Venture Capital Fantasy Portfolio: Build a Track Record With $0

A venture capital fantasy portfolio is a simulated fund: you pick real early-stage companies you would invest in, record a hypothetical check size and entry valuation, write a short memo for each, and track what happens over the following months and years. It costs nothing, and done seriously it is one of the few ways someone outside the industry can show dated, written evidence of investment judgment.

Done lazily, it's a list of logos. Done with rules, it's a track record.

In our view, the exercise is most credible when it follows real rules: a fixed fund size, dated decisions you can't revise, written reasoning, and honest scoring. This guide sets out one version of those rules, a 20-company template, how to score it, how to write it up for interviews, and the mistakes that sink it. It pairs with how to get a job in venture capital and the venture capital case study interview guide.

Why a venture capital fantasy portfolio can work

Venture hiring is largely a judgment problem. Firms can't easily see judgment on a resume, so they look for proxies: deals you've been near, memos you've written, angel checks you've made. A fantasy portfolio is a proxy you can build yourself.

The research on investor decision-making suggests what to demonstrate. In the survey of 885 institutional VCs by Paul Gompers, Will Gornall, Steven Kaplan and Ilya Strebulaev, published as NBER Working Paper 22587, 95 percent of firms rated the management team an important factor in selecting investments and 47 percent rated it the single most important one, and firms considered roughly 100 opportunities for every deal they closed. A portfolio that screens widely, chooses a few, and reasons specifically about founders mirrors that process.

There's a second benefit. Venture returns are extremely concentrated: Chris Dixon's analysis at Andreessen Horowitz of Horsley Bridge data found that about 6 percent of investments, representing 4.5 percent of dollars invested, generated roughly 60 percent of total returns. Run a paper portfolio for a year or two and you'll feel that in your gut. Most of your picks will stall. The exercise is about finding the one or two that don't.

Fantasy portfolio versus investment thesis

These are two documents, and interviewers often ask about both. An investment thesis is the argument: what you would back, why it should outperform, and what edge you have in finding it. A fantasy portfolio is the evidence: the dated decisions you made under that thesis, and what happened next.

Write the thesis first, then let the portfolio test it. If after 20 picks your companies don't match your stated thesis, one of the two probably needs revising. Saying which in an interview can itself be a strong answer.

Ground rules that make it credible

Fix the fund. Choose a size and strategy and write it down: for example, $10M, 20 initial checks of $250K to $350K with the rest held back for fees and follow-ons, pre-seed and seed, one or two sectors. Constraints make choices meaningful. Portfolio size is a choice too, and we think it's worth making on purpose rather than by accident (our take on why there's no magic number).
Only real, investable companies, raising or recently closed at your stage.
Date every decision. Record the date you "invested," the round, and the valuation or cap as reported. Once logged, treat it as fixed.
Write a memo for every pick. One page: what the company does, why the team, why now, the key risk, the price, and the path to a fund-returning outcome.
Log your passes too. For every investment, note two or three companies you passed on, with a sentence of reasoning. Passes are a large part of the judgment.
Track quarterly: follow-on rounds, valuation changes, product news, team changes, shutdowns.
Score honestly. A conservative approach is to mark to market on announced rounds only and count shutdowns and silence as losses.

Where to find companies, and the tools to use

You probably don't need paid data. Three free sources, all live in September 2026, cover most of an early-stage funnel: the YC Startup Directory, Y Combinator's public list of funded companies across verticals from biotech to developer tools, which is the densest source of stage-appropriate names; Product Hunt, which posts new launches daily with community rankings and catches companies before a round is announced; and firm portfolio pages and round announcements, which teach you each fund's real stage and sector.

For the portfolio itself, a spreadsheet plus one document per memo is usually enough. We'd skip tools that lock your record inside a product you may lose access to. The point is a dated file you still have in two years.

A 20-company template

One simple setup is a spreadsheet with these columns and one memo document per company:

Company, sector, stage, date of decision.
Round size and post-money valuation or SAFE cap (as reported; note the source).
Your hypothetical check and implied ownership.
One-line thesis.
Key risk.
Last update date and current status: up round, flat, down, shut down, no news.
Current mark (last valuation multiplied by your ownership, adjusted for estimated dilution).

Use realistic 2026 numbers. Carta's State of Pre-Seed report for Q2 2026, published in August 2026, put the average pre-seed SAFE or note at $276,000, up 27 percent year over year, across more than 11,500 instruments. At Series A, Carta's State of Private Markets report for Q1 2026 put the median Series A valuation for a non-AI startup at $55 million against $300 million for AI foundational model companies. That's a useful check if everything in your portfolio is priced like the latter.

Don't skip ownership. It's what makes the returns math work. As an illustration, using a check size from the example fund above, a $300K check into a $6M post-money round is 5 percent; the same check into a $20M post-money round is 1.5 percent. After three later rounds at roughly 18 percent dilution each, that 5 percent falls to about 2.8 percent.

How to score a venture capital fantasy portfolio

The industry's own metrics work at paper scale. Three are worth computing:

TVPI (total value to paid-in): current marks plus any hypothetical distributions, divided by capital deployed, the same measure AngelList's explainer describes for real funds. A portfolio with $5M "deployed" and marks totaling $9M has a TVPI of 1.8x.
Loss ratio: share of companies (and of dollars) marked at or below cost, including shutdowns.
Concentration: what share of total value comes from your top one or two companies. In a power-law business, a high number here isn't a warning sign. It's the pattern.

