One truth about markets is this:
There are multiple ways to make money.
Discretionary traders usually discover their path and then spend years mastering it.
For some, it’s chart reading.
For others, technical indicators.
For many, pure price action.
If you talk to enough discretionary traders, you’ll quickly realize something important:
Everyone makes money differently.
That itself tells us something fundamental — there are infinite ways of extracting alpha from markets.
How algotrading changes the game
In algotrading, this idea expands even further.
Machines don’t think like humans.
Algos can take any form of data as input and convert it into trades:
-
Price & volume
-
Options data
-
Fundamental data
-
Macros
-
Weather data
-
News
-
Satellite imagery (yes, even that)
So by the very nature of markets, there can exist millions or even billions of distinct alphas, each structurally different from the other.
This is powerful.
Because if you can keep discovering new alphas, you don’t need to worry too much about one edge disappearing overnight.
Which brings us to the real risk.
What is alpha decay?
Alpha decay is the idea that:
What worked in the past may not work in the future.
Even if an algo has worked beautifully for years, there is no guarantee it will continue to do so.
Why does alpha decay happen?
-
When an edge becomes crowded
-
When market structure changes
-
When regimes shift
-
When liquidity, volatility, or participant behavior changes
Large quant firms try to protect their alphas because once too many people do the same thing, returns get competed away.
But leakage is just one reason.
Markets are a complex system with infinite interacting variables. Trying to fully explain why an alpha dies is often a futile exercise.
The important part is this:
Alpha decay is inevitable.
The real risk most traders underestimate
The biggest risk isn’t drawdowns.
The biggest risk is betting everything on a single idea.
Imagine deploying all your capital into one algo:
-
It works for years
-
You grow confident
-
Capital scales
-
And then… it slowly stops working
Now you’re back to where you started — or worse.
This is not a rare event.
This is a structural risk.
How to navigate alpha decay
The most robust way to deal with alpha decay is diversification across ideas, not just instruments.
Think of it like a business.
If your business has:
-
Multiple products
-
Multiple revenue streams
-
Multiple customer segments
It’s unlikely everything breaks at the same time.
Markets work the same way.
If you run a pool of uncorrelated or structurally different algos, the probability that all of them decay together is significantly lower.
Some will underperform.
Some will decay.
Some will work better in certain regimes.
That’s fine.
The portfolio survives.
The institutional lesson
At an institutional level, risk is managed not by finding one perfect algo but by running hundreds, thousands, or even tens of thousands of small edges.
Think in terms of:
- 100 → 1,000 → 10,000 ways of making money
That’s how risk is actually reduced over time.
Bottom line
-
Alpha decay is real and unavoidable
-
Single-algo portfolios are fragile
-
Diversification across ideas, not just assets, is key
-
Large algo pools outperform single “hero” strategies over time
The goal isn’t to find the best algo.
The goal is to build a system that survives when individual alphas fail.
That’s how risk is truly reduced in algotrading.
PS: We’re releasing an app update soon that lets you analyze correlations between algos and combine them to see how they perform together as a portfolio.