Markets Don’t Care About Data Anymore? feat. Ben Hunt & Cem Karsan | U Got Options | Ep.9

T
Top Traders Unplugged Feb 04, 2026

Audio Brief

Show transcript
This episode explores the fundamental transition of financial markets from a regime based on economic data to one driven almost entirely by narrative adoption and behavioral game theory. There are four key takeaways from this conversation regarding the application of AI in investing and the mechanics of modern market movements. First, investors must shift their focus from playing the cards to playing the player. Traditional finance focuses on the cards, which are fundamental data points like earnings, GDP, and cash flow. However, modern investing requires a focus on the players, or the behavioral psychology of market participants. Success in the current environment depends less on analyzing the objective reality of an asset and more on predicting how other participants will perceive and react to information. This shift has been accelerated by central bank policies that have effectively trained investors to prioritize policy communication over hard economic data. Second, the true value of Artificial Intelligence in finance lies in measuring narrative density rather than predicting prices. Large Language Models should not be treated as oracles that can answer open-ended questions about market direction. Instead, they function best as tools for inference within unstructured data. The correct approach is to provide a specific narrative framework, such as the stickiness of inflation, and use AI to rigorously measure the loudness and prevalence of that story across news and transcripts. This allows investors to quantify exactly how much a specific idea is dominating the market psyche. Third, market timing relies on distinguishing between Private Knowledge and Common Knowledge. A fact can be objectively true, known as Private Knowledge, yet have zero impact on asset prices. Prices only re-rate when that information transforms into Common Knowledge, which is defined as information that everyone knows that everyone else knows. The most profitable strategy involves identifying dormant stories—facts that are true but currently ignored—and waiting for the inflection point where narrative volume begins to rise. Investors should not trade against a loud narrative even if the data contradicts it; they must wait for the narrative density to peak and roll over. Fourth, corporate valuation has become a function of storytelling. In this narrative-based regime, a CEO's primary value add has shifted from maximizing operating leverage to engineering a story that the market rewards with a higher multiple. When analyzing management teams, investors should assess the ability of leadership to sell a truthy story—one that feels right to the herd—because the multiple attached to a stock is now largely a derivative of narrative success rather than just operational performance. Ultimately, this discussion suggests that alpha generation now requires quantifying the crowd's attention and waiting for the specific moment when private insights transition into consensus reality.

Episode Overview

  • This episode explores the transition of financial markets from a fundamentals-based regime ("playing the cards") to a narrative-based regime ("playing the players").
  • Guest Ben Hunt explains how to use AI and Large Language Models (LLMs) not as crystal balls, but as tools to measure the "density" and "loudness" of market narratives.
  • The discussion covers why factual truth matters less than "truthiness" in modern investing and how Central Bank policy has trained investors to ignore economic data.
  • Listeners will learn practical frameworks for identifying "dormant" stories and timing market entries based on when private knowledge transforms into "Common Knowledge."

Key Concepts

  • Cards vs. Players (Game Theory): Traditional investing focuses on "the cards" (fundamental data like earnings and GDP). Modern investing requires focusing on "the players" (behavioral psychology). Success now depends on predicting how other participants will perceive and react to information, rather than the information's objective reality.

  • Inference and Unstructured Data: The true value of AI in finance is not in basic sentiment analysis, but in "inference"—finding signal within unstructured data (language, news, transcripts). This involves identifying specific narrative structures and mapping how ideas evolve from obscurity to dominance.

  • Context Engineering with LLMs: Large Language Models should act as operating systems for context, not oracles. You cannot ask an AI "What will the market do?" because it will hallucinate. Instead, you must provide the specific narrative framework (the context) and use the AI to measure the prevalence and volume of that specific story over time.

  • Truthiness vs. Truth: Markets are driven by "truthiness"—stories that feel right to the herd—rather than capital-T Truth. A narrative does not need to be factually accurate to move asset prices; it only needs to resonate with the current psychological state of the market.

