Podcast Summary: "When Signals Matter More Than Stories"

Podcast: Top Traders Unplugged – Systematic Investor Series
Episode: SI383
Host: Niels Kaastrup-Larsen
Guest: Nick Baltas
Date: January 17, 2026

Episode Overview

This episode dives deep into the mechanics and philosophy of systematic trend following, exploring the complex interplay between market narratives, signals, and investment strategy design. Host Niels Kaastrup-Larsen and guest Nick Baltas examine current market conditions, reflect on lessons from recent years, and discuss several recent research papers that push the boundaries of trend following methodologies. The episode serves both as a practical update and a thoughtful academic exploration, offering insights for portfolio builders, asset allocators, and those interested in robust systematic investing.

Key Discussions & Insights

1. Narratives vs. Signals in Investing

• Narratives' Influence: Nick is fascinated by how investor narratives and themes shape asset allocation decisions, often before price moves are reflected (03:10).
  • Quote: “I’m spending a bit more time these days looking into the way investor narrative and thematics are impacting asset allocation...” – Nick (03:10)
• Can signals preempt stories?: The hosts discuss whether earlier access to narrative data could provide an "edge" by signaling momentum before it appears in prices (03:45).
• Cautionary Note: Niels doubles down on the value of letting price-based signals (rather than predictions or stories) drive trend following, cautioning against a model that relies on anticipating market narratives (06:00).
  • Quote: “I’m reminded ... how important it is not to have any kind of prediction in your model.” – Niels (05:50)

Timestamps:

• Narrative & signals: 03:10 – 09:30

2. Market Update & Performance Review

• Trends in 2025–2026: Both 2025 and the start of 2026 showed strong conditions for trend, with performance notably concentrated in metals and equities, while fixed-income lagged (11:00–15:00).
  • Quote: “There were a few markets that exhibited strong trends ... probably equities and then precious metals, that’s it.” – Nick (12:22)
• Dispersion Among Trend Programs: High dispersion in program outcomes was driven by the universe and speed of trend signals used—smaller, more concentrated universes and slower speeds often outperformed (12:00–16:00).

Timestamps:

• 2025–2026 markets & trend update: 11:00–15:00
• Index returns and breakdown: 15:10–18:00

3. Universe vs. Speed – What Drives Dispersion?

• Key Empirical Findings:
  • Speed: Over the past three years, slower trend-following speeds consistently outperformed—unprecedented in the last 25 years (19:28–22:00).
  • Universe: Very concentrated (small) universes delivered top performance recently, again rare in long-term historical analysis (22:00–24:12).
  • Historically, small universes were usually lowest performing for a decade post-2001 tech bubble, cautioning against drawing long-term conclusions from recent outperformance (24:12–25:10).
  • Quote: “Pretty much since 2016, small universes have been, you know, the better performer ... but ... through 2009, being small ... was the worst universe every single year for something like seven years in a row.” – Niels (24:12)
  • Caveat: Performance patterns can flip unpredictably; discipline and diversification are more robust than chasing recent winners (25:10–29:00).
  • Quote: “Reacting to out or underperformance is the wrong recipe ... I think discipline is important. I think trust to the process is important.” – Nick (28:09)

Timestamps:

• In-depth on speed & universe: 18:43–24:30
• Philosophy on strategy design: 25:10–31:10

4. Sector Contributions & Diversification

• Sector Profitability: All sectors have produced profits over two decades, but leadership shifts. Diversified portfolios remain intuitively superior over time, even if concentration sometimes outperforms (29:12–31:10).
  • Quote: “The truly diversified portfolios ... over time is a better ... more intuitive way of doing trend following ... not trying to be too concentrated.” – Niels (30:52)

5. Review of Recent Research Papers

A. Makita Paper: “Trend Following & Defensive Allocations”

• Practical Overview: Makita’s latest publication offers a user-friendly overview, focusing on design decisions: universe, speed, alternative signals, stop loss mechanisms, and risk allocations (32:10–36:00).
• Non-Trend Signals: Incorporating mean reversion and macro indicators (non-trend signals) helps reduce allocation loss aversion, smoothing returns and making it easier for investors to stick with trend strategies (36:00–38:07).
  • Quote: “Should we be able to allow allocations in trend following be maintained by those non trend components? I think we eventually make the investment process more complete ...” – Nick (37:49)
• Institutional Access: Recognition that institutions can now access trend following through various wrappers—CTAs, mutual funds, ETFs, QIS, etc.—signals increased mainstream adoption.

