Crazy Wisdom
Episode #468: Forecasting the Market’s Weather: Events, AI, and the Future of Trading
- AI & Agents
- Knowledge & Learning
- Level Fields
- artificial intelligence
- financial markets
- event-driven investing
- CEO departures
- earnings reports
- data analysis
- market prediction
This video isn't available here. Try YouTube, or listen to the audio below.
Watch on YouTubeAbout this episode
Timestamps
00:00 – Andrew introduces Level Fields and explains how it identifies event-driven stock movements using AI.
05:00 – Discussion of LLMs vs. custom models, and how Level Fields prioritized financial specificity over general AI.
10:00 – Stewart asks about ontologies and knowledge graphs; Andrew describes early experiences building rule-based systems.
15:00 – They explore the founder’s role in translating problems, UX challenges, and how user expectations shape product design.
20:00 – Insight into feedback collection, including a unique refund policy aimed at improving user understanding.
25:00 – Andrew breaks down the complexities of user segmentation, churn, and adapting the product for different investor types.
30:00 – A look into event types in the market, especially crypto-related announcements and their impact on equities.
35:00 – Philosophical turn on narrative vs. fundamentals in finance; how news and groupthink drive large-scale moves.
40:00 – Reflection on crypto parallels to dot-com era, and the long-term potential of blockchain infrastructure.
45:00 – Deep dive into machine persuasion, LLM training risks, and the influence of opinionated data in financial AI.
50:00 – Final thoughts on momentum algos, market manipulation, and the need for transparent, structured data.
Key Insights
- Event-Based Investing as Market Forecasting: Andrew Einhorn describes Level Fields as a system for interpreting the market’s weather—detecting recurring events like CEO departures or earnings beats to predict price movements. This approach reframes volatility as something intelligible, giving investors a clearer sense of timing and direction.
- Building Custom AI for Finance: Rejecting generic large language models, Einhorn’s team developed proprietary AI trained exclusively on financial documents. By narrowing the scope, they increased precision and reduced noise, enabling the platform to focus only on events that truly impact share price behavior.
- Teaching Through Signals, Not Just Showing: Stewart Alsop notes how Level Fields does more than surface opportunities—it educates. By linking cause and effect in financial movements, the platform helps users build intuition, transforming confusion into understanding through repeated exposure to clear, data-backed patterns.
- User Expectation vs. Product Vision: Initially, Level Fields emphasized an event-centric UX, but users sought more familiar tools like ticker searches and watchlists. This tension revealed that even innovative technologies must accommodate habitual user flows before inviting them into new ways of thinking.
- Friction as a Path to Clarity: To elicit meaningful feedback, Level Fields implemented a refund policy that required users to explain what didn’t work. The result wasn’t just better UX insights—it also surfaced emotional blockages around investing and design, sharpening the team’s understanding of what users truly needed.
- Narrative as a Volatile Market Force: Einhorn points out that groupthink in finance stems from shared academic training, creating reflexive investment patterns tied to economic narratives. These surface-level cycles obscure the deeper, steadier signals that Level Fields seeks to highlight through its data model.
- AI’s Risk of Amplifying Noise: Alsop and Einhorn explore the darker corners of machine persuasion and LLM-generated content. Since models are trained on public data, including biased and speculative sources, they risk reinforcing distortions. In response, Level Fields emphasizes curated, high-integrity inputs grounded in financial fact.
Episode transcript
Welcome to the Crazy Wisdom Podcast. This podcast is for you. If you have an insane drive to find the truth of things, it's not the good answers that we seek, but the good questions. I interview a range of different guests from many different fields, all with the intention to uncover the simple truths that are hidden in plain sight. Most people don't want to go there. I go there, my guests go there, and you benefit. Please let me know if you enjoy these episodes and as always, subscribe on itunes, Spotify or wherever you listen to the podcasts. Welcome to the Crazy Wisdom Podcast. I've got Andrew Einhorn here and he is the CEO and founders of Level Fields. Welcome to the show.
Level Fields is a research automation platform focused on financials and trying to find trade ideas easily using artificial intelligence. So we saw this kind of hard part about investing that's in the environment where you need to have a very steep learning curve typically to get started investing in equities. We wanted to help change that. We also saw that there was a lot of opinion based stock picking going on and some of it good, much of it not so good. So we wanted to make a real data driven approach to the market, but also give people an opportunity to do something other than buy and hold for the rest of their life, because not everyone's capable of doing that because they're in retirement, nearing retirement, need access to their cash. Right. So wanted to have a way to kind of come in and out of the market or find emerging companies that are doing well, you know, either the, that three months, the six months, or the 12 months. And so we use events to do that. Yeah, well, I think we all kind of are familiar with the fact that a headline will drive a share price movement in the stock market. What we're probably not as familiar with is that the same types of events happen again and again and again. So CEO leaves, CEO arrives. Right. Like that's going to happen a lot of times. And so if you have a data set of what the market normally reacts to that particular event, you can start to predict in the future how the market will react. Just like you could to say, well, the weather in Texas in December varies between, know, 45 degrees and 65 degrees, and the average is 55 degrees. Therefore, next December, I know if I'm planning my vacation, what to wear. It's kind of the same with these events and, and stock price movements. Right. We're trying to give a forecast to say, hey, over here, it's Rainy, it's going to rain for three days. You know, this is the average amount of precipitation. So that's, that's how far the stock typically goes down. And then over here we've got sunny skies, it's going to be 85 degrees. We think it's going to look like this for the next three months. Buy, hold and then get out at that point. And so we're using this massive data set, this massive analytic data set of lots of different types of events to just help people make better decisions that are based on data and not just like a knee jerk reaction to a headline.
Your first full transcript is free. After that, an email opens every transcript in the index — a list of readers we can write to, not a guest book.