  • The Common Knowledge Game: A critical distinction exists between "Private Knowledge" (what you know is true) and "Common Knowledge" (what everyone knows that everyone else knows). Market prices only shift when information becomes Common Knowledge. A "dormant" story creates opportunity, but timing the trade requires waiting for the "inflection point" where the story becomes loud.

  • Valuation as Storytelling: In the current regime, corporate valuation multiples are not a measure of operational efficiency, but a metric of narrative success. A CEO's primary job has shifted from maximizing operating leverage to telling a story that the market rewards with a higher multiple.

Quotes

  • At 0:04:35 - "It's not just playing the cards, it's playing the player... It's all about playing the player." - Hunt establishes the foundational thesis that psychology and game theory now outweigh fundamental analysis.

  • At 0:06:46 - "The real fertile ground, the undiscovered country, is in applying that same level of rigor to unstructured data." - Highlighting the shift from analyzing spreadsheets (structured data) to analyzing human language (unstructured data).

  • At 0:08:44 - "I have no idea what Truth is with a capital T. But I can tell you when a story is truthy, when it sounds right. And that's what clicks for humans." - Explaining why factual accuracy is often irrelevant to immediate market movements.

  • At 0:15:11 - "You can't ask AI to do it. Don't ask AI. You ask one of the guys around here who's been trading something for ten years." - Emphasizing that human expertise is required to define the narratives before AI can be used to measure them.

  • At 0:18:56 - "The stories we had about fundamentals didn't matter anymore... It was the story that central banks were telling." - Identifying the post-2009 regime change where Central Bank narratives overrode economic reality.

  • At 0:26:24 - "We're crazy positive... The density of these stories is off the chart. You haven't seen this density of stories saying 'consumer spending is going to surprise to the upside' since 2021." - Using the disconnect between weak consumer data and bullish media stories to illustrate how markets price "expectations" over "reality."

  • At 0:31:12 - "The easiest use of what we do is look for a story that's dormant... and be watching for it. And when the dormant story starts to get undormant, then you get involved." - Outlining a concrete trading strategy: identify a true but ignored story, then wait for the narrative volume to rise.

  • At 0:39:44 - "It's not what you think. It's not even what you think that other people think. It's what we all think that we all think. That's what Common Knowledge is." - Defining the specific type of consensus required to move markets.

  • At 0:50:04 - "Central bankers said, 'We're going to start using our words for effect. Not for what we really think... but we're using them intentionally to try to change investor behavior.'" - Describing "Open Mouth Operations," where policy makers use communication as a tool for market manipulation.

  • At 0:50:35 - "Today's CEO... you're not being evaluated on 'can I get an additional turn of operating leverage?'... It's 'Can I tell a story that gets a multiple?' Because multiple is story." - A key insight into corporate incentives: valuation is now a function of narrative capability.

Takeaways

  • Practice Context Engineering: Do not use AI to ask open-ended prediction questions. Instead, explicitly define a narrative framework (e.g., "inflation is sticky") and use LLMs to strictly measure the volume and sentiment of that specific topic in the news.

  • Trade the "Rollover," Not the Data: Even if your fundamental data is correct (e.g., "the consumer is weak"), do not short the market while the opposing narrative (e.g., "consumer rebound") is loud and rising. Wait for the narrative density to peak and begin to roll over before positioning.

  • Hunt for Dormant Stories: The highest "alpha" comes from identifying facts that are true but currently ignored by the market (dormant). Monitor these stories for the moment they become "patient zero" and start trending; this is your entry signal.

  • Evaluate Management on Storytelling: When analyzing companies, assess the CEO's ability to drive a narrative. In the current market structure, a management team's ability to sell a "truthy" story drives the stock multiple more than operational tweaks.

  • Distinguish Private vs. Common Knowledge: continually ask yourself: "Is this something I know, or is this something everyone knows that everyone knows?" You cannot trade on Private Knowledge until you see signs it is converting into Common Knowledge.

  • Monitor the "Fiscal Dominance" Narrative: Keep a close watch on the "Sell America" / repatriation of foreign assets story. While currently dormant (a "melting iceberg"), this is identified as a massive potential risk that could rapidly re-price markets if it enters the Common Knowledge phase.