Timestamps:

• Makita paper discussion: 31:50–38:07

B. Moskovich et al.: “Non-Linear Time Series Momentum”

• Main Idea: Investigates whether the relationship between recent returns (trend signals) and future returns is best captured with non-linear (sigmoid) rather than simple linear or binary functions (40:22–50:33).
• Method: Uses neural networks to let the data “speak”, revealing that position sizing should flatten out in extreme signals (tails), sometimes even suggesting mean reversion.
  • Quote: “The novel thing ... is that if there are those nonlinearities ... maybe we let the data speak and we use a neural network ... to uncover those relationships.” – Nick (40:52)
• Implications: Neural nets confirm theoretical intuition but add complexity; whether the gain justifies the cost over classic parametric approaches remains to be seen.

Timestamps:

• Nonlinear trend signals: 40:22–50:33

C. Kerr (ATP): “On the Autonomy of Trend”

• Mathematical Deep Dive: Explores what truly drives trend following performance and its crisis alpha properties (51:30–55:01).
• Key Finding: Even if returns are IID (no serial correlation), conditional on observing a significant drawdown, trend following can still generate positive expected returns—a mathematical foundation for trend’s crisis-resilience.
  • Quote: “They show that beyond specific thresholds ... a trend follower can still deliver positive return. And specifically in the drawdown, specifically as a crisis alpha ...” – Nick (55:02)
• Connection to Practice: Supports the original definition of “crisis alpha”, emphasizing trend’s value exactly when it’s needed most.

Timestamps:

• Kerr paper discussion: 51:30–56:37

Notable Quotes & Memorable Moments

• On the Value of Signal Over Story:

  • Niels: “A lot of people don’t like the fact that we have no story and we have no view ... but just the fact that we’re super disciplined at following what has happened rather than trying to get ahead of ourselves.” (05:50)

• On Recent Market Trends:

  • Nick: “There were a few markets that exhibited strong trends ... probably equities and then precious metals, that’s it.” (12:22)
  • Niels: “Despite, you know, whether people ended up plus 5, plus 10 or minus 5 doesn’t really matter. But ... the return distribution actually looked pretty familiar to trend followers.” (15:40)

• On Speed & Universe Lessons:

  • Nick: “For the last three years ... if you were slow, you would have outperformed ... there is no other time ... that three years in a row being slow would have allowed you to outperform.” (19:49)
  • Niels: “I think we need to be very cautious of making too many conclusions based on people saying ... we can outperform the CTA index ... based on your completely independent type of data here.” (24:59)

• On Theoretical Insight from Kerr’s Paper:

  • Nick: “... even an IID process can allow you still to be profitable with the trend following strategy is an important finding.” (55:11)

Segment Timeline Reference

• 00:00–03:10: Intros, personal anecdotes, travel woes
• 03:10–09:30: Narratives, signals, and digestion of market information
• 11:00–15:00: Performance review/market update for trend following
• 18:43–29:12: Speed vs. universe; recent empirical findings and implications
• 29:12–31:10: Sector analysis/diversification philosophy
• 31:50–38:07: Makita paper discussion – design & defensive overlays
• 40:22–50:33: Moskovich paper – nonlinearities in trend following
• 51:30–56:37: Kerr paper – mathematical rationale for crisis alpha
• 56:37–End: Acknowledgements, wrap-up

Tone & Style

Insightful yet accessible, with technical complexity where needed. The hosts are candid about what is known, unknown, and where true edge or robustness comes from—a consistent emphasis on data-informed discipline, skepticism of market stories, and long-term, resilient strategy design.

For Further Reading

• [Makita Consultants: Trend Following Papers (2024/25)]
• [Tobias Moskowitz et al.: “Non-Linear Time Series Momentum” (Dec 2025)]
• [Christian Kerr: “On the Autonomy of Trend” (ATP, Dec 2025)]
• Top Traders Unplugged Ultimate Investment Book Guide

For those continuing their learning journey, this episode is a comprehensive reference for state-of-the-art trend following, blending practical market wisdom with the latest research.