Why Profitable Trading Tricks Are Now Disappearing in Months Instead of Years

For decades, a clever trading idea could make money for five to seven years before enough other investors copied it and profits dried up. Not anymore. New research shows that artificial intelligence has collapsed this timeline to just 18 months. More than 70% of big institutional trading desks now use AI, and that's changed the speed of the game entirely.

When an investing idea stops making money, traders call that its "half-life"—how long before its usefulness is cut in half. The math is straightforward: add more competitors using the same AI tools, and that half-life shrinks fast. At today's adoption rates, traditional quantitative strategies (that's finance speak for trading plans built on mathematical patterns) face extinction within quarters, not years.

Three Ways AI Speeds Up the Death of Profitable Trades

The research points to three overlapping forces eroding profits at machine speed. The first is crowding. When multiple AI systems spot the same market weakness and pile in to exploit it, the advantage evaporates almost instantly. It's like a bargain everyone's heard about by noon—the deal's already gone.

The second is something harder to see: performative signal erosion. As thousands of traders execute the same strategy, their buying and selling actually change the market patterns that made the strategy work in the first place. The profit opportunity was real, but exploiting it destroyed it.

The third force is what researchers call "Red Queen competition," named after Through the Looking Glass. Just to keep pace, firms must constantly upgrade their AI just to stay even. They're running faster just to stand still.

These three forces don't take turns—they hit simultaneously. A momentum trade that might have worked for three to four years now faces crowding within weeks, watch the pattern shift under its feet, and get outpaced by faster algorithms hunting the next angle.

When One Broken Trick Breaks the Next One

Here's where things get genuinely destabilizing: the collapse of one profitable strategy can trigger a domino effect. When high-frequency trading that exploited mean reversion (quick, temporary price swings) stopped working, traders rushed to find new angles—and that competition accelerated the decay of other strategies they turned to instead.

This isn't gradual erosion. The research suggests it's closer to a phase transition—like ice melting into water rather than slowly dampening. Once AI adoption hits certain thresholds in particular markets, the behavior flips from linear to chaotic.

The August 2007 financial crisis offered an early preview. Multiple hedge funds unwound crowded trades at once, feeding a cascade of losses. The difference now: AI systems can discover, exploit, and oversaturate an opportunity in hours instead of weeks. Speed is measured in machine time, not human time.

The Confusing Evidence on Whether Markets Are Getting Smarter

Here's the puzzle: the data doesn't paint one clear picture. Federal Reserve research testing different language models (AI programs designed to understand text and markets) found mixed results—some performed decently, others not. That suggests AI's edge varies depending on what strategy you're running.

At the same time, studies of AI-powered mutual funds show they outperform human managers. They pick stocks better and trade less often, which means lower fees and smaller tax hits. Fewer emotional mistakes helps too.

Yet other academic papers argue that algorithmic trading can actually undermine market efficiency. Separately, research also shows that algorithmic trading improves efficiency. How can both be true?

The likely answer is that it depends on who's using the AI and how well they've built their systems. Early adopters with the best infrastructure are pulling profits out of markets still learning to adapt. Latecomers face markets that are already saturated and far more efficient. That creates winners and losers, but it doesn't settle whether AI makes markets better or worse overall.

What This Means for Active Investors

The research has real implications for how professional investors need to think about their work. Building a portfolio around trading patterns that worked in the past is now a recipe for overestimating future profits. Those historical patterns won't hold up the same way.

Diversification—the old idea that different trading strategies don't all fail at the same time—is also becoming less reliable. When signal half-lives compress, strategies start moving together during crises, and their insurance value evaporates.

For professional trading firms, the pressure is mounting. They can no longer develop a strategy, roll it out, and harvest profits for years. Instead, they need to keep refreshing their algorithms constantly and invest heavily in research just to tread water.

The broader question worth taking seriously: if the useful life of a trading advantage shrinks to months, what happens to the economics of active management? If most profits flow only to the fastest-moving firms, how do traditional asset managers justify their fees? As human stock-pickers have declined substantially, with expert traders "almost non-existent in the new millennium," markets increasingly rely on algorithmic competition rather than human judgment to set prices fairly.

The Theoretical Endgame

The mathematics suggests a future where almost everyone uses AI and signal half-lives approach zero. In that extreme case, excess returns exist only in the microsecond between discovering an opportunity and a thousand algorithms simultaneously exploiting it. Markets would approach something close to perfect efficiency—but through machine speed, not human wisdom.

Whether markets can actually function that way is an open question the research doesn't answer. The math gets you to the mechanism but not to the policy question: what should change, and how?

What's clear is that we're in a transition period. The compression of profitable trading ideas from years down to months isn't just incremental change. It's a fundamental shift in how markets operate, with consequences rippling far beyond individual trades to the basic business model of professional investing itself.