“Television won’t be able to hold on to any market it captures after the first six months. People will soon get tired of staring at a plywood box every night”.

…………Darryl Zanuck (20th Century Fox Film Producer), 1946

Markets defy our models, politics rewrites itself overnight, technology leaps unpredictably, and AI sharpens—but cannot solve—our uncertainty. Still, forecasting remains essential: not to know tomorrow, but to navigate it.

In every cycle, from calm to chaotic, investors return to the same ritual: trying to peer around the corner of tomorrow. We know the future won’t sit still. We know markets can reverse on a whisper, elections can tilt on a headline, and new technologies can arrive years early—or years late. Yet the urge to forecast persists. Not because prediction grants certainty, but because without it the world becomes unmanageable. Forecasting is our way of imposing shape on motion.

The Human Instinct to Chase Tomorrow

Humans have always lived in negotiation with the unknown. Long before there were bond markets or central banks, there were seasons, storms, and scarcity. To survive, early humans learned to anticipate—to read the sky, the soil, the migration of animals.

Prediction was survival

Today, the threats are abstract, but the circuitry is unchanged. Uncertainty still triggers ancient alarm bells. We forecast because not forecasting feels dangerous. But the modern unknown is not seasonal—it is systemic, reactive, and shaped by human behavior, the most unpredictable variable of all.

Where Forecasting Breaks Down

Our failures are consistent. We miss turning points, underestimate nonlinear shocks, and cling to tidy stories over messy probabilities.

Consider markets. Every major investment bank publishes year-end S&P 500 targets. And every year, the index refuses to obey.

In 2020, no model predicted both a historic crash and the fastest rebound in U.S. history. In 2008, no institution foresaw the full scale of the global financial collapse.

In politics, the blind spots are notorious. Brexit was not supposed to pass. Trump was not supposed to win. Populist movements across Europe were dismissed as tails of the distribution—even as they became the distribution.

In economics, the pattern repeats. The IMF and OECD consistently misproject global growth. The Federal Reserve’s own dot plot has repeatedly failed to predict its future rate path. In 2021, nearly every major institution insisted inflation would stay low. Eighteen months later, inflation hit forty-year highs.

In technology, timelines are chronically wrong. Experts claimed self-driving cars were imminent in 2018. They still aren’t.

Meanwhile, the explosion of generative AI in 2022–2023 blindsided nearly everyone—from Silicon Valley to policy circles.

But these failures are not accidents—they are symptoms.

Tetlock and the Problem With “Expert” Forecasting

In one of the most important findings in modern decision science, psychologist Philip Tetlock tracked more than 28,000 political and economic predictions over two decades.

His conclusion was devastating:

Experts performed barely better than chance.

Specialists performed worse than generalists.

The most confident forecasters were the least accurate.

Tetlock’s work exposed a fundamental flaw: experts mistake narratives for probabilities. They anchor to ideology, tell coherent stories, and underestimate uncertainty. The more certain they sounded, the worse they performed.

This wasn’t a failure of intelligence—it was a failure of psychology.

Kahneman, Tversky, and the Narrative Brain

Research by Daniel Kahneman and Amos Tversky showed that humans are wired for story, not statistics.

---Probabilities feel abstract; stories feel real.

---We overweight recent events (recency bias).

---We ignore large deviations (normalcy bias).

---We misjudge rare events (availability bias).

---We confuse precision with accuracy.

Give a human a 70% probability and we turn it into certainty. Give a model a fat-tailed distribution and we compress it into a line.

We want a narrative. We want to be right. We want the world to make sense.

And so we forecast stories—not futures.

Human Behavior: The Ultimate Wild Card

Forecasting breaks in complex systems because humans adapt faster than models update.

---A single geopolitical spark can reorder global risk sentiment in minutes.

---A viral clip can erase a political campaign overnight.

---A shift in central bank language—not policy, language—can whipsaw bond markets.

---A meme can launch a crypto rally that defies valuation.

This instability is not a bug—it is the engine. Human behavior is fluid, reactive, emotional, opportunistic. And since humans create markets, politics, and technology, the systems themselves become impossible to model statically.

Tomorrow changes because we change.

Does AI Improve Forecasting?

AI is often treated as the antidote to human bias. It helps—but not in the way people imagine.

AI excels at recognizing patterns in stable environments: credit modeling, weather forecasting, logistics, fraud detection. It processes noise better than we do, weighs signals more consistently, and sees nonlinearities our intuition misses.

But AI inherits a fatal constraint:

It learns from the past.

When the future diverges sharply from historical patterns—as it often does in markets, politics, and innovation—AI becomes only marginally more accurate than humans. It cannot foresee what has never happened.

AI is a compass, not a prophecy.

A powerful assistant, not a replacement for judgment.

Where Humans Actually Shine

Humans excel when systems are stable and feedback is immediate. A pilot anticipates turbulence. A radiologist detects malignancy. A meteorologist nails a three-day forecast. These worlds have rules.

Markets do not.

Politics does not.

Geopolitics does not.

Innovation does not.

The more adaptive the system, the worse deterministic forecasting becomes.

Why We Forecast Anyway

If we’re so consistently wrong, why keep forecasting?

Because forecasting isn’t about being right—it’s about being prepared.

A forecast forces us to articulate assumptions, identify unknowns, weigh risks, and imagine multiple futures. It sharpens decision-making even when the forecast itself proves wrong.

Forecasting is the discipline of living with uncertainty. It is the structure we build around the unknowable.

Investment Takeaways: Navigating a World That Won’t Sit Still

For investors, the lesson is not to abandon forecasting—but to reframe it.

Treat forecasts as hypotheses, not destinations.

Favor ranges over point estimates.

Use probabilities, not narratives.

Revise quickly as conditions change.

Avoid falling in love with elegant explanations.

Stay humble—because the world will not stay predictable.

The markets do not reward certainty.

They reward adaptability, resilience, and speed of adjustment.

The goal is not to know the future --- The goal is to be ready for the futures that arrive.

Forecasts don’t reveal tomorrow   -   They reveal the forecaster.

 

John Vidas, Portfolio Manager

December 2025