insider dynamics

    Who can sell when, and how much overhang to expect.

  5. 5

    Sector momentum

    Whether the broader category is in favor or out.

  6. 6

    Narrative & founder fit

    Story durability, founder track record, market timing.

Each prediction comes with the model's confidence score, the human override (if any), and the reasoning — so you can decide whether you agree.

What you'll get

Selective IPO predictions

Only the IPOs we have real conviction on — with confidence score, rationale, and the specific risks that could break the thesis.

Plain-English breakdowns

Why this IPO, what the comparables say, what could go wrong, and what to watch for in the weeks after listing.

This is research and analysis, not investment advice. IPOs.fyi is not a registered investment adviser or broker-dealer. Always do your own research and consult with a qualified financial advisor before making investment decisions.

Be the first to know when we launch.

We're opening this up to a small group of investors. Drop your email if you're interested and we'll let you know the moment we go live.