91ÖÆƬ³§

By Jeremy Alexis

In the early 2000s, I was promised a future of Long Tails—an era where everything, from consumer offerings to curated experiences, would be hyper-customized to my whims. Picture a sailing trip filled exclusively with Beagle puppies, catered by Harold’s Fried Chicken, and serenaded by New Order playing songs they'd written solely for me. Instead, we got chaotic swings in stock markets, wildly uneven economic growth, universities wrestling existential crises, and an abundance of truly terrible music. Except perhaps for the music part, these phenomena are evidence we're living not in Long Tail times, but in Fat Tail times—an era defined by unpredictability, extreme events, and the failure of traditional predictive models.

I have recently been helping to stand up a Center for Decision Quality at 91ÖÆƬ³§ Tech. As part of our curricular efforts, our board of advisors gave me a very specific challenge: teach undergraduates to "think probabilistically." To simplify this lofty goal, we distilled the task into the three Bs of probabilistic thinking: Bayesian, Bias, and Bell Curves. Bayesian thinking starts from a baseline of facts—priors—and revises those assumptions with every new piece of evidence. A massive oversimplification, yes, but directionally correct. Bias refers to cognitive blind spots that sabotage rational decision-making—particularly overconfidence (we trust our predictions too much) and recency bias (overweighting current information). Each deserves a dedicated section, or perhaps an entire library. Today, though, we're zeroing in on Bell Curves because understanding them can illuminate the chaos of our current age.

There are two types of bell curves: Normal and Fat Tail. Normal curves imply predictability, narrow outcomes, and a future we can reliably plan for. Fat Tail curves are different beasts entirely. They signify unpredictability, massive deviations, and outcomes so uncertain traditional forecasting tools simply collapse.

Someday, my hope is someone will read this and say, "None of this applies anymore." That'll mean we've reached a time of comfortable certainty, where events fit neatly into predictable patterns. It also means someone actually bothered to read this article in the future—a double victory. But for now, to paraphrase Jules from Pulp Fiction, "I'd like that. But that ain't the truth."

In Fat Tail times, planning becomes tricky, surprises frequent, and our traditional tools quickly become obsolete. But paralysis isn't an option. We must adapt—shifting our tools, our outlook, and our expectations. The pivotal shift involves embracing irregular operations not as random crises to be merely reacted to, but as a normal category of operations with established, battle-tested protocols.

Visiting United Airlines' Network Operations Center (NOC) feels like stepping into the heart of a spy thriller: massive screens flicker with real-time data while professionals speak in urgent whispers, monitoring weather patterns, flights in the air, even social media chatter. An airline doesn't invest heavily in a NOC for clear skies and smooth sailing. It's purpose-built to spring into action when emergencies hit or uncertainty peaks. The system was specifically engineered to swiftly respond to irregular operations, ensuring the broader network continues running smoothly, and ripple effects are tightly contained.

Airlines have always been designed for Irregular Operations (IROPs). Businesses and academic institutions can—and should—do the same. Fat Tail times are here to stay, at least for a while. Adopting the following strategies will help your organization thrive amidst uncertainty:

• Redundancy: Favor strategic buffers over relentless asset utilization. Think spare aircraft ready to deploy, extra staff on-call, or backup servers idling quietly—yes, it costs money, but less than catastrophic failure.
• Interoperability: Systems and processes should seamlessly swap in and out. Picture Lego-like compatibility in technology or standardized roles that allow teams to shift dynamically during disruptions.
• Distributed Decision-Making: Empower frontline workers to make decisions on the spot, rather than waiting for instructions from distant central command. It's about trusting pilots to make calls mid-flight instead of awaiting clearance from headquarters.
• Cross-Training: Cultivate versatility in your teams. The flight attendant who can step into gate management during chaos or an engineer who knows enough customer service to diffuse tensions on-site. Cross-trained teams handle shocks better.
• Save the System, Not Every Flight: Focus first on preserving overall stability. Canceling one problematic flight early might prevent cascading delays across dozens of others. Think of it as sacrificing a pawn to protect the queen.
• Chaos Engineering: Constantly stress-test systems in controlled environments. Netflix deliberately shuts off servers randomly to prepare its systems for genuine disruptions. Adopt the same mindset: rehearse chaos, simulate breakdowns, and build robustness through continuous, controlled disruption.
• Scenario Planning: Regularly develop and rehearse detailed scenarios for possible future events—both good and bad. Think of it as strategic storytelling: mapping out alternate futures, practicing responses, and preparing teams to pivot swiftly when reality deviates from the plan.

Adopting these strategies means not only surviving Fat Tail times but thriving through them—ready, resilient, and, dare I say, relaxed when uncertainty strikes.