Top Traders Unplugged: CTAs—The Unexpected Heroes of Market Crises

SI343: Systematic Investor Series with Yoav Git

Date: April 12, 2025
Host: Niels Kaastrup-Larsen
Guest: Yoav Git

Overview of the Episode

In this episode, Niels and Yoav provide a deep dive into recent turbulent market conditions, focusing on the surprising behavior of bonds and traditional safe havens, and how systematic trend following (CTAs) have performed during this period. They examine the reliability and adaptability of trend following strategies, the structural challenges facing CTAs, the myth of "crisis alpha," and how portfolio diversification—and correlation—affect managers' results in times of stress.

Key Discussion Points & Insights

1. Market Stress: Bonds, Yields, and Safe Havens

Timestamps: 02:58–11:50

• Unusual Bond Market Moves:

  • Niels reflects on the historic spike in US 10-year yields, the biggest three-day jump since 2001, despite expectations of Treasuries as a "safe haven" ([03:00]).
  • Yoav explains the impact of global debt ownership and trade tensions on the bond rout, noting that US financing heavily relies on foreign buyers ([05:16]):

    “If you start a trade war, this can quite happily escalate into capital war.” (Yoav, 06:00)

  • Sharp divergence in yields across countries: yields up in the US and UK, down in Japan.

• Crumbling Correlations:

  • Discussion about how traditional correlations—stocks down, bonds up—failed, especially in 2022 and now ([08:29]).
  • Investors relying on 60/40 portfolios face challenges when inflation becomes a dominant market driver, creating positive correlation among asset classes.

• Evolving Perceptions of Safe Havens:

  • Gold outperformed other commodities as a relative safe haven; cryptos discounted as too volatile ([11:06]).
  • Niels argues that CTAs/trend following, by virtue of adaptability, may now function as modern “safe assets.”

2. Recent CTA/Trend Following Performance in Crisis

Timestamps: 11:50–21:29

• Portfolio P&L Breakdown ([11:50–15:14]):

  • Equities and currencies hurt CTA portfolios; fixed income mixed, with massive divergences between countries.
  • Energies were a relative bright spot—longer-term managers may have benefited from holding shorts as the selloff hit.

• Volatility, Risk Management & Positioning:

  • Sudden reversals forced aggressive risk reduction among CTAs.
  • Many managers, particularly slower traders, held onto long equity exposure amid rapid moves ([19:16]).

• CTA Indexes Update

  • B Top 50 Index and Soc Gen CTA/Trend indexes suffered large losses for April and YTD (e.g., Soc Gen Trend Index down almost 10% for the year as of Tuesday) ([17:10–19:48]).
  • Short-term trader indexes were doing better but still negative.

3. The Crisis Alpha Debate: When Do CTAs Actually Shine?

Timestamps: 21:29–33:01

• Performance During Drawdowns

  • Citing Twitter research (attribution: Tyler Loving), Niels summarizes how CTAs tend to lag in the initial leg of equity crises ([21:29–27:12]):

    “What’s very clear is that when it happens very quickly, CTAs are struggling the most.” (Niels, 25:04)

  • Over 10 S&P 10% drawdowns since 2000, average CTA performance was ~-3%. The “crisis alpha” typically emerges in the -10% to -20% S&P drawdown leg.

• Structural Evolution

  • Yoav elaborates: CTAs today trade slower than in the early 2000s and often can’t turn positions quickly when crises start from all-time highs ([27:12]):

    “It’s not just the speed of the crisis, it’s also the speed of CTAs…The bulk of the money has migrated to trading slower.”—Yoav (27:58)

  • Value of diversification: In a “cross-asset” crisis, even highly diversified futures portfolios can’t fully insulate from sharp, correlated reversals.

• Portfolio Structure Lessons

  • The old “commodities provide the best offset in equity crises” lesson is reaffirmed ([30:04–33:01]).
  • Need for rigorous risk management and real diversification, especially when crisis correlation spikes across traditional “safe” assets.

4. Trade Execution, Market Impact, and Diversification

Timestamps: 34:26–45:00

• Reflections on Diversification Within CTAs ([34:59–45:02]):

  • Yoav highlights two ways to lower market impact: mixing signal timeframes and occasionally trading further out the futures curve ([34:59]).
  • “Correlation at Entry vs. Exit”: Exits are highly correlated due to risk management triggers—everybody de-risks together when volatility spikes:

    “Entry is rather smooth, but exit, we all tend to exit at the same time.” (Yoav, 35:39)

  • “Unexpected volatility” when trends are extended is a measurable sign of CTA crowding and impact.
  • Wide dispersion among trend signals: adapting the type of trend following predictor and time frame to the market context can improve alpha.

