Data Update 1 for 2026: The Push and Pull of Data

Data Deep Dive: 2026 Analysis & The Future of Data-Driven Decision Making

Key Concepts:

• Signal in the Noise: Identifying meaningful patterns and insights within large, often contradictory, datasets.
• Structural Change: Shifts in the underlying economic or market conditions that invalidate historical patterns (like mean reversion).
• Mean Reversion: The tendency of asset prices or market indicators to return to their historical average.
• Bias (in Data & Analysis): Systematic errors in data collection, interpretation, or analysis stemming from preconceptions or incentives.
• False Precision: The illusion of accuracy created by adding unnecessary decimal places to numerical data.
• Gresham’s Law (applied to Data): The tendency for bad or misleading data to displace good, reliable data.
• Data Overload: The difficulty of processing and interpreting information when presented with an excessive amount of data.
• Demodern Bot/AI Integration: Utilizing AI to automate data processing and analysis tasks.

I. The Evolving Landscape of Data (1993-2026)

The speaker begins by outlining his three-decade-long practice of analyzing data at the start of each year. Initially driven by personal need for corporate financial analysis in the 1990s (starting with roughly half a dozen datasets sourced from Value Line on CD-ROM), this practice has expanded to encompass over 200 datasets today. He acknowledges a paradoxical feeling: despite unprecedented access to data – both quantitative and qualitative – he feels less at ease than he did 40 years ago. This sets the stage for a discussion of the benefits and drawbacks of the current data-rich environment.

II. The Good Side of Data: Four Key Benefits

The speaker identifies four primary advantages of leveraging data effectively:

1. Signal in the Noise: The ability to discern meaningful patterns from contradictory information. Financial analysis often involves conflicting data points, and the skill lies in identifying the most reliable signals.
2. Coping Mechanism for Uncertainty: Data provides a framework for understanding and addressing uncertainty by framing it as a range of possible outcomes. Tools like simulations and visualizations aid in incorporating uncertainty into decision-making. The current global economic order’s breakdown increases the need for this approach.
3. Prescription for Tunnel Vision: Data, particularly when viewed across time and companies, provides perspective and helps analysts avoid getting bogged down in irrelevant details.
4. Shield Against Disinformation: In an era of widespread misinformation (found not just on social media but also in established financial news sources like CNBC and the Wall Street Journal), data serves as a crucial verification tool. The speaker emphasizes the importance of independent verification rather than relying solely on expert opinions.

III. The Bad Side of Data: Potential Pitfalls

Despite its benefits, the speaker cautions against the inherent dangers of relying on data without critical thinking:

1. The Illusion of Control: Attaching numbers to things can create a false sense of control, even when the underlying situation remains uncertain. The tendency to add decimals to numbers is cited as an example of “false precision.”
2. Bias: Acknowledging that objectivity in finance is a myth, the speaker highlights the pervasive influence of bias. This bias can stem from pre-existing beliefs, incentives (reward/punishment based on outcomes), and can manifest subconsciously in data selection. He references the adage "lies, damn lies, and statistics" to illustrate how data can be manipulated to support pre-conceived notions.
3. The Trap of Mean Reversion: The speaker critiques traditional value investing strategies based solely on mean reversion (buying low PE stocks expecting them to revert to the average). He argues that this approach fails to account for structural change – fundamental shifts in the economic landscape that render historical averages irrelevant. The current debate surrounding AI valuations (overpriced vs. a new world order) exemplifies this issue.
4. The Diminishing Voice of Expertise: Over-reliance on data can lead to a passive role for analysts, simply letting the data “speak for itself” and abdicating responsibility for interpretation and judgment. The phrase "I didn't do it, the data did it" is used to illustrate this danger.

IV. Data in 2025: A Snapshot

The speaker details the composition of his 2025 dataset, which includes 48,156 companies globally. Key statistics include:

• Geographical Breakdown:
  • US: 69.8 trillion market cap (47% of global equities) – a slight decrease from 48.7% the previous year, with a 16.8% increase in value.
  • Global Equities: 26.4 trillion increase (21.5% growth).
  • Latin America & Asia: Outperformed the US in dollar terms.
  • India: The worst-performing market in dollar terms (partly due to rupee depreciation).
• Sectoral Breakdown:
  • Technology: 32.8 trillion market cap (22% of global equities).
  • Materials: Best performing sector (37.8% growth).
  • Energy & Consumer Staples: Worst performing sectors (5% and 8.12% growth, respectively).

V. Data Methodology & Principles

The speaker outlines his approach to data collection and analysis:

• Data Sources: S&P Capital IQ, Bloomberg Terminal, and various online subscriptions (both free and paid). He emphasizes that almost all the data used is publicly available, with the value of paid services lying in convenience and consolidation.
• Data Variables: Covers a wide range of areas including corporate governance, operating metrics, hurdle rates, accounting returns, financing, dividends, valuation, R&D, and working capital.
• Key Principles:
  • Consistency: Definitions of variables are aligned with his understanding of corporate finance and valuation (e.g., treating lease commitments as debt and R&D as capital expenditure).
  • Transparency: Detailed definitions of each variable are publicly available for scrutiny.
• Data Aggregation: Data is aggregated at the industry level (95 industry breakdowns) to reduce sampling bias and protect company-level confidentiality.
• Market Cap Weighted Averages: Industry averages are calculated using a market cap-weighted approach, summing market capitalization and net income to provide a more representative figure.

VI. The Future: AI and the Role of the Analyst

The speaker anticipates a significant impact from AI on the data analysis landscape. He notes that much of his current work is mechanical and rule-based, tasks that an AI “demodern bot” could perform more efficiently. He plans to integrate AI more actively into his data update process, potentially automating the entire process in the future. He predicts that businesses built around data collection and processing will face disruption as AI lowers costs and increases efficiency. He also suggests that quantitative investing will need to evolve, requiring more than just mathematical prowess to generate excess returns.

VII. Concluding Remarks

The speaker encourages users to explore his data (available on his website) but emphasizes the importance of critical thinking, independent verification, and taking ownership of the analysis. He stresses that data should be a starting point, not an ending point, and that conclusions should remain the user’s own. He acknowledges the potential for errors and encourages feedback.

Notable Quote:

“When you feel more comfortable when you can attach a number to something…sometimes just getting a number on something makes you feel like you're in control of it, even though that's not true.”

Actionable Insights:

• Question Data Sources: Always consider the biases and motivations of the data provider.
• Don't Accept Data at Face Value: Understand how variables are defined and calculated.
• Look Beyond Historical Averages: Be aware of structural changes that may invalidate past trends.
• Embrace Critical Thinking: Don't let data replace human judgment and expertise.
• Explore and Verify: Cross-check data with multiple sources.