CHAPTERS
The tweet that sparked a VC identity debate
Martín explains the viral tweet that triggered “venture Twitter” and why it landed as more than a simple hot take. The group sets up the core tension: non-consensus can be a source of alpha, but ignoring what the market thinks can be dangerous for both investors and founders.
- •Martín’s claim: non-consensus investing is “dangerous” if you’re blind to how other VCs perceive a company
- •The tweet was not an argument for consensus investing—just a warning against ignoring consensus
- •Early markets may be more efficient than investors like to admit
- •The conversation is framed as an existential question about how venture works today
Defining (and misdefining) “consensus”: hot rounds vs. true market belief
The panel challenges the ambiguous meaning of “consensus,” arguing that people often confuse a tough fundraise with a non-consensus company. They critique anecdotal winner lists and emphasize that many supposed “non-consensus” winners had elite founders, top investors, and premium pricing throughout.
- •“Consensus” is ill-defined; lists of winners can be cherry-picked without data
- •Hard rounds don’t necessarily mean the market lacked conviction
- •Examples like Anduril/Scale are debated as “non-consensus” given founder quality and investor interest
- •Key distinction: look for great companies, not “good deals” versus other investors
Are hot rounds predictive? How to measure the relationship between heat and outcomes
They explore whether fast, competitive up-rounds correlate with success—or just reflect momentum from prior “hot” rounds. The group proposes doing real analysis (basket/correlation) rather than relying on narratives.
- •Hypothesis: best predictor of a hot up-round is that the prior round was hot
- •If true, it suggests inductive market efficiency (earlier investors correctly anticipate future demand)
- •Anecdotes cut both ways; rigorous cohort/basket analysis is needed
- •Question posed: do high-priced rounds still underprice the very best companies?
Fundamentals vs. perception: two ways markets can drive returns
Martín distinguishes between a “productive asset” view (returns come from business fundamentals) and a “perception” view (returns can come from investor sentiment independent of fundamentals). Leo adds that sector fashion cycles show how valuations swing even when fundamentals don’t.
- •Productive asset view: smart investors identify real quality and price accordingly
- •Perception view: market narrative and human opinion can create outcomes even if fundamentals lag
- •Sector cycles (e-commerce, etc.) show shifting appetite without massive fundamental change
- •Implication: returns may depend on both company performance and capital market psychology
Founder perspective: the fundraising trap and the frugality advantage
They discuss how founders experience “non-consensus” as a real operational risk because survival often depends on follow-on capital. Leo counters that being non-consensus can create discipline—hot-money companies may overhire/overspend and collapse when growth slows.
- •Founders feel pressure to “look consensus” to raise even if their product is non-consensus
- •Leo’s upside: hard raises force cash efficiency and frugality
- •Hot rounds can enable shallow diligence and momentum-based markups
- •Martín’s warning: many companies fail from “indigestion, not starvation” (too much money too early)
Efficiency over time—and how the AI wave distorts it
The panel examines whether venture markets are becoming more efficient as the investor base grows. They argue efficiency improves for overlooked companies (more chance someone “gets it”), while competition can make hot-company pricing feel extreme; AI is cited as a live example of both hype and real demand.
- •More investors can make it easier for non-consensus companies to find at least one backer
- •Hot companies may get bid up far beyond intrinsic expectations (founder-friendly, investor-painful)
- •AI boom: speculative capital in unclear models while strong non-AI infra struggles to raise
- •Even amid hype, real growth signals exist (OpenAI/Anthropic/Cursor cited)
Personal startup anecdote: exuberance, drought, and eventual big outcome
Martín recounts raising a high-priced seed round pre-2008, then facing a financing freeze during the recession, then returning to hot rounds as the business showed life—ending in a major acquisition that later proved strategically justified. The story is used to question whether early exuberance is “wrong” or just early recognition of potential.
- •2007: very hot early fundraising before clear product-market fit
- •2008 recession: couldn’t raise; near-death experience despite early hype
- •Later: renewed investor enthusiasm as traction emerged; eventual large acquisition
- •Retrospective ambiguity: market overexuberance vs. early identification of promising initial conditions
How seed investors underwrite the path to consensus (milestones, not miracles)
Martín presses Leo on how a seed investor predicts a non-consensus company will become fundable later. Leo explains the seed bet is often about hitting specific milestones that unlock larger checks—especially in deep tech where “working” businesses may be far in the future.
- •Seed underwriting often targets a milestone path to the next financing, not immediate traction
- •Deep tech: Series A/B may still be pre-product; the bet is whether milestones can be hit
- •Capital plan matters: raising $10M next is easier to underwrite than needing a $50–100M “top 5%” Series A
- •Being aware of what future investors will demand is part of seed diligence
Hype, TAM, and unit economics: humanoids, autonomy, and distortion by ‘infinite markets’
They use humanoid robotics and autonomous vehicles to illustrate how huge TAM narratives can justify almost any valuation, even when unit economics are unclear. Martín argues he can’t underwrite businesses without a believable standalone scaling story; Leo notes mega-TAM logic can warp early-stage decision-making.
- •Massive TAM (e.g., human labor) can make any seed price feel “defensible,” distorting discipline
- •Martín emphasizes unit economics and standalone viability over M&A-optionality investing
- •Autonomous vehicles example: enormous spend with unit economics still not clearly venture-grade
- •Pattern: investors extrapolate from a few “model” AI companies to unrelated spaces without proof
Outcomes are bigger, so should funds and prices be bigger too? SoftBank/Tiger as experiments
The conversation shifts to fund mechanics: if outcomes have expanded dramatically, perhaps prices are still “too low,” but scaling that strategy requires much larger pools of capital. They discuss SoftBank/Tiger’s mixed results and whether failure was due to price, macro cycles, or execution/positioning.
- •Claim: outsized outcomes could imply even expensive rounds are still underpriced for true winners
- •Constraint: fund size and LP capital limit how many large bets can be placed
- •SoftBank/Tiger attempted the ‘more capital at higher prices’ model with mixed outcomes
- •Open question: would the strategy work better with Silicon Valley insiders and different timing?
Competitive identity, cost of capital, and why ‘pure consensus’ would feel like PE
Erik reframes the debate: much of VC identity is tied to being non-consensus, but in a fully efficient world competition shifts to who can accept lower returns (cost of capital). Martín argues public markets reward predictability over innovation, and venture’s pro-growth bias is socially valuable despite misalignments.
- •If everything becomes consensus, investing becomes a cost-of-capital game (who needs the lowest multiple)
- •Erik: some firms fear efficiency because it’s harder to compete and win deals
- •Martín: public markets prioritize predictability, potentially stifling innovation; venture funds upside growth
- •Ecosystem vs. individual incentives: more capital and competition can be better for progress even if harder for any one VC
What the data should answer—and seed vs. multi-stage: who really wins early rounds?
They close by outlining the analyses they want to run: whether winners were priced above/below stage medians and where total returns actually come from. The final thread debates whether multi-stage firms have “won seed,” concluding it depends on founder profile and round competitiveness, with multi-stage advantaged in obvious-repeat-founder cases.
- •Proposed studies: pricing of winners vs. peers; and whether most returns come from high-priced companies
- •Shared lesson: price arbitrage is often less important than simply being in the winner
- •Stage and check size shape strategy: it’s harder to be ‘non-consensus’ with very large checks
- •Multi-stage advantage at seed is strongest for proven founders/obvious deals; less obvious bets remain seed-fund dominated
