Why some edges decay while others endure
Two Ways to Hunt
The cheetah is the fastest land animal on Earth. It can accelerate from rest to seventy miles per hour in three seconds. Its spine flexes like a spring. Its claws grip the ground like cleats. Every feature of its body has been sculpted by millions of years of evolution for a single purpose: catching prey.
Yet the cheetah fails more often than it succeeds. Most hunts end without a kill. The gazelle escapes.
This is not a design flaw. It is the signature of an arms race. The cheetah is fast because slow cheetahs starved. The gazelle is fast because slow gazelles were eaten. Each species has been sharpened by the other. Neither can rest. Neither can optimise once and stop. The race continues because the other side keeps running.
But not all predators hunt this way.
The spider builds a web and waits. It does not predict where the fly will be. It does not chase, anticipate, or extrapolate trajectories. It constructs a structure that catches whatever arrives. The spider’s success does not depend on outsmarting its prey. It depends on being present, prepared, and positioned when something blunders into the silk.
These two hunting strategies represent fundamentally different relationships with uncertainty. The cheetah predicts. The spider responds. The cheetah’s success depends on being faster and smarter than its prey. The spider’s success depends on the inevitability that something, somewhere, will move.
Markets reward both strategies. But only one is subject to the arms race.
The Predictive Hunter
Financial markets are filled with cheetahs.
Every strategy that attempts to forecast future prices is a predictive hunter. Statistical arbitrage models that anticipate which stocks will outperform. Factor timing strategies that predict when value will beat growth. Macro funds that position for economic shifts before they arrive. Merger arbitrageurs who assess the probability that a deal will close.
These strategies share a common structure: they form a view about what will happen and position accordingly. They are making predictions, whether explicit or implicit, about future states of the world.
Predictive strategies are subject to the arms race because the prey learns.
Consider merger arbitrage. When an acquisition is announced, the target’s stock typically trades below the deal price, reflecting the risk that the deal might fail. Early arbitrageurs captured this spread reliably. The strategy was straightforward: buy the target, wait for the deal to close, collect the premium.
Then competition intensified. Spreads compressed as more capital competed for the same opportunities. The easy deals offered thin returns. The attractive spreads existed only in deals with genuine uncertainty. The arbitrageur who once harvested a reliable premium now faced a choice between thin margins and real risk.
The prey had learned to run faster.
Consider quantitative equity strategies. Early practitioners discovered patterns in stock returns that persisted long enough to exploit. Value stocks outperformed. Small caps outperformed. Stocks with positive momentum outperformed on certain horizons. These patterns were documented, published, and deployed.
Then the patterns became crowded. As more capital chased the same signals, the entry points degraded. The premiums compressed. The easy gains were captured earlier by faster participants. What once required insight now required infrastructure, speed, and scale. The hunting ground had not disappeared, but the prey had become far harder to catch.
This is the pattern for all predictive strategies. They work until others notice. Capital flows in. Returns compress. The edge that once seemed robust erodes under the weight of competition.
Why Predictive Alpha Decays
In the language of finance, alpha is the return a strategy generates above what could be explained by passive exposure to risk. Alpha is the reward for skill, insight, or edge.
Predictive alpha decays because markets learn.
When a strategy discovers a predictable pattern, it begins to exploit it. The exploitation changes the market. Prices adjust faster. The pattern that generated the alpha weakens. Other participants observe the success and deploy similar approaches. The alpha gets divided among more hunters. Eventually, what was once a rich hunting ground becomes picked clean.
This is not failure. It is the natural consequence of success in a competitive ecosystem. The cheetah does not catch the gazelle easily precisely because cheetahs have been catching gazelles for millions of years. The strategy that worked last decade may struggle this decade because markets have been adapting to it.
The half-life of predictive alpha has been shortening. Strategies that once persisted for years now erode in months. Computing power increases. Data becomes more accessible. The tools for finding patterns become widely distributed. The cheetahs multiply, and the gazelles evolve.
The Red Queen Effect describes this dynamic precisely. In Lewis Carroll’s Through the Looking-Glass, the Red Queen tells Alice: “It takes all the running you can do, to keep in the same place.” Predictive strategies must continuously evolve just to maintain their relative position. Standing still means falling behind.
A quantitative fund develops a signal that predicts short-term returns. The signal works. But its success leaves traces in the data. Other funds detect the pattern. They develop their own versions. The signal becomes crowded. Returns decay. The fund must develop a new signal just to maintain its previous performance.
It is running to stay in place.
The Spider’s Web
But there is another way to hunt.
Trend-following does not predict where prices will go. It responds to where prices are going. The trend follower does not forecast the direction of the next move. They observe the direction of the current move and align with it. The distinction is subtle but profound.
Mean-reversion strategies operate similarly at their own horizons. They do not predict that a stretched price will snap back at a particular moment. They observe that prices have moved far from some reference point and position for the structural tendency toward reversion. They respond to what is, not what will be.
These are spiders, not cheetahs.
The spider’s web works because movement is inevitable. Flies do not learn to stop flying. They cannot. Flight is what makes them flies. The spider does not need to predict which fly, or when, or from which direction. It needs only to build the web and wait.
Trend and mean-reversion persist because they harvest structural features created by trading itself. When traders act on information, they move prices. These moves do not happen instantaneously. They unfold over time as different participants react, creating trends. When prices overshoot, the stretched positioning eventually reverses, creating mean-reversion.
As long as there are traders, whether human or algorithmic, these structural regularities will exist. They are not inefficiencies to be arbitraged away. They are the unavoidable byproducts of the trading process itself. The arms race between predictive strategies does not erode them. If anything, it feeds them.
This is why trend-following has worked across decades, across asset classes, across market regimes. It is not because trend followers are smarter than the market. It is because they have built a web that harvests the inevitable. The fly does not learn to avoid silk. It cannot. Movement is its nature.
The Arms Race Has Boundaries
The distinction matters because it determines whether you are running in an arms race or harvesting what the race produces.
Predictive strategies compete directly with other predictive strategies. Every improvement on one side creates selection pressure on the other. The gazelles get faster. The cheetahs get faster. The gap between them remains roughly constant. The race has no finish line.
Structural response strategies do not compete in the same way. The trend follower is not trying to predict better than other trend followers. They are positioning to capture moves that emerge from the collective activity of all market participants, including the cheetahs chasing the gazelles.
When a predictive strategy works and then fails, it often creates exactly the kind of price movement that trend-following harvests. The crowded trade that unwinds. The regime shift that surprises the forecasters. The correlation spike that breaks the models. These disruptions are meat for the spider’s web.
The arms race, in other words, feeds the process-based hunters. The more intensely predictive strategies compete, the more they create the dislocations and trends that structural strategies harvest. The spider does not compete with the cheetah. The spider eats what the chase disturbs.
The Cost of Being a Cheetah
Predation is expensive.
A cheetah’s hunt consumes enormous energy. The acceleration, the sprint, the explosive effort: these cost calories that must be replenished. A failed hunt is not just a missed meal. It is a net loss. Too many failed hunts and the cheetah weakens, slows, and eventually cannot hunt at all.
Predictive strategies carry similar costs. They require research, technology, and personnel. They require infrastructure to execute and risk systems to monitor. Every trade incurs transaction costs. Every position carries risk. The hunt is not free.
When the hunting ground is rich, these costs are easily covered. When the hunting ground is depleted, the costs become burdensome. A strategy that once generated comfortable returns after costs may find itself barely breaking even as alpha decays and expenses remain fixed.
This creates a survivorship dynamic. When prey is abundant, predators multiply. When prey becomes scarce, predators starve. The weakest exit. The population contracts. Eventually, with fewer predators hunting, opportunities recover, and the cycle begins again.
The spider faces different economics. The web, once built, costs little to maintain. The spider waits. It does not exhaust itself in failed chases. When the hunting is thin, it conserves energy. When something arrives, it acts. The cost structure matches the uncertainty of the environment.
Apex Predators and Their Vulnerabilities
Every ecosystem has apex predators. They sit at the top of the food chain. They dominate their environment. They seem invincible.
But apex predators carry hidden vulnerabilities.
Their position at the top means they are the most specialised. They have evolved to exploit a particular niche with maximum efficiency. When that niche changes, they are often the least able to adapt. The very features that made them dominant become liabilities in a new environment.
Markets have apex predators too. The strategy that dominated a decade. The fund that seemed invincible. The approach that attracted the largest pools of capital.
These apex strategies often fail suddenly and dramatically. Their size makes them slow to adapt. Their success breeds overconfidence. Their specialisation leaves them exposed when conditions shift. The regime change that a smaller, more agile participant might navigate destroys the apex predator.
Long-Term Capital Management was an apex predator. It had the best talent, the most sophisticated models, the deepest relationships. It dominated its niche. Then the niche changed. The positions that had been reliably profitable became correlated. The leverage that had amplified returns amplified losses. The apex predator did not survive.
The lesson is not that size is bad or that success is dangerous. The lesson is that dominance in one regime does not guarantee survival in the next. The cheetah that perfectly adapted to one savanna may starve when the grasslands shift.
The Reframe
The question is not whether you are a predator. Everyone in markets is hunting something.
The question is what kind of predator you are.
If you are a cheetah, know what that means. Your edge depends on being faster, smarter, and better-resourced than your competition. Your alpha will decay. Your prey will learn. You must continuously evolve just to maintain your position. The race never stops.
If you are a spider, know what that means. Your edge depends on structure, patience, and the inevitability of movement. You are not trying to outsmart the market. You are positioning to harvest what the market produces through its own activity. Your returns will be irregular. Your waiting will be long. But the flies will never learn to stop flying.
Most participants are some combination. The portfolio that blends predictive and structural strategies. The approach that forecasts when it can and responds when it cannot. The key is knowing which part of your process faces the arms race and which part harvests its byproducts.
Alpha is temporary. Adaptation is durable.
The race never ends.
Keep running, or build a web.
This is the fifth article in a series exploring markets as living systems. Previously: “The Murmuration,” on how collective motion emerges without a leader. Next: “The Forest Fire,” on why destruction enables renewal.
Want the theoretical foundation for why trend following works?
The Fractals of Finance: Determinism, Adaptation and the Geometry of Markets bridges complexity science with practical trading implementation. With a foreword by Jerry Parker, original Turtle Trader.
Available now on Amazon in paperback, hardcover, and Kindle.
