Quantitative: 1. Bid-ask spreads 2. Market breadth 3. Cross-Sectional Return Dispersion 4. Average Pairwise Correlation 5. Liquidity on smaller projects 6. Volatility on other coins than BTC 7. Trading expectancy on simple models 8. Momentum persistence ... Qualitative: 1. Talking to other trading teams 2. Talking to allocators 3. Talking to capital introducers 4. Potentially - measuring the activity on X (we don't do that) The thing is that usually, the biggest returns can be made from the change of the regimes. Being out completely is probably not the smartest way because one can miss the initial momentum wave.

Crypto is cyclical: in the ratio between retail and institutional capital. During hype periods, retail participation rises. More retail = more uninformed capital. More uninformed capital = more market inefficiencies. More inefficiencies = higher PnL potential. Then the cycle reverses. Retail disappears = edges shrink and expectancy falls. During these periods, weaker trading teams and funds also disappear. They are no longer extracting returns from retail flow. That is the real crypto cycle. You can observe it in the data. You can also see it behind the scenes in the behavior, performance, and survival rate of crypto trading teams and funds.

The current crypto market is bad. Sentiment is extremely bearish. So we checked the full historical record and searched for the closest analogue. 16 metrics across 5 equally weighted blocks: 1. Volatility breadth: mean, median, 10% trimmed mean, and share of Top40 above 75%, 100%, and 150% annualized volatility. 2. Dispersion: P90−P10 volatility spread and the five most volatile coins’ contribution to total volatility. 3. Correlation: average pairwise Top40 correlation and average correlation to BTC, both over 30 days. 4. Downside: annualized negative-return semivolatility, average percentage of coins declining together, and frequencies of ≤−10% and ≤−20% daily returns. 5. Persistence: 30-day volatility-of-volatility and its coefficient of variation. The strongest historical analogue for March–July 2026 is: 2023-05-20 to 2023-10-15 A low-volatility regime with very limited participation and almost no market interest. What followed was a major expansion in volatility and momentum.

It is actually $1.2 billion versus his current net worth of $300 million. Of course: 1. Massive hindsight bias. 2. With this philosophy, he would not have shorted the housing market in the first place. Still, it is a good example of the cost of betting against the structurally long-biased equity market.