• Capacity as a Source of 'Alpha':

  • Smaller managers can deploy meaningful risk to less liquid, diversifying markets (metals, ags), whereas mega-CTAs are forced to concentrate in the most liquid (and correlated) sectors.

    “Having 500 markets or 50 markets is less important than having volatility reduction. That’s really what is key.”—Yoav (45:15)

  • Alpha from trading “off the beaten path” is as much about lower correlation as about higher returns.

5. Are All CTAs the Same? Manager Dispersion & Correlation

Timestamps: 50:06–57:26

• Manager Dispersion Data ([50:06–56:04]):
  • Recent CFM and Goldman Sachs analyses show that pairwise CTA correlations remain ~50–60%—far higher than in most macro investing.
  • Despite industry expansion and more tradeable markets, the majority of managers are highly correlated, all exposed to the same core trends:

    “If you look at CTAs…it’s completely shocking…We end up very highly correlated.” (Yoav, 53:10)

  • Key to differentiating: willingness to allocate risk to off-benchmark or “non-standard” markets, though capacity becomes the limiting factor for large players.

6. Expectations, Trend Speeds & the Misconception of CTA Positioning

Timestamps: 57:27–69:18

• Banks' “Front-Running CTA” Reports ([57:59–64:16]):

  • Investment banks regularly try to “predict” CTA buying/selling flows. Niels and Yoav are skeptical of these simplistic model-based calls, given model diversity and real-world uncertainty.

    “I’m not a big fan of [front-running CTA positions]…they probably have some idea of how these models are built, but they don’t have the full picture.” (Niels, 63:25)

• Why Aren’t CTAs Short Equities Yet?

  • Time vs. Price: Most trend-followers, now slower and coming from all-time highs, wouldn’t have flipped their book short despite sharp drawdowns ([60:41]).
  • The design of signal lookback—one month to three months—means the fastest, largest reversals often occur before the models are able to react.

    “Trend following is really about two things. It’s about the price, but it’s also about the time.”—Niels (59:07)

• Practical Takeaways for Investors:

  • Set expectations: Don’t expect CTAs to deliver “instant” crisis alpha in rapid market meltdowns.
  • Diversification and robust risk management remain the core strengths of managed futures, over time.

Notable Quotes & Memorable Moments

• On safe havens:

  “I was very thankful that it wasn’t crypto that would have been insane as a safe haven.” (Yoav, 11:10)

• On structural market risks:

  “If you start a trade war, this can quite happily escalate into capital war.” (Yoav, 06:00)

• On the “myth” that CTAs protect instantly in a crisis:

  “It is super important…people should not expect CTAs or trend followers to be the first responders. We’re not the ones who can deliver the first positive offset when a crisis begin. It comes later.” (Niels, 25:18)

• On the reality of dispersion among CTAs:

  “It’s like Alice in Wonderland. You have to run twice as fast just to stay, just to make any sort of progress.” (Yoav, 56:04)

Key Segment Timestamps

• [03:00] – US bond market turmoil, yield surge, safe haven inversion
• [08:29] – Correlation breakdown, 60/40 obsolescence
• [11:06] – Gold, CTAs, and evolving safe havens
• [15:14] – Country-by-country fixed income/FX behavior in crisis
• [21:29] – “Crisis alpha” and trend follower performance in S&P drawdowns
• [34:59] – Moritz’s Substack post: execution, market impact, timing of trades
• [45:02] – Capacity, portfolio construction, and true diversification
• [50:06] – Dispersion between managers, persistent high correlations
• [57:27] – Explaining “time vs. price” in trend following exits and expectations
• [63:25] – Skepticism of “frontrunning CTA” narratives

Tone & Language

The episode is frank, data-rich, and attentive to nuance—consistent with Top Traders Unplugged’s style. Niels and Yoav blend industry humor with technical detail, offering both strategic insight (“what matters isn’t the number of markets, it’s the correlation structure!”) and practical portfolio lessons. Their tone is skeptical of simple narratives, especially around “crisis alpha,” and they encourage listeners to look beyond headlines and focus on real-world risk management, structural diversification, and strategy design.

Summary Takeaways

• Traditional safe havens (bonds) can fail—and have recently.
• CTAs are not “first responders” in fast equity selloffs—crisis alpha materializes if downturns persist.
• Trend following performance is highly path-dependent: signal speed, market structure, and correlations matter.
• Capacity constraints drive mega-CTAs into correlated, less diversifying trades; smaller managers can find extra alpha in “off-index” markets.
• Portfolio risk control, robust signal design, and adaptive diversification are more important than chasing short-term performance.
• Despite much talk, actual dispersion among major CTAs remains low. The “trend engine” remains resilient, but investors must be realistic about how and when it works.

Next episode: Dr. Kathryn Kaminski returns to discuss new research on Crisis Alpha and the role of CTAs in turbulent markets.