Probably my favourite entry in the Market Wizards series. The caliber of the interviewees is superb and the book is full of wisdom from titans of the hedge fund industry, who are not only excellent traders, but thoughtful individuals with battle-tested mental models and the ability to communicate their knowledge.

One of the things that becomes clear across the Market Wizards series (see The New Market Wizards) is that while each individual finds their comparative advantage in a different area, they often share the same core beliefs: a strong empirical leaning, control of their emotions, and an almost paranoid focus on not just decision-making, but the decision-making process. My favourite interviews: Colm O’Shea, Jamie Mai, Mike Platt.

Key ideas

• Reality is all that matters
  • sounds obvious but we rarely behave that way.
  • e.g. Soros has zero attachment to his ideas and just cares about reality
  • Mike Platt’s core strategy was based on the few things we know “for certain” about markets: markets trend, diversification works.
• Correlation is a quantity derived from asset movement.
  • making statements about correlation only makes sense w.r.t your view of the assets (Dalio)
  • look for situations where markets are extrapolating past correlations (Jamie Mai)
• Asymmetries (Jamie Mai):
  • "Markets tend to overdiscount the uncertainty related to identified risks, but underdiscount risks that have not yet been expressly identified". Once there is an identified risk, look into taking the other side. But before it's identified, you can get convex exposure by betting on it
  • market participants extrapolate the past and often assume symmetric uncertainty
  • but idiosyncratic fundamentals can create asymmetries that might not be priced in
• Options mispricings (Jamie Mai)
  • mkt often extrapolates recent low vol – complacency.
  • long-dated equity index options can be used to make bets on interest rates (attractive with low rates)
  • options maths works better over short intervals.
    • e.g often modelled that vol grows as sqrt t – but do we truly believe that 9y std is only 3x greater than 1y std?
    • note: sigma ~ sqrt t requires the stochastic process to be GBM. For Levy alpha-stable distributions, it grows more quickly e.g. sigma ~ t^0.6
• Look over your past trades maniacally: do more of what works and less of what doesn’t (Steve Clark)

Highlights

• Colm O’Shea (Comac Capital, ex Balyasny, Soros) – Discretionary Macro
• Ray Dalio (Bridgewater) – Macro
• Scott Ramsey (Denali Capital) – Discretionary Macro/Futures
• Jaffray Woodriff – pattern recognition CTA
• Edward Thorp (Princeton-Newport) – Genius. See A Man For All Markets
• Jamie Mai (Cornwall Capital, The Big Short) – Special Sits
• Michael Platt (BlueCrest) – Macro/rates
• Steve Clark (Omni Global) – Equity
• Martin Taylor – EM Equities
• Tom Claugus – Equity
• Joe Vidich (Equity)
• Kevin Daly – Long-only Equity
• Jimmy Balodimas (First New York) – Equity
• Joel Greenblatt (Gotham Funds) – Equity