Lean conservative: mark up only on priced rounds announced by the company or reported by a credible outlet, estimate roughly 15 to 25 percent dilution per subsequent round, and treat a company with no news for 18 months as impaired. For a sense of what "good" looks like, Cambridge Associates reported that its US venture capital index returned 21.1 percent in calendar year 2025, its best year since 2021, though a two-year paper portfolio can't be compared to it directly.

The point of scoring isn't to claim a return. It's to make you confront your misses, so that in an interview you can say "my loss ratio was 45 percent and here is what I learned about why."

"But paper picks prove nothing"

This is the strongest objection, and it's fair. No money is at risk. Nobody had to win the allocation. There was no founder to convince, no partner to argue with, and you can quietly drop the file if it goes badly. Some investors will discount it for exactly those reasons.

But a dated memo still shows how you reason, and a logged pass that turned out right is hard to fake after the fact. Our read: a fantasy portfolio won't prove you can invest. It can prove you think like someone who could, and that you kept score honestly when nobody made you. That's more than most applicants bring.

Turning the portfolio into interview material

A fantasy portfolio is most useful if you can talk about it. Prepare:

A one-page summary: fund strategy, number of picks, dates, headline TVPI and loss ratio, top two companies, biggest miss.
Three memos you're proud of, including at least one pass that turned out right.
The post-mortem: two picks that went wrong and what the memo missed. Investors tend to trust people who can describe their own errors precisely.
A view that changed because of the portfolio.

No firm publishes a rubric for mock portfolios and few ask for one by name, so treat this as what the exercise is designed to prove rather than a stated hiring requirement: that your picks were investable at the firm's stage, that the memos reason about the team specifically, that a pass log exists, and that you tracked outcomes rather than building a list. A portfolio started 18 months ago with quarterly updates likely carries more weight than one built last week.

Publishing often helps. A short write-up per quarter, on a personal site or newsletter, creates a dated public record and attracts founders and investors. Several of the investors listed in best venture capital blogs began as people writing publicly about companies they liked.

Common mistakes

Picking famous companies. A portfolio of companies that already raised at $100M valuations shows taste, not sourcing. Stay at the stage you want to work in.
No entry prices. Without a valuation, there's little return math and less discipline.
Rewriting history. Changing dates or memos after the fact undermines the exercise. Keep the original files.
Only tracking winners. Logging each pick and pass keeps the record honest.
Treating it as done. The value compounds with time; a portfolio without updates is a list.
Not talking to founders. Even a paper investor can ask for a 20-minute call.

A fantasy portfolio is practice, and the limit of practice is that nobody depends on the answer. In 1752vc's Venture Fellow program the same work is live: across 8 weeks of live virtual sessions, Fellows source companies for partner funds and run due diligence on real ones, and they earn payouts on deals they source and carry on select deals, so a thin memo has a cost attached. About half of 1752vc's deal flow is sourced by Fellows, which is the difference between a pick on paper and one a fund acts on. The portfolio is also strong material for applications to a venture capital fellowship or a scout program.

The bottom line

A fantasy portfolio is only as good as its rules. Fix the fund, date and price every pick, write the memo, log the passes, and score yourself harder than anyone else would. Then keep it running long enough for the misses to show up.

Anyone can name the winners today.

Few can show they named them two years ago.

Key takeaways

A venture capital fantasy portfolio works best, in our view, as a paper fund with fixed rules: set fund size and strategy, pick real companies, date and price every decision, write a memo, log passes, and track outcomes.
It is the evidence for your investment thesis, not a substitute for it. One approach is to write the thesis first, then let 20 dated picks test whether you actually follow it.
Consider scoring with TVPI, loss ratio, and concentration, using conservative marks based only on announced rounds and estimated dilution.
A one-page summary, three memos, and an honest post-mortem help the portfolio work as interview material.
It helps to keep it at the stage you want to work in, avoid rewriting history, update quarterly, and talk to founders where you can.

Frequently asked questions

What is a venture capital fantasy portfolio?

It is a simulated venture fund in which you select real early-stage companies you would back, record a hypothetical check size and entry valuation, write a short memo for each, and track the outcomes over time. People trying to break into VC use it to build dated evidence of investment judgment without capital.

How do I build a mock VC portfolio?

One approach: fix a fund size and strategy, choose 15 to 25 real companies raising at your target stage, record the date, round, and valuation for each, write a one-page memo, log the companies you passed on, and update the portfolio every quarter with follow-on rounds, shutdowns, and marks.

Do VC firms care about a fantasy portfolio?

Few firms ask for one by name. What they respond to is what it reveals: whether you can source at the right stage, reason about founders, price a deal, and confront your misses. A dated, tracked portfolio with memos and a pass log gives you far more to say in an interview than a list of well-known company names.

How do you score a fantasy VC portfolio?

A common approach uses total value to paid-in (TVPI), loss ratio, and concentration. Mark companies up only on announced priced rounds, assume roughly 15 to 25 percent dilution per subsequent round, and count shutdowns and long silences as losses. Treat the result as a rough guide, not a claimed return.

How many companies should be in a fantasy venture portfolio?

In our view, fifteen to twenty-five is typical for a seed-style paper fund. That is enough for the power-law pattern of venture returns to show up, and few enough that you can write a real memo for each and track them quarterly.

Sources

Disclaimer: This guide is for general education only and is not legal, tax or investment advice. Laws, market data and program terms change, so it may not reflect the latest developments or fit your situation. Treat it as a starting point, not a source of truth, and talk to a qualified lawyer, accountant or financial adviser before you make decisions.