Understanding Diversification and Concentration Strategies

I used to think indexing was lazy. Now I recommend it to every new manager I meet. That shift came into focus at a recent breakfast. 15 GPs, one LP, no agenda. We were halfway through coffee and eggs… Then it started Two old friends, seated at the far end of the table, got into it. “Spray and pray is lazy. You’ve given up on outperformance before writing the first check.” “Concentration is arrogance. You’re betting on your gut over the math.” Forks paused mid-air. Side conversations stopped. Indexing vs. Concentration. A debate as old as venture. The Indexer continued… “Dispersion is high in VC… concentrating is too risky. Our pre-seed index outperforms the average in VC and every other asset class.” The Concentrator: “Dispersion is a feature, not a bug. Our LPs back dozens of funds. They already have an index. They don’t need another.” The Indexer: “And your structure is your weakness. How many great deals didn’t fit your check size or ownership target? You’re building a portfolio of your second and third choices.” The Concentrator: “Tiny checks aren’t the answer. You’re promising pro rata access you can’t deliver. Meaningful ownership creates meaningful co-invest.” The Indexer: “LPs want to co-invest in the best. Indexes have outliers. You can’t double down on winners if you don’t catch them first.” The table was quiet. GPs stared into their mugs. Then, a calm voice cut through the noise Our guest LP finally spoke. “You’re both right… and wrong. Indexing isn’t about humility. Concentration isn’t about arrogance. They’re just different phases. Indexing is where you build your edge. You get reps. You develop differentiated sourcing, selection, and support. You earn the right to lead. Concentration is where you use it. When you know what you’re good at, When you see the pattern And your edge becomes durable You shift from follower to leader. Fred Wilson said it took him ten years to become halfway decent at Venture. Brad Feld made 40+ angel bets before becoming a VC... and later said he should’ve made more. Index early. Learn fast. And when signal is strong... It’s time for conviction. The best LPs don’t need an average. They want upside. The reason they allocate to venture, More than real estate, credit, or even buyout, is simple: Asymmetric return. Much like GPs that invest in startups… The most we can lose is our investment. The upside has no ceiling. We have 26x, 32x, and 48x funds. And persistence in VC is higher than in other asset classes. If we find an extraordinary manager They’re more likely to outperform over many vintages. We’re not hunting for funds. We’re hunting for talent. So, if you’re early in your investing journey? Don’t try to be a sniper before you’ve learned to shoot. Spray and pray works To learn. To build. To earn your edge. Then? Pounce. There is alpha in the average… Find it And success will find you.

Most VCs believe diversification reduces risk. But a decade of Latin American data tells a completely different story. I analyzed Series A performance from multiple VCs across a 10-year period (2014-2024) and found something that challenges everything we think we know about portfolio construction. The most successful investors aren't the ones casting wide nets. They're the ones who've mastered surgical precision. The numbers are striking. Sequoia Capital made exactly 2 investments in Latin America and achieved a 100% unicorn hit rate with Nubank and Rappi. GIC also hit 100% with 2 investments, while General Atlantic achieved 50% with 4 investments. This contrasts sharply with high-volume strategies. MONASHEES made 81 investments for a 4.9% conversion rate, KASZEK invested in 63 companies for 9.5% conversion, and Valor Capital Group deployed across 57 investments for a 5.3% hit rate. These patterns challenge Dan G.'s research showing VCs need 50+ investments to optimize for unicorn outcomes. His analysis works for mature markets but breaks down in emerging markets where opportunity scarcity changes everything. Latin America produced only 39 unicorns over 10 years (3.9 per year). The US produced 974 unicorns in the same period (88.5 per year). When there are only 3-4 real opportunities annually, diversifying across 100+ companies becomes mathematically impossible. But here's the plot twist. Even in mature markets, selection can outperform diversification. Index Ventures maintains a concentrated approach with roughly 4.2 seed and 5.1 Series A investments per year. Their concentrated approach paid off with their Revolut Series A generating 2,500x returns and Figma's seed returning 18x their entire fund size. The Thiel Fellowship backs just 20 fellows per year and achieves a 5.9% unicorn rate—3-10x higher than baseline performance. In public markets, just 7 companies (the "Magnificent 7") drove most S&P 500 returns while 493 companies barely moved. This suggests that concentration of value is universal across asset classes. The question is understanding when and where each approach tends to generate the best risk-adjusted returns for the specific market structure at hand. Ultimately, the market doesn't care about your investment thesis. It only cares about results. See the full analysis with detailed breakdowns in my latest article. If you're interested in data-driven insights on venture capital and fintech in emerging markets, feel free to follow and subscribe for more analysis like this.

Was at an event last week and a common topic that came up was the age old debate between Concentration vs Diversification. Which is better? A common belief of asset allocation is that everyone "should" concentrate to grow wealth, and then diversify to maintain wealth. I don't believe this to be sound advice for the masses, as I believe the first half of this statement to be a misnomer. The majority of investors do not have an edge, due to the natural zero sum game of alpha. And the success of those that have concentrated, in isolation does not account for the survivorship bias of those that concentrated and failed. Historic failures are both, less recorded in the data, and less spread through word of mouth. Failures quietly go unnoticed while success gets free advertisement via the tournament effect. Now alternatively, if one has a true edge, then I'd moderately agree with this statement. If one does have an edge, the correct decision is to add concentration in those areas, particularly where one spends the most of their time and effort. Further breakdown: Concentration is a double-edge sword, that if correct, yields outsized gains Diversification is a quasi-free lunch, that yields steady risk-reward benefits Why do they work? Concentration is applying one's edge, akin to the 80/20 rule. Diversification is benefiting from the law of large numbers. Under what conditions do both fail? Concentration fails if one is wrong. Diversification fails if one seeks short-term outlier outperformance. Why do they fail? Perfect prediction is impossible. Predicting nothing delivers only market returns. Why do them? We concentrate to express what we know. We diversify to express what we don't know for sure. And of course, "It ain't what you don't know that gets you in trouble. It's what you know for sure that just ain't so" -Twain So then what return profile can they deliver if done well? Edges can create periodic asymmetry. Diversification can create persistent asymmetry. What are the outcomes of an approach that balances both? You win if you’re right. You tie if you’re wrong. How allocate to achieve this balance? Align your concentration, to the level of your current edge. Then calibrate to diversify around concentrations. Summary: Neither 100% concentration nor 100% diversification is ideal. In the long run, a time-varying, edge-dependent middle path is best. Get the right balance, somewhere in between, not on either side.

DFA says diversification wins in most scenarios and calls single-stock bets a "lottery ticket." "Addressing Concentrated Stock Positions in Client Portfolios" (June 2024) compares: 1) a low volatility single stock 2) a high volatility single stock 3) diversifying w/ an exchange fund (cost = 60 bps) 4) selling and investing the proceeds in an ETF (cost = 4 bps) In their experiment, diversification wins in bad markets. Concentration wins in good markets. The upshot is that diversification wins *most* of the time. Hard to look away from that high vol single stock lottery ticket, though. Take that bet if you want, but do so with eyes wide open. --- Addressing Concentrated Stock Positions in Client Portfolios Rizova, Savina, and Namiko Saito. June 2024. Refer to the paper for the experiment's methodology.

Portfolio construction for early-stage investing: the perennial debate between concentration vs diversification For VC investing at earlier stages, specifically pre-seed through Series A, investors tend to pick from one of the two following approaches: 👉 Building out a larger, diversified portfolio. The underlying logic here is built on the power law nature of venture. The earlier stages is particularly risky - per PitchBook's 2021 Q3 report, VC Returns by Series: Part III, out of nearly 37,000 companies that have raised their first round of venture funding, less than 1,000 of those businesses were acquired or went public after having raised 4 rounds. As a result, having a small check across a large number of companies developing a product that or technology that will underpin commercial success a decade later helps boost the likelihood of investing into fund returnners (companies that will generate returns that are more than the size of the fund that has invested in it), at least statistically. 👉 Running a small, concentrated portfolio At pre-seed and seed, there's also a relatively smaller number of VC investors that are firm believers in the value of making fewer bets with high conviction. Despite the inherent risk from investing in nascent startups, these investors tend to be derisk through: 1️⃣ being highly specialized This usually entails the investor having a technical background (deep tech and biotech as examples) and in-depth understanding of a (often relatively niche) industry segment. 2️⃣ leveraging unique access to a founder network This is in line with sector specialization. An investor in this case might have exclusive access to a founder community through personal connections, partnership with higher education institutions, etc. An additional motivation comes from their stance on the investor-founder relationship. Some founders prefer working with investors with a concentrated portfolio because they want to spend more time with their investors, developing deeper relationship and working together to solve problems during the entrepreneurial journey. When you're part of a large, diversified portfolio and don't perform as well as some of the front-runners (at least at a certain point in time), some founders could feel left out when the investor doubles down on their best bet and spends less time helping them pivot to a more effective path. What's your take on concentration vs diversification?