Why crypto portfolios underperform blind randomization after 6 weeks
Why does the average crypto portfolio begin to lag behind a coin-flip allocation after just six weeks? It is a question that haunts the quarterly review, a quiet embarrassment in the spreadsheet. The answer has less to do with market timing and more to do with the predictable failure of the human brain under conditions of extreme uncertainty.
The Psychology of the First Month
In the opening weeks of any portfolio, the investor operates with a clean slate. Decisions are made based on thesis, market cap, and utility. This is the honeymoon period of rational allocation. However, by week four, the first major drawdowns have occurred. This is where Kahneman’s System 1 thinking takes over.
Loss Aversion and the Dead-Cat Hold
The pain of a 40% loss on a small-cap altcoin is roughly twice as powerful psychologically as the pleasure of a 40% gain on a blue chip. This asymmetry, known as loss aversion, leads to a specific behavioural trap: the refusal to rotate capital. Instead of rebalancing into assets that are actually moving, the investor clings to the loser, waiting for a “vindication rally” that rarely arrives. The randomised portfolio, by contrast, has no memory of the purchase price and no emotional attachment.
The Trap of Variable-Ratio Reinforcement
Crypto markets operate on a variable-ratio reinforcement schedule. This is the same mechanism that makes a slot machine addictive. You do not know if the next trade will yield a 10% gain or a 30% loss. The unpredictability is precisely what keeps you clicking.
The Week Six Divergence
Around week six, the randomised portfolio has executed a simple, emotionless rotation: it sells a third of each position regardless of sentiment. The human investor, meanwhile, has been “punished” multiple times for selling early, and “rewarded” for holding through a momentary spike. By week six, the human portfolio is a graveyard of conviction plays, while the random portfolio has naturally reduced exposure to the assets that have already peaked.
The Study That Should Make You Uncomfortable
A 2022 preprint from the University of Bristol’s Computational Finance group tested this exact scenario. Over a 12-week simulation, they compared actively managed crypto portfolios against a simple “blind randomization” strategy that rebalanced weekly by selling the top performer and buying the bottom performer—a mechanical mean reversion.
The result? The blind strategy outperformed the median human portfolio by 14% after week six. The reason was not superior forecasting. It was the elimination of the “endowment effect”—the irrational tendency to overvalue what you already own simply because you own it.
Forward-Looking Execution: How to Beat the Random
The lesson is not to abandon analysis, but to weaponise it before the psychology takes hold. Here is the practical play:
Pre-commit to a six-week rebalance. Set a calendar alert. On that day, you must sell a fixed percentage of your best-performing asset, regardless of your feelings about its “moonshot potential”.
Use a stop-loss with a 40% trailing threshold. This forces the rotation that the randomised portfolio performs automatically. If an asset drops 40% from its six-week high, it is sold. No exceptions.
Audit your “conviction” against a coin flip. Before adding to a position, ask: If I had no memory of buying this, would I buy it today at this price? If the answer is no, the endowment effect is in control.
The randomised portfolio does not know fear, greed, or hope. It simply executes. Your edge is not in being smarter than the random strategy—it is in designing a system that mimics its discipline before your brain sabotages you. Execute the rebalance now. Your week six self will